Double Heck Route to a Dibenzoxepine and Convergent Suzuki Cross-Coupling Strategy for the Synthesis of an MR Antagonist

Article abstract of DOI:10.1021/acs.oprd.6b00368

A practical pilot plant convergent synthesis of MR antagonist LY2623091 was established. For synthesis convergence, a vinyl bromide geometric isomer and chiral alaninol derivative were required building blocks. Key to the synthesis route development is a stereoselective synthesis of the E-vinyl bromide via a sequential double Heck reaction, Suzuki-Miyaura cross-coupling of the vinyl bromide, a selective nitro reduction, and a highly sensitive cyanamide hydrolysis to the urea. Improvements in yield and processing were accomplished by two sets of telescoping methods which decreased the manufacturing time and provided purity enhancements.

Full text of DOI:10.1021/acs.oprd.6b00368

Subscriber access provided by University of Newcastle, Australia
Full Paper
Double Heck Route to a Dibenzoxepine and Convergent Suzuki
Cross-Coupling Strategy for Synthesis of an MR Antagonist
Marvin M. Hansen, Neil J. Kallman, Thomas Mitchell Koenig, Ryan J Linder, Rachel N
Richey, John R Rizzo, Jeffrey A Ward, Hannah Yu, Tony Y. Zhang, and David Mitchell
Org. Process Res. Dev., Just Accepted Manuscript • DOI: 10.1021/acs.oprd.6b00368 • Publication Date (Web): 19 Dec 2016
Downloaded from http://pubs.acs.org on December 20, 2016
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 free 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 accessible to all
readers and 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.
Organic Process Research & Development 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 30
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Double Heck Route to a Dibenzoxepine and
Convergent Suzuki CrossꢀCoupling Strategy for
Synthesis of an MR Antagonist
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
Marvin M. Hansen, Neil J. Kallman, Thomas M. Koenig, Ryan J. Linder, Rachel N. Richey, John
R. Rizzo, Jeffrey A. Ward, Hannah Yu, Tony Y. Zhang, and David Mitchell*
Small Molecule Design and Development, Lilly Research Laboratories, Eli Lilly and Company,
Indianapolis, Indiana 46285
ACS Paragon Plus Environment
1

Organic Process Research & Development
Page 2 of 30
1
2
3
4
5
6
7
Table of Contents Graphic
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
ACS Paragon Plus Environment
2

Page 3 of 30
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
ABSTRACT: A practical pilot plant convergent synthesis of MR antagonist LY2623091 was
established. For synthesis convergence, a vinyl bromide geometric isomer and chiral alaninol
derivative were required building blocks. Key to the synthesis route development is a
stereoselective synthesis of the Eꢀvinyl bromide via a sequential double Heck reaction; Suzukiꢀ
Miyaura crossꢀcoupling of the vinyl bromide; a selective nitro reduction and a highly sensitive
cyanamide hydrolysis to the urea. Improvements in yield and processing were accomplished by
two sets of telescoping methods which decreased manufacturing time and provided purity
enhancements.
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: MR antagonist LY2623091, Heck reaction, vinyl bromide, alaninol, SuzukiꢀMiyaura
crossꢀcoupling, cyanamide hydrolysis.
ACS Paragon Plus Environment
3

Organic Process Research & Development
Page 4 of 30
1
2
3
4
INTRODUCTION
5
6
7
8
9
Mineralocorticoid receptor (MR) antagonists have been evaluated as treatment for
hypertension, congestive heart failures and chronic kidney disease including diabetic
nephropathy.1 Marketed MR antagonists include eplerenone, spironolactone and nifedipine to
name a few.2 More recently, LY2623091 (1, Figure 1), an MR antagonist, was evaluated as
therapy for resistant hypertension with an effective amount of tadalafil.3 In an effort to prepare
active pharmaceutical ingredient (API) for clinical evaluation, a multiꢀkilogram synthesis was
developed which was based on a crossꢀcoupling retrosynthesis of vinyl bromide 2 and
arylbromide 3 (Figure 1). This article describes the development of a practical means for
preparing 1 on pilot plant scale.
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. Retrosynthesis of 1.
The first generation synthesis of 1 is presented in Scheme 1. The process was efficient
such that it afforded multiple kilogram quantities of API for early animal toxicology and clinical
evaluation. Vinyl bromide 2 was available as a building block from related drug discovery
programs in our laboratories.4 In addition, multiple means for its preparation were attainable. A
Miyaura borylation5 provided vinylboronate 4 which served as a coupling partner for aryl
ACS Paragon Plus Environment
4

Page 5 of 30
Organic Process Research & Development
1
2
bromide 5. The SuzukiꢀMiyaura crossꢀcoupling6 gave the single Eꢀgeometric isomer 6.
Reduction of the nitro group and reaction of the resulting aniline 7 in solution with a
carbonimidate resulted in the cyanoguanidine 8. Hydrolysis of cyanamide functionality of 8 as
the final step gave compound 1 which was further purified by crystallization to afford the API.
3
4
5
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
For synthesis route design, the first generation route was selected for optimization.
Advantages of this approach included the following: a) the route was highly convergent as it
dissected the target into two approximately equal parts; b) we had multiple experiences
preparing analogs of vinyl bromide 2; c) there was no need to install the chiral methyl center as
the moiety was available as an alpha raw material from alaninol; and d) the crossꢀcoupling of
vinyl boronate 4 and aryl bromide 3 established and retained the required Eꢀgeometric isomer.
Our work plan also involved stream lining the first three steps as they were all metal catalyzed
processes. The hydrobromide salt of 3 (compound 5) was corrosive to plant equipment as well as
being an irritant due to the salt form. Therefore, optimizing the synthesis also included the
search for an alternate salt form. The remaining development plan included impurity rejection in
crystallization of 1 as well as overall step reduction and waste minimization.
ACS Paragon Plus Environment
5

Organic Process Research & Development
Page 6 of 30
1
2
3
4
5
6
7
Scheme 1. First Generation Synthesis Route
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
RESULTS and DISCUSSION
1. Synthesis of vinyl bromide 2 via a sequential double Heck reaction. Preparation of
vinyl bromide 2 focused on a requirement to selectively prepare the Eꢀgeometric isomer. In the
initial drug discovery route, this was achieved with low stereocontrol (2:1) by bromination of the
alkene as shown in Scheme 2.4a,b Several different fluoro substitution patterns on the
dibenzoxapine ring system were required for the discovery structureꢀ activity relationship (SAR)
efforts. The data from preparing these analogs were utilized in the development evaluation
process.
ACS Paragon Plus Environment
6

Page 7 of 30
Organic Process Research & Development
1
2
3
4
Scheme 2. Discovery Route to Vinyl Bromide 2
5
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
It was envisioned that a double Heck reaction7 disconnection as shown in Scheme 3
might provide an expeditious and stereoselective synthesis of the dibenzoxepine ring system
from 9 to 10. The literature has an example of a similar double Heck disconnection for
preparation of 6ꢀmembered rings, and includes one example of a 7ꢀmembered ring, lacking the
oxygen in the bridge, as shown in Scheme 3.8 The literature example showed that formation of
the 7ꢀmembered ring was more difficult than the 6ꢀmembered ring, as evidenced by formation of
the double intermolecular Heck product, and did not address the stereochemistry question.
However, the rapid assembly of two bonds in a single step made this disconnection a top priority
for further evaluation.
Scheme 3. Proposed Double Heck Route to Dibenzoxepine Ring System
ACS Paragon Plus Environment
7

Organic Process Research & Development
Page 8 of 30
1
2
3
4
F
O
O
5
6
OEt
F
7
8
9
O
Pd(OAc)2 (cat), TBAB,
NaOAc, NMP
Br
EtO2C
I
9
10
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
Literature Precedent
O
OEt
+
Pd(OAc)2 (2.5 mol%),
TBAB, NaOAc, NMP,
49% isolated mixture
Br
CO2Et
EtO2C
Br
6.5 : 3.5
CO2Et
The fluorinated biaryl ethers 11 and 12 were explored first, utilizing the more readily
attainable dibromo ether precursors (Scheme 4). The double Heck reaction with 11 proceeded to
afford a 7:1 mixture of the desired Eꢀisomer relative the Zꢀisomer along with 18% of the double
intermolecular Heck byproduct that was also observed in the literature example above. Efforts to
suppress the intermolecular Heck byproduct through reaction dilution or control of the ethyl
acrylate stoichiometry were not successful. The stereocontrol can be explained by the
mechanistic work of Heck,7b assuming that the more electron deficient aryl bromide reacts first,
followed by cis addition in the second Heck reaction, then rotation and cis elimination (Scheme
4). Despite formation of the double intermolecular Heck byproduct, the rapid assembly of the
dibenzoxepine ring in one step with acceptable stereocontrol was gratifying, and crystallization
afforded purified product in 50% yield. Applying the same double Heck reaction to the
regioisomeric biaryl ether 12 gave a higher yield (68%) as shown in Scheme 4. The lower level
of impurities formed in this case may be related to the more electon deficient aryl bromide in the
left hand ring, due to inductive withdrawl from the metaꢀF group, without any donation from the
fluorine lone pair of electrons.9
ACS Paragon Plus Environment
8

Page 9 of 30
Organic Process Research & Development
1
2
3
4
Scheme 4. Double Heck Reactions of Fluorinated Dibromo Ether Substrates
5
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
With this strong precedent provided by two successful double Heck reactions in the
difluoro ether series, the same strategy was applied to the monofluoro analog in Scheme 5. Iodo
bromo substrate 9 was prepared, as the left hand aromatic ring lacks a fluoro electron
withdrawing group to help direct the regiochemistry of the initial Heck reaction. It was hoped
that the iodo bromo substrate would give the highest possible stereocontrol, by virtue of a
regiocontrolled initial intermolecular Heck reaction. Surprisingly, the one pot double Heck
ACS Paragon Plus Environment
9

Organic Process Research & Development
Page 10 of 30
1
2
3
4
5
6
7
8
9
reaction required higher temperature (145ºC) than the dibromo analogs 11 and 12. Further study
showed that the first Heck reaction proceeded at only 60ºC due to the more reactive aryl iodide
9, and the second step (intramolecular rection) was the slow step. If the initial Heck product 14
was isolated, and then resubjected to Heck conditions in a separate step, the second Heck
reaction proceeded at 120ºC in higher overall yield (77 versus 61% during development), and
provided acrylate 10 in higher purity (99.2 versus 96.4 area%). Spiking studies showed that
iodide was a catalyst poison for the second Heck reaction, and this explains the higher
temperature required for the one pot process. The sequential double Heck approach provided
good control of the alkene stereochemistry and delivered dibenzoxepine 10 as shown in Scheme
5 during laboratory development.
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
Scheme 5. Preparation of 2 by Sequential and Double Heck Reactionsa
a) All yields are pilot plant isolated transformations except as noted.
ACS Paragon Plus Environment
10

Page 11 of 30
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Details for the overall optimization and plant scale production of 2 for each step in
Scheme 5 are as follows. Plant scale alkylation to provide ether 9 involved the use of aqueous
K2CO3 (0.7 g/1 mL) rather than the solid reagent in order to control a mild exotherm. The
aqueous K2CO3 solution was added dropꢀwise to initiate the reaction. After reaction completion,
the reaction was quenched by charging water directly to the reaction mixture. This approach for
preparing 9 provided 88.2% yield after isolation by filtration and drying.
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
Based on laboratory development, the sequential two step Heck reaction was selected for
pilot plant processing due to higher overall yield and catalyst performance. The conditions for
the intermolecular Heck reaction of 9 with ethyl acrylate proved sensitive to oxygen on gram
scale. The quality of catalyst also played an important role in the reaction performance. The
catalyst loading amount was screened, 1.5% w/w was enough for the reaction completion at 65ꢀ
70oC after 5 h. During lab trials of pilot plant materials, two separate reactions failed to reach
completion. In one case only 59% of 14 was observed after 2 h. An additional catalyst charged
only increased the inꢀprocess yield to 63%. When a similar pattern was observed for a second
reaction, the temperature was increased to 100ºC. Although an 82% yield of 14 was observed,
12% of compound 10 was also detected. The increased reaction temperature may have
contributed to formation of compound 10. After investigating the reaction parameters, two
different lots of Pd(OAc)2 catalyst performed without any of the problems observed previously.
It was concluded that catalyst lot variability was the problem. Interestingly, up to 20% water
content (based on w/w of 9) did not negatively impact the reaction performance. Compound 14
was consistently obtained in 93% yield with satisfactory Pd(OAc)2 catalyst. With isolated
compound 14 free from residual iodide or ethyl acrylate, the second sequential Heck reaction
(intramolecular) proceeded smoothly to afford cyclized acrylate 10. The plant scale conversion
ACS Paragon Plus Environment
11

Organic Process Research & Development
Page 12 of 30
1
2
3
4
5
6
7
8
9
of iodo bromo ether 14 to ester 10 proceeded in only 53% overall yield, but afforded 10 with
high purity (99.7 area %) after crystallization. Further development is needed to understand the
lower plant yield relative to higher yields (77%) observed during development of the reaction.
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
Initially ester 10 was hydrolyzed to the acrylic acid, which was isolated and treated with
NBS and catalytic LiOAc per the procedure developed by Roy and coworkers.10 Although this
provided the vinyl bromide 2 in high yield and with retention of stereochemistry, isolation of the
acrylic acid proved unnecessary. Ester hydrolysis with excess LiOH (Li+ carboxylates react
more rapidly with NBS than Na+ carboxylates per Roy10) and adjustment of the pH with HOAc
gave the Li carboxylate in situ, and this reacted cleanly to afford 2 in a oneꢀpot process. If the
pH was not adjusted with HOAc, the NBS was decomposed by excess LiOH and lower
conversion resulted.
2. Synthesis of aryl bromide 3 salt. Arylbromide 3 was prepared by a five steps process
starting with the Boc protected, chiral Rꢀ(ꢀ) alaninol (Scheme 6). The alcohol was converted to
mesylate 15. Reacting 15 with morpholine gave compound 16 which was deprotected and an
ammonium dihydrochloride salt 17 was isolated. During laboratory development, all three steps
were telescoped due to weak UV adsorption of the intermediates which caused a challenge for in
process analytical monitoring. For example, HPLC, TLC and 1H NMR monitoring all provided
inconsistencies for the reaction end point. In the formation of 15, HPLC indicated an incomplete
reaction compared to 1H NMR. Focusing on the methyl doublet shift, 1H NMR was utilized for
reaction monitoring. A brief solvent survey indicated that EtOAc was best for all three reactions.
Some decomposition of the intermediates was noted with CH2Cl2 solutions. In 2ꢀMeTHF, a
challenging emulsification was observed during aqueous extraction. Phase separation at times
took several hours. Neither decomposition nor emulsification was observed with EtOAc.
ACS Paragon Plus Environment
12

Page 13 of 30
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
However, preparation of the hydrochloride salt after removal of Boc protecting group indicated
that EtOAc was not the best solvent for preparing 17. Only a 42% overall yield of the threeꢀ
steptransformation was obtained when EtOAc was utilized as solvent. During salt formation in
EtOAc, some loss of 17 was observed in the filtrate. Addition of two volumes of MTBE avoided
product filtration lost The three step telescoped transformations: mesylate formation, amination
of morpholine and Boc deprotection were robust providing diꢀHCl salt 17 in 60% 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
The SNAr reaction of 17 and 4ꢀbromoꢀ1ꢀfluoroꢀ2ꢀnitrobenzene in EtOAc and Et3N was
robust providing freebase 3 in very good yield (93%). CaCl2 dihydrate was used as a fluoride
trapping agent. Performing the reaction in other solvents provided lower yields. For example 2ꢀ
MeTHF under the same conditions as EtOAc provided <88% yield. For the Step 4 SNAr
reaction, it was important to minimize the levels of other nucleophilic amines (e.g. alaninol, from
BOC cleavage in Step 1, and morpholine, from Step 2), as these could compete with amine 17 in
Step 4. The levels of these other amines were minimized by careful monitoring and control in
the steps preceding Step 4.
Compound 3 as a free base is not a solid. Therefore, the first generation synthesis
utilized a crystalline HBr salt. However, forming the HBr salt using a solution of HBr in acetic
acid was corrosive to plant equipment. Evaluation of possible acid salts of 3 that were
crystalline involved oxalic acid, tartaric acid and fumaric acid. Among these salts, the oxalate
salt 18 (oxalate salt of 3)gave the best purity, and minimized product loss during isolation. The
formation of 18 was also conducive to EtOAc as solvent. A 2ꢀMeTHF solution of oxalic acid
was added to the EtOAc solution of free base of compound 3. Therefore, the salt formation was
conducted in a mixture of 2ꢀMeTHF and EtOAc. EtOAc was utilized as a final rinse of the salt.
ACS Paragon Plus Environment
13

Organic Process Research & Development
Page 14 of 30
1
2
3
4
5
6
Since the oxalate salt provided similar quality (99.7% purity) to the hydrobromide salt but was
not corrosive to plant equipment, it was the salt of choice for future large scale processing.
7
8
9
Scheme 6. Preparation of Aryl Bromide 3 Oxalate Salt (18)a
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
a) all yields are isolated pilot plant transformations.
3. Second generation synthesis of 1. Having a robust pilot plant syntheses of key
intermediates 2 and 3 (as the oxalate salt 18), we turned our attention to the final API sequence
(Scheme 7). In the first generation synthesis, the organometallic transformations of Steps 1 and
Steps 2 were performed in dioxane with catalytic palladium at 80ꢀ85oC. Each reaction required a
different catalyst system. Our strategy was to perform a single workꢀup. For optimization, the
two reactions were streamlined in dioxane such that isolation and a separate workꢀup for each
step would be avoided.
ACS Paragon Plus Environment
14

Page 15 of 30
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Vinyl bromide 2 was converted to a vinyl boronate ester 4 by a Miyaura borylation
reaction. It was found that addition of trace H2O (approximately 1 w/w % relative to KOAc) was
important for reproducible reaction rates, especially on larger scale. It is believed that on lab
scale adequate H2O is present to provide enhanced KOAc solubility, but on larger scale where
less adventitious moisture is present, a spike of H2O is necessary to aid in solubilization of
KOAc. Filtration of the reaction mixture to remove solids was the only purification undertaken
before a Suzuki crossꢀcoupling with aryl bromide 18, catalytic Pd(OAc)2, PPh3, one equivalent
of K2CO3, water and no additional solvent. Compound 6 was isolated by an EtOAc/water
extractive work up. A crystallization with MTBE and drying provided 6 in 56 % yield over two
steps.
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 nitro reduction of 6 utilized 5% PtꢀC, EtOAc as solvent, Et3N as an additive and 50
psi H2atmosphere at 30ꢀ40oC. Although there is no rationale for Et3N impacting the reaction,
there is a reaction rate association. For example, performing the nitro reduction with 0.1
equivalents of Et3N the reaction is completed in 20 h. Under the same reaction conditions except
without Et3N, 10 % of the unreduced nitro group was observed. A total of 48 h was required for
reaction completion. As the subsequent guanidine ring forming step for 7 was performed in
EtOAc, the completed hydrogenation reaction solution was filtered and used directly in the next
step.
Commerically available diphenyl cyanocarbonimidate11 was reacted with diamine 7 in
EtOAc solution from the previous reaction at reflux (75ꢀ80oC) for 20 h. After completion of the
reaction, the mixture was concentrated, crystallized from MTBE and dried to provide
cyanoguanidine 8 as a crude intermediate. At this stage, a metal scavenger was utilized to
remove any residual metal since the three prior steps had utilized Pd or Pt metals. Silica thiol
ACS Paragon Plus Environment
15

Organic Process Research & Development
Page 16 of 30
1
2
3
4
5
6
7
8
9
was selected as the metal scavenger of choice. However, 8 was not readily soluble in EtOAc.
The solubility of 8 in EtOAc at ambient temperature was only 15 g/L. A solvent screen indicated
that CH2Cl2 provided enhanced solubility at 100 g/L. Therefore, 8 was dissolved in CH2Cl2 and
treated with silica thiol (5% w/w) for residual metal removal. The resulting residual Pd was 26
ppm while residual Pt was <10 ppm. After residual metal removal, 8 was recrystallized from
MTBE to provide an 80% isolated 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
The final transformation in the second generation synthesis involved hydrolysis of the
cyanamide functionality of 8 to the indazole urea 1. A major development challenge was
circumventing complete hydrolysis to guanidine 19. The original hydrolysis procedure involved
addition of 8 as a solid to a CH2Cl2/ TFA/water mixture at 0ꢀ25oC for 3 h. In an effort to
improve the process, a solution of 8 during addition to the reactor was preferred. A solubility
screen consisting of THF, AcOH and CH2Cl2 indicated that CH2Cl2 or THF provided increased
solubility of 8. Upon further development, the rate of hydrolysis in THF as solvent was much
slower when compared to CH2Cl2. The reaction time was 10 h in THF vs 3 h in CH2Cl2. Due to
concern that longer reaction time may lead to increased hydrolyzed byꢀproducts, CH2Cl2 was
selected as solvent. The optimized reaction involved forming an aqueous CH2Cl2 mixture of 8 to
which TFA was added at 0oC and the hydrolysis reaction was allowed to proceed for 3 h, at
ambient temperature. Extractive work up and crystallization with EtOAc followed by drying
resulted in 89 % yield of 1.
ACS Paragon Plus Environment
16

Page 17 of 30
Organic Process Research & Development
1
2
3
4
Scheme 7. Second Generation Synthesis Routea
5
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
a) all yields are pilot plant isolated transformations.
CONCLUSION
A highly convergent pilot plant synthesis for MR antagonist LY2623091 (1) was
developed. The synthesis is based on both the first generation synthesis approach and
preparation of related MR analogs. However, the first generation synthesis did not posess
complete stereocontrol of the triꢀsubstituted olefin. Incorporating and developing a double Heck
reaction for geometric control of the Eꢀolefin isomer was key to the synthesis success. Execution
of the sequential Heck reaction on a pilot plant scale provided complete control of the
stereoisomer in moderate yield.
ACS Paragon Plus Environment
17

Organic Process Research & Development
Page 18 of 30
1
2
3
4
5
6
7
8
9
In addition to the geometric isomer control, a chiral alaninol derivative as a building
block was prepared. An important HBr to oxalate salt switch was required to avoid plant
equipment corrosion from the HBr salt. Other methodologies developed for synthesis success
include Suzuki crossꢀcoupling, amination, nitro reduction, indazole formation and hydrolysis of a
challenging cyanamide. Overall, multiꢀkilogram quantities of API were obtained to support
clinical trials.
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
General. All commercial reagents were used as received without further purification. Known
compounds were prepared according to reported procedures or were purchased from commercial
sources. Melting points were determined by DSC using TA Instruments Discovery Series SA+
Differential Scanning Calorimeter, running the latest version of TRIOS software. IR spectra
were recorded with Nicolet 6700 FTꢀIR with Smart OMNI sampler. All NMR data were
collected with a 400 MHz Varian Spectrometer ATB probe. HRMS analyses were performed
with a Thermo LTQ Orbitrap Discovery, using Electrospray ionization (ESI) technique, in the
positive ion mode. Analytical methods for purity (wt% and area% impurities) was developed
utilizing an HPLC (Shimadzu LCꢀ20A HPLC with PDA detector or equivalent) and measured
against an authentic external standard. Assay: Shimadzu LCꢀ20A HPLC with PDA detector or
equivalent, Waters XBridge Phenyl, 4.6x150 mm, 3.5 um, 40ºC, 1.5 mL/min; 240 nm; 5 uL
injection; A:0.2% Perchloric acid in 95 : 5 MilliQ water : MeCN; B: MeCN; gradient: 0% B to
74% B in 20 mins; to 100% B in 1 min; hold at 100% B for 3 mins; Diluent: MeCN.
1-Bromo-4-fluoro-2-((2-iodobenzyl)oxy)benzene (9). DMF (320 L), 2ꢀbromoꢀ5ꢀ
fluorophenol, 39.4 kg, 206 mol), and 2ꢀiodobenzyl bromide, (60.0 kg, 202 mol) were combined
at 15ºC to 25ºC under an inert atmosphere. K2CO3 (44.4 kg, 321 mol) was dissolved in water
ACS Paragon Plus Environment
18

Page 19 of 30
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
(60 L) and added to the DMF mixture at 15ºC to 30ºC over 1.5 h. After stirring for 12.5 h, assay
showed no 2ꢀiodobenzyl bromide. Water (603 L) was added over 2 h, and the mixture stirred for
another 2 h. The slurry was filtered and washed with water, before the wet cake was dried,
resulting in 9 (73.6 kg, 98.6 wt%, 178 mol, 88.2% yield). Mp 89.8ꢀ92.3 ºC. IR (neat, ATR):
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
3072 (w), 1482 (s), 1451 (s), 1294 (s), 1294 (s) cmꢀ1. H NMR (399 MHz, DMSOꢀd6) δ 5.12 (s,
2H), 6.81 (td, J = 8.49, 2.77 Hz, 1H), 7.14 (td, J = 7.64, 1.65 Hz, 1H), 7.18 (dd, J = 10.90, 2.82
Hz, 1H), 7.46 (td, J = 7.52, 0.92 Hz, 1H), 7.60 (dd, J = 7.64, 1.41 Hz, 1H), 7.62 (dd, J = 8.66,
6.23 Hz, 1H), 7.92 (dd, J = 7.83, 0.83 Hz, 1H). 13C NMR (100 MHz, DMSOꢀd6) δ 74.5, 99.2,
102.4 (d, J = 27.1 Hz), 105.8 (d, J = 3.4 Hz), 108.9 (d, J = 22.5 Hz), 128.5, 129.8, 130.3, 133.6
(d, J = 9.9 Hz), 138.0, 139.2, 155.4 (d, J = 10.7 Hz), 162.2 (d, J = 244.3 Hz). GCMS: m/z
[M]+ calcd for C13H9BrFIO: 405.88600; found: 405.88620.
Ethyl (E)-3-(2-((2-bromo-5-fluorophenoxy)methyl)phenyl)acrylate (14). Nꢀmethylꢀ2ꢀ
pyrrolidone (778 L), compound 9 (73.4 kg, 177 mol), NaOAc (23.0 kg, 280 mol), and TBAB
(59.0 kg, 183 mol) were combined. The mixture was degassed by sparging N2 gas through the
mixture for 1 h. Pd(OAc)2 (1.17 kg, 5 mol) was placed in a charge box, degassed with N2 gas
three times, before adding it, and heating the mixture to 60ºC to 65ºC. Separately, Nꢀmethylꢀ2ꢀ
pyrrolidone (14.7 L) was combined with ethyl acrylate (19.9 kg, 199 mol) and added to the
mixture at 60ºC to 65ºC over 1.25 h. The combined mixture was then heated to 70ºC to 75ºC for
5 h. Assay showed nonꢀdetect level of 9. The mixture was cooled to 0ºC to 5ºC before adding
this mixture to another vessel containing water (1854 L) at 0ºC to 10ºC over 13.5 h. The
combined mixture was adjusted to 20ºC to 30ºC and stirred for 2.5 h. The slurry was filtered in
three sections, washing each section with water (50 – 60 L). The reactor was rinsed with water
(90 L) to completely transfer the product. The combined wet cake (203 kg) was returned to the
ACS Paragon Plus Environment
19

Organic Process Research & Development
Page 20 of 30
1
2
3
4
5
6
7
8
9
reactor and slurried in water (750 L) for 1 h, before filtering, washing with water, and drying at
50ºC to 55ºC for 58 h. This resulted in 14 (70.2 kg, 89.3 wt%, 165 mol, 93.0 % yield). Mp 84.9ꢀ
89.4 ºC. IR (neat, ATR): 2979 (w), 1711 (vs), 1314 (vs), 1296 (vs), 1290 (s), 1184 (vs), 1167
(s) cmꢀ1. 1H NMR (399 MHz, DMSOꢀd6) δ 1.20 (t, J = 7.10 Hz, 3H), 4.15 (q, J = 7.10 Hz, 2H),
5.35 (s, 2H), 6.55 (d, J = 15.86 Hz, 1H), 6.80 (td, J = 8.47, 2.82 Hz, 1H), 7.30 (dd, J= 11.00,
2.72 Hz, 1H), 7.44 (quind, J = 7.25, 1.46 Hz, 2H), 7.60 (t, J = 6.62 Hz, 1H), 7.61 (d, J = 6.42 Hz,
1H), 7.84 (dd, J = 7.35, 1.41 Hz, 1H), 7.94 (d, J = 15.86 Hz, 1H). 13C NMR (100 MHz, DMSOꢀ
d6) δ 14.1, 60.0, 68.9, 102.8 (d, J = 27.3 Hz), 105.9 (d, J = 3.4 Hz), 108.9 (d, J = 22.5 Hz), 120.2,
127.1, 129.0, 129.8, 130.1, 133.5 (br d, J = 9.7 Hz), 133.5, 134.8, 141.0, 155.3 (d, J = 10.9 Hz),
162.1 (d, J = 244.1 Hz), 165.9. HRMS: m/z [M+H]+ calcd for C18H17BrFO3: 379.03396;
found: 379.03436.
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
Ethyl (E)-2-(3-fluorodibenzo[b,e]oxepin-11(6H)-ylidene)acetate (10). Nꢀmethylꢀ2ꢀ
pyrrolidone (723 L), compound 14 (67.5 kg, 159 mol), NaOAc (22.0 kg, 268 mol), and TBAB
(58.0 kg, 180 mol) were combined in a preꢀtreated reactor and degassed by bubbling with N2 gas
for 1 h. Pd(OAc)2 (1.45 kg, 6.47 mol) was added. The atmosphere was exchanged with N2gas
three times before heating the mixture to 115ºC to 125ºC for 5 h. Assay showed 0.5% 14. The
mixture was adjusted to 20ºC to 30ºC. In a separate tank was added water (829 L), adjusted to
20ºC to 30ºC. The reaction mixture was added to the water over 0.75 h. EtOAc (783 L) was
added to the mixture. After stirring for 0.75 h, the phases were separated. The aq phase was
back extracted with EtOAc two times (610 L, followed by 636 L). The combined organic
extracts were washed with brine (457 L). Activated charcoal (16.5 kg) in EtOAc (31 L) was
added, and the mixture was stirred at 20ºC to 30ºC for 4 h, before filtering over diatomaceous
earth (12.3 kg) and rinsing with EtOAc (110 L). The filtrate was concentrated under reduced
ACS Paragon Plus Environment
20

Page 21 of 30
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
pressure at < 35ºC (to 110 L) before methanol (446 L) was added. This mixture was then
concentrated under reduced pressure at < 35ºC (to 213 L). Assay showed 0.1% EtOAc. The
mixture was heated to 60ºC to 70ºC and held for 1 h to ensure dissolution of the solids. Nꢀ
heptane (801 L) was added over 7 h to the mixture at 60ºC to 70ºC. After stirring for 1 h at this
temperature, the slurry was cooled to 25ºC to 30ºC over 4 h, cooled to ꢀ10ºC to ꢀ5ºC over 3 h,
and held at this temperature for 14 h. The slurry was filtered, washing the wet cake with nꢀ
heptane (133 L). The wet cake was dried under vacuum at 35ºC to 40ºC for 36 h. This resulted
in 10 (27.9 kg, 97.5 wt%, 99.7 area%, 91 mol, 57.4% yield). Mp 104.5ꢀ105.4 ºC. IR (neat,
ATR): 2989 (vw), 1716 (vs), 1287 (vs), 1173 (vs), 1165 (vs) cmꢀ1. 1H NMR (399 MHz,
DMSOꢀd6) δ 1.04 (t, J = 7.10 Hz, 3H), 3.97 (q, J = 7.10 Hz, 2H), 5.20 (br s, 2H), 6.36 (s, 1H),
6.66 (dd, J = 10.51, 2.63 Hz, 1H), 6.80 (td, J = 8.37, 2.63 Hz, 1H), 7.22 (dd, J = 7.44, 0.92 Hz,
1H), 7.34 (td, J = 7.44, 1.36 Hz, 1H), 7.39 (td, J = 7.37, 1.31 Hz, 1H), 7.45 (dd, J = 8.81, 6.86
Hz, 1H), 7.49 (dd, J = 7.30, 1.07 Hz, 1H). 13C NMR (100 MHz, DMSOꢀd6) δ 13.8, 59.7, 69.5,
105.9 (d, J = 24.4 Hz), 108.5 (d, J = 21.6 Hz), 120.1, 121.6 (d, J = 3.1 Hz), 127.5, 128.0, 128.5,
128.8, 131.6 (d, J = 9.9 Hz), 131.8, 140.0, 152.7, 156.7 (d, J = 12.4 Hz), 163.4 (d, J = 246.4 Hz),
164.7. HRMS: m/z [M+H]+ calcd for C18H16FO3 : 299.10780; found: 299.10798.
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
(E)-11-(Bromomethylene)-3-fluoro-6,11-dihydrodibenzo[b,e]oxepine (2). Isopropanol
(284 L), water (110 L), 11 (27.9 kg, 97.5 wt%, 91 mol), and LiOH.H2O (8.0 kg, 190 mol) were
combined and heated to 80ºC to 85ºC for 2 h. Assay showed non detect levels of 11. While at
40ºC to 45ºC, acetic acid (11.2 kg, 186 mol) was added over 0.7 h, rinsing with isopropanol (7.6
L). The mixture was stirred for 1 h at 40ºC to 45ºC before Nꢀbromosuccinimide (17.7 kg, 99
mol) was added in portions over 1.5 h to maintain 40ºC to 45ºC. After cooling to 20ºC to 30ºC,
the mixture was stirred for 7.5 h. 4% aq NaHSO3 (73 L, 28 mol) was added to quench the
ACS Paragon Plus Environment
21

Organic Process Research & Development
Page 22 of 30
1
2
3
4
5
6
7
8
9
reaction over 0.75 h and stirred for another 1 h at 20ºC to 30ºC. 7% aq NaHCO3 (70 L, 61.6
mol) was added over 0.5 h and stirred for another 1 h at 20ºC to 30ºC. Water (261 L) was added
over 0.5 h and the mixture was stirred for another 1 h at 20ºC to 30ºC. The slurry was filtered,
washing the wet cake with aq isopropanol (29 L). The wet cake was dried at 35ºC to 40ºC for 29
h. This resulted in 2 (27.2 kg, 96.4 wt%, 98.4 area%, 85 mol, 94.0% yield) as a white powder.
Mp 115.8ꢀ116.8 ºC. IR (neat, ATR): 3060 (vw), 1492 (vs), 1462 (s), 1249 (vs), 1160 (s), 1145
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
(s), 1025 (s), 817 (vs), 791 (s) cmꢀ1. H NMR (399 MHz, DMSOꢀd6) δ 5.03 (br s, 1H), 5.33 (br
s, 1H), 6.64 (dd, J = 10.61, 2.63 Hz, 1H), 6.78 (td, J = 8.37, 2.63 Hz, 1H), 7.13 (s, 1H), 7.36 (dd,
J = 8.66, 6.91 Hz, 1H), 7.38 (dd, J = 7.01, 1.95 Hz, 1H), 7.39ꢀ7.42 (m, 1H), 7.44 (br td, J = 7.40,
1.40 Hz, 1H), 7.53 (dd, J = 6.76, 1.31 Hz, 1H). 13C NMR (100 MHz, DMSOꢀd6) δ 69.7, 106.2
(d, J = 24.4 Hz), 107.8, 108.7 (d, J = 21.4 Hz), 121.5 (d, J = 3.1 Hz), 128.4, 129.0, 129.2, 129.2,
132.2 (d, J = 9.9 Hz), 133.5, 140.4, 142.4, 156.2 (d, J = 12.4 Hz), 163.3 (d, J = 245.3 Hz).
HRMS: m/z [M+H]+ calcd for C15H11BrFO : 304.99718; found: 304.99730.
(R)-2-((tert-Butoxycarbonyl)amino)propyl methanesulfonate (15). tertꢀbutyl (R)ꢀ(1ꢀ
hydroxypropanꢀ2ꢀyl)carbamate, Bocꢀalaninol, (38.4 kg, 219 mol) was combined with EtOAc
(123 L) and Et3N (44.4 kg, 439 mol), forming a solution. This solution was cooled to 0ºC before
methanesulfonyl chloride (30.3 kg, 264 mol) in EtOAc (38 L) was added over 6 h, maintaining ꢀ
5ºC to +5ºC. After the addition was complete, the mixture was stirred for 2 h. Assay showed
0.04% Bocꢀalaninol remaining. The mixture was adjusted to 15ºC before the reaction was
quenched with 7% aq NaHCO3 (115 L). While maintaining 5ºC to 15ºC, the phases were
separated, and the organic phase was washed twice with brine (154 L). The organic phase was
kept < 10ºC, dried over Na2SO4 (50 kg) and filtered, rinsing with EtOAc (38 L).
ACS Paragon Plus Environment
22

Page 23 of 30
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
tert-Butyl (R)-(1-morpholinopropan-2-yl)carbamate (16). To the solution of 15 was added
morpholine (116 kg, 1330 mol). This solution was heated to 52ºC to 55ºC for 16 h. Assay
showed no 15 remaining. The solution was cooled to 20ºC to 25ºC, before it was washed twice
with 10% brine (154 L). The organic phase was concentrated (77 L – 115 L) at < 30ºC and
utilized as a solution in the next step.
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
(R)-1-Morpholinopropan-2-amine dihydrochloride (17). The solution of 16 was cooled to
0ºC to 5ºC before HCl in EtOAc (41 kg) was added over 4 h, maintaining 0ºC to 5ºC. After
stirring for 20 h at 0ºC to 5ºC, assay showed 0.4% 16. MTBE (162 L) was added over 3.5 h.
While at 0ºC to 3ºC, the slurry was stirred for 5 h, before it was filtered, washed with MTBE (39
L), and dried at 50ºC to 55ºC, resulting in 17 (28.6 kg, 132 mol, 60.2% yield over the three
1
steps). H NMR (399 MHz, acetonitrileꢀd3) δ 1.39 (d, J = 6.71 Hz, 3H), 3.12ꢀ3.18 (m, 2H), 3.28
(br dd, J = 14.60, 3.70 Hz, 3H), 3.49 (dd, J = 14.45, 8.51 Hz, 1H), 3.81ꢀ3.86 (m, 1H), 3.87ꢀ3.94
(m, 2H). 13C NMR (100 MHz, acetonitrileꢀd3) δ 17.7, 43.5, 44.2, 53.9, 60.7, 64.5, 64.6. HRMS:
m/z [M+H]+ calcd for C7H17N2O: 145.13354; found: 145.13335.
(R)-4-Bromo-N-(1-morpholinopropan-2-yl)-2-nitroaniline oxalate (18). Compound 17 (18.3
kg, 84 mol), calcium chloride (8.9 kg, 80 mol), 5ꢀbromoꢀ2ꢀfluoronitrobenze, 18.8 kg, 85 mol),
EtOAc (182 L), and Et3N (34.4 kg, 340 mol) were combined and heated to 72ºC for 26 h. Assay
showed 0.1% 17 remaining. The mixture was cooled to 20ºC to 25ºC and filtered. The filtrate
was washed with 20% aq NH4Cl (73 L) four times. Oxalic acid (8.5 kg, 94 mol) was dissolved
in 2ꢀmethyltetrahydrofuran (38 L) and added over 0.7 h to the product solution at 20ºC. This
slurry was stirred at 20ºC for 20 h, before it was filtered. The wet cake was dried at 40ºC to
45ºC for 16 h, resulting in 18 (27.6 kg, 94.7 wt%, 60 mol, 71.2% yield). Mp 191.7ꢀ192.2 ºC.
1
IR (neat, ATR): 3363 (w), 1617 (s), 1518 (s), 1280 (s), 1176 (s), 1154 (s), 1133 (s) cmꢀ1. H
ACS Paragon Plus Environment
23

Organic Process Research & Development
Page 24 of 30
1
2
3
4
5
6
7
8
9
NMR (399 MHz, DMSOꢀd6) δ 1.20 (t, J = 7.10 Hz, 3H), 4.15 (q, J = 7.10 Hz, 2H), 5.35 (s, 2H),
6.55 (d, J = 15.86 Hz, 1H), 6.80 (td, J = 8.47, 2.82 Hz, 1H), 7.30 (dd, J = 11.00, 2.72 Hz, 1H),
7.44 (quind, J = 7.25, 1.46 Hz, 2H), 7.60 (t, J = 6.62 Hz, 1H), 7.61 (d, J = 6.42 Hz, 1H), 7.84
(dd, J = 7.35, 1.41 Hz, 1H), 7.94 (d, J = 15.86 Hz, 1H). 13C NMR (100 MHz, DMSOꢀd6) δ 18.7,
44.9, 53.0, 61.9, 65.3, 105.4, 117.5, 128.0, 131.8, 138.8, 143.5, 162.1. HRMS: m/z [M+H]+
calcd for C13H19BrN3O3 : 344.06043; found: 344.06041.
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
(E)-2-((3-Fluorodibenzo[b,e]oxepin-11(6H)-ylidene)methyl)-4,4,5,5-tetramethyl-1,3,2-
dioxaborolane (4). Compound 2 (33.0 kg, 96.4 wt%, 104 mol), bis(pinacolato)diboron (30.5 kg,
120 mol), and dioxane (279 L) were combined and stirred until all the solids dissolved.
Tricyclohexylphosphine (1.9 kg, 6 mol), KOAc (11.7 kg, 119 mol), Pd2(dba)3 (1 kg, 1.0 mol),
and water (0.16 kg, 8.8 mol) were added. The mixture was heated at 80ºC to 85ºC for 27 h,
resulting in 8% residual 2. The mixture was cooled to 20ºC to 30ºC, filtered, washing the wet
cake with dioxane (59 L). Assay of the filtrate showed that it contained 4 (30.4 kg, 86 mol, 83%
yield).
(R,E)-4-((3-Fluorodibenzo[b,e]oxepin-11(6H)-ylidene)methyl)-N-(1-morpholinopropan-2-
yl)-2-nitroaniline (6). To the solution of compound 4 was added K2CO3 (39.4 kg, 285 mol),
water (140 L), and 18 (43.1 kg, 94.7 wt%, 94 mol). The mixture was stirred for 0.5 h before
PPh3 (0.07 kg, 0.27 mol) and Pd(OAc)2 (0.195 kg, 0.87 mol) were added. The mixture was
heated at 80ºC to 85ºC for 24 h. Assay showed 0.05% 4 remaining. After cooling to 20ºC to
30ºC, the mixture was diluted with EtOAc (334 L) and brine (92 L). After stirring for 1 h, the
mixture was filtered, washing the wet cake with EtOAc (50 L). The organic phase was
separated. The aqueous phase was extracted with EtOAc (60 L). The organic phases were
combined and concentrated. MTBE (176 L) was added, and the mixture concentrated . This
ACS Paragon Plus Environment
24

Page 25 of 30
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
was repeated with additional MTBE (136 L) and concentrating. MTBE (312 L) was added.
Assay showed 4.5% dioxane remaining in the organic phase. The mixture was stirred for 8 h at
20ºC to 30ºC, before the mixture was filtered, washing the wet cake with MTBE (48 L). The wet
cake was dried at 45ºC to 50ºC for 16 h, resulting in 6 (30.2 kg, 94.9 wt%, 98.4 area%, 58 mol,
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
56.2 % yield across the two steps) as a brown solid. H NMR (399 MHz, DMSOꢀd6) δ 1.15 (d, J
= 6.13 Hz, 3H), 2.38 (br t, J = 3.94 Hz, 4H), 2.40ꢀ2.47 (m, 2H), 3.50 (br t, J = 4.38 Hz, 4H), 3.90
(quin, J = 6.54 Hz, 1H), 4.93ꢀ5.15 (m, 1H), 5.50ꢀ5.70 (m, 1H), 6.61 (dd, J = 10.61, 2.63 Hz, 1H),
6.80 (td, J = 8.37, 2.72 Hz, 1H), 6.89 (d, J = 9.24 Hz, 1H), 6.93 (s, 1H), 7.02ꢀ7.09 (m, 2H), 7.28
(td, J = 7.54, 1.07 Hz, 1H), 7.35ꢀ7.42 (m, 1H), 7.57ꢀ7.63 (m, 2H), 7.86 (d, J = 1.95 Hz, 1H), 8.18
(br d, J = 6.62 Hz, 1H). 13C NMR (100 MHz, DMSOꢀd6) δ 18.8 (s), 45.2 (s), 53.6 (s), 63.0 (s),
66.2 (s), 69.6 (s), 105.5 (br d, J=24.4 Hz), 108.1 (br d, J=21.4 Hz), 114.7 (br s), 123.5 (s), 123.7
(d, J = 3.1 Hz), 126.8 (s), 127.4 (s), 128.1 (s), 128.4 (s), 128.9 (s), 129.4 (s), 130.5 (s), 131.1 (br
d, J = 9.2 Hz), 133.7 (s), 136.3 (s), 136.8 (br s), 141.1 (s), 143.6 (s), 156.0 (d, J = 12.2 Hz), 162.3
(d, J = 243.8 Hz). HRMS: m/z [M+H]+ calcd for C28H29FN3O4: 490.21366; found:
490.21238.
(R,E)-4-((3-Fluorodibenzo[b,e]oxepin-11(6H)-ylidene)methyl)-N1-(1-morpholinopropan-
2-yl)benzene-1,2-diamine (7). Compound 6 (29.5 kg, 94.9 wt%, 57 mol), EtOAc (297 L), Et3N
(0.6 kg, 5 mol) and 5% PtꢀC (1.5 kg, 0.4 mol) were combined, degassed twice with N2 and once
with H2, before being placed under 50 – 55 psi H2 and heating to 40ºC to 45ºC for 20 h. Assay
showed no 6 remaining. The mixture was cooled to 20ºC to 30ºC before the mixture was
filtered, rinsing the wet cake with EtOAc (74 L). Assay of the filtrate showed that it contained 7
(26.0 kg, 56 mol, 99% yield). IR (neat, ATR): 3302 (w br), 1516 (s), 1493 (s), 1299 (s), 1140
1
(s), 1116 (s), 1069 (s) cmꢀ1. H NMR (399 MHz, DMSOꢀd6) δ 1.07 (d, J = 6.13 Hz, 3H), 1.16 (t,
ACS Paragon Plus Environment
25

Organic Process Research & Development
Page 26 of 30
1
2
3
4
5
6
7
8
9
J = 7.10 Hz, 1H), 1.98 (s, 1H), 2.19 (br dd, J = 12.12, 7.15 Hz, 1H), 2.25ꢀ2.46 (m, 5H), 3.44ꢀ3.57
(m, 1H), 3.52 (t, J = 4.57 Hz, 3H), 4.02 (q, J = 7.10 Hz, 1H), 4.32 (s, 1H), 4.34 (br d, J = 6.62
Hz, 1H), 4.99 (br s, 1H), 5.53 (br s, 1H), 6.17 (dd, J = 8.37, 1.65 Hz, 1H), 6.23 (d, J = 8.47 Hz,
1H), 6.32 (d, J = 1.75 Hz, 1H), 6.56 (dd, J = 10.61, 2.63 Hz, 1H), 6.67 (s, 1H), 6.75 (td, J = 8.34,
2.68 Hz, 1H), 7.10 (d, J = 6.91 Hz, 1H), 7.26 (td, J = 7.52, 1.12 Hz, 1H), 7.33 (td, J = 7.42, 1.12
Hz, 1H), 7.46 (dd, J = 8.66, 7.01 Hz, 1H), 7.54 (d, J = 7.30 Hz, 1H). 13C NMR (100 MHz,
DMSOꢀd6) δ 14.1 (s), 19.3 (s), 20.7 (s), 44.8 (s), 53.6 (s), 59.7 (s), 63.9 (s), 66.2 (s), 69.5 (s),
105.3 (d, J = 24.0 Hz), 107.9 (d, J = 21.0 Hz), 109.6 (s), 115.6 (s), 119.5 (s), 124.6 (s), 124.9 (d,
J = 3.1 Hz), 127.7 (d, J = 5.0 Hz), 128.6 (s), 129.0 (s), 130.8 (d, J = 9.5 Hz), 131.3 (s), 132.4 (s),
133.3 (s), 134.7 (s), 134.9 (s), 142.3 (s), 155.7 (d, J = 11.8 Hz), 161.9 (d, J = 243.4 Hz. HRMS:
m/z [M+H]+ calcd for C28H31FN3O2 : 460.23948; found: 460.23868.
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
N-((E)-5-((E)-(3-Fluorodibenzo[b,e]oxepin-11(6H)-ylidene)methyl)-1-((R)-1-
morpholinopropan-2-yl)-1,3-dihydro-2H-benzo[d]imidazol-2-ylidene)cyanamide (8). To the
filtrate of 7 was added diphenyl Nꢀcyanocarbonimidate (14.5 kg, 60 mol). The mixture was
heated at reflux (76ºC) for 18 h. Assay showed 0.3% 7 remaining. The mixture was cooled to
20ºC to 30ºC and concentrated. MTBE (159 L) was added, and the mixture concentrated again.
MTBE (562 L) was added. Assay showed 1.5% EtOAc remaining. The mixture was stirred for
8 h, before the slurry was filtered, washing the wet cake with MTBE (43 L). The wet cake
(25.75 kg) was combined with CH2Cl2 (426 L) and stirred until a solution resulted. Silica thiol
(1.51 kg, 0.058 g / g 7) was added. The mixture was heated to 40ºC to 45ºC for 5 h, before the
mixture was cooled to 20ºC to 30ºC and filtered, washing the wet cake with CH2Cl2 (30 L). The
filtrate was concentrated MTBE (150 L) was added. This mixture was concentrated before
MTBE (566 L) was added. Assay showed 2.4% CH2Cl2. The slurry was stirred at 20ºC to 30ºC
ACS Paragon Plus Environment
26

Page 27 of 30
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
for 10 h and filtered, washing the wet cake with MTBE (81 L). The wet cake was dried at 45ºC
to 50ºC for 14 h, resulting in 8 (23.4 kg, 99.4 wt%, 45 mol, 79.6% yield across the two steps) as
an offꢀwhite solid. Mp 227.8ꢀ231.5 ºC. IR (neat, ATR): 2182 (s), 1637 (s), 1615 (s) cmꢀ1. 1H
NMR (399 MHz, DMSOꢀd6) δ 1.38 (d, J = 6.91 Hz, 3H), 2.12ꢀ2.23 (m, 2H), 2.40ꢀ2.47 (m, 2H),
2.93 (br dd, J = 12.80, 10.07 Hz, 1H), 3.31 (s, 2H), 3.33ꢀ3.39 (m, 4H), 4.58ꢀ4.75 (m, 1H), 5.06
(br s, 1H), 5.60 (br s, 1H), 6.62 (dd, J = 10.51, 2.63 Hz, 1H), 6.80 (td, J = 8.37, 2.63 Hz, 1H),
6.84ꢀ6.93 (m, 2H), 7.00 (d, J = 7.40 Hz, 1H), 7.02 (s, 1H), 7.25 (td, J = 7.54, 0.97 Hz, 1H), 7.34
(d, J = 8.47 Hz, 1H), 7.39 (td, J = 7.52, 1.02 Hz, 1H), 7.56ꢀ7.66 (m, 2H), 7.56ꢀ7.66 (m, 2H),
12.51 (br s, 1H). 13C NMR (100 MHz, DMSOꢀd6) δ 15.9, 47.2, 53.2, 59.9, 66.1, 69.5, 105.5 (br
d, J = 24.0 Hz), 108.1 (br d, J = 21.2 Hz), 110.2, 110.7, 117.5, 123.9 (d, J = 2.9 Hz), 124.2,
127.4, 128.4, 128.8, 129.3, 129.8, 130.7, 131.2 (br d, J = 9.7 Hz), 133.5, 136.9, 141.2, 154.1,
156.0 (d, J = 12.0 Hz), 162.3 (d, J = 244.1 Hz). HRMS: m/z [M+H]+ calcd for C30H29FN5O2 :
510.22998; found: 510.22876.
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-((E)-5-((E)-(3-Fluorodibenzo[b,e]oxepin-11(6H)-ylidene)methyl)-1-((R)-1-
morpholinopropan-2-yl)-1,3-dihydro-2H-benzo[d]imidazol-2-ylidene)urea (1). Compound 8
(23.0 kg, 99.4 wt%, 44 mol) was combined with CH2Cl2 (68 L), resulting in a slurry. Water (19
L) was added, and the slurry cooled to 0ºC to 5ºC. Trifluoroacetic acid (107 kg, 938 mol) was
added over 2.5 h, maintaining 0ºC to 5ºC. The mixture was warmed to 20ºC to 25ºC and stirred
for 4 h. Assay showed no compound 8 remaining. CH2Cl2 (229 L) was added, and the mixture
cooled to 0ºC to 10ºC. While maintaining this temperature, 5 N aq NaOH (172 kg, 141 L, 705
mol) was added, adjusting the pH to 8 to 9, resulting in an emulsion. The mixture was warmed
to 20ºC and brine (100 L) was added, removing the emulsion. The organic phase was separated.
The aq phase was extracted with CH2Cl2 (96 L). The organic phases were combined, washed
ACS Paragon Plus Environment
27

Organic Process Research & Development
Page 28 of 30
1
2
3
4
5
6
7
8
9
with brine (100 L), and dried over Na2SO4 (46 kg, 2 g / g, 8) by stirring for 4 h, resulting in a
K.F. of 0.08%. The mixture was filtered, washing the wet cake with CH2Cl2 (27 L). While
maintaining < 35ºC, the mixture was concentrated. EtOAc (124 L) was added, and the mixture
concentrated. More EtOAc (127 L) was added, and the mixture was concentrated. EtOAc (110
L) was added to adjust to the total volume. Assay showed 400 ppm CH2Cl2. Therefore, the
mixture was concentrated before EtOAc (106 L) was added to adjust the total volume. Assay
now showed 30 ppm CH2Cl2. The mixture was adjusted to 0ºC to 10ºC and stirred for 2 h. Seed
(200 g, 0.9 wt% seed) was added, and the mixture stirred at 0ºC to 10ºC for 12 h. The mixture
was concentrated and stirred for 2 – 3 h at 0 – 10ºC, and filtered, washing the wet cake with
EtOAc (39 L). The wet cake was dried at 45ºC to 50ºC for 37 h, resulting in 1 (21.2 kg, 99 wt%,
99.5 area%, 1.04% EtOAc, 39 mol, 88.6% yield) as an offꢀwhite solid. Mp 200.8ꢀ207.8 ºC. IR
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
(neat, ATR): 3468 (vw br), 1697 (vs), 1581 (vs), 1491 (vs), 1437 (s) cmꢀ1. H NMR (399 MHz,
DMSOꢀd6) δ 1.38 (br d, J = 6.91 Hz, 1H), 1.43 (d, J = 6.81 Hz, 2H), 2.17ꢀ2.29 (m, 2H), 2.36ꢀ
2.47 (m, 1H), 2.62 (dd, J = 13.38, 5.01 Hz, 1H), 2.78 (dd, J = 13.33, 7.88 Hz, 1H), 3.34ꢀ3.40 (m,
1H), 3.41ꢀ3.50 (m, 3H), 4.72ꢀ4.85 (m, 1H), 4.86ꢀ4.97 (m, 1H), 4.99ꢀ5.17 (m, 1H), 5.48ꢀ5.72 (m,
1H), 5.87ꢀ6.10 (m, 1H), 6.61 (dd, J = 10.51, 2.63 Hz, 1H), 6.80 (td, J = 8.30, 2.68 Hz, 2H), 6.94ꢀ
7.09 (m, 3H), 7.23 (br t, J = 7.54 Hz, 1H), 7.30ꢀ7.41 (m, 2H), 7.54ꢀ7.62 (m, 2H), 9.93 (s, 1H),
11.72 (s, 1H). 13C NMR (100 MHz, DMSOꢀd6) δ 16.7, 17.3, 49.2, 53.8, 54.1, 61.3, 66.4, 66.6,
70.1, 105.9 (d, J = 24.4 Hz), 108.5 (d, J = 21.0 Hz), 111.0, 118.0, 124.7 (d, J = 2.6 Hz), 128.0,
128.6, 129.3, 129.7, 130.1, 131.3, 131.5 (br d, J = 9.5 Hz), 131.6, 134.0, 136.1, 136.6, 140.9,
142.2, 149.6, 155.2, 156.4 (d, J = 11.8 Hz), 162.7 (br d, J = 243.8 Hz), 165.1. HRMS: m/z
[M+H]+ calcd for C30H31FN5O3 : 528.24054; found: 528.23986.
ACS Paragon Plus Environment
28

Page 29 of 30
Organic Process Research & Development
1
2
3
4
ACKNOWLEDGMENT
5
6
7
8
Contributions from the following are greatly appreciated: Venkatraghavan Vasudevan, Steven
Sproull, Jeff Roberts, Scott Mauldin, WuXi PharmaTech, Carbogen and Cambridge Majors
Laboratories.
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
FUNDING
All research funding sources for all authors were provided by Eli Lilly and Company, Lilly
Research Laboratories, Indianapolis, Indiana 46285.
SUPPORTING INFORMATION
1
Supplementing information include a general preparative experimental conditions, NMR and
13NMR for compounds 1, 2, 6, 7, 8, 9, 10, 14, 17, and 18. IR spectrum for compounds 1, 2, 7, 8,
9, 10, 14, and 18 are also included as supplementing information.
REFERENCES
1. a) Glicklich, D.; Frishmen, W. H. Drugs, 2015, 7, 473ꢀ485. b) Lavall, D.; Selzer, C.;
Schuster, P.; Lenski, M.; Adam, O.; Schäfers, HꢀJ.; Böhm, M.; Laufs, U. J. Biol Chem. 2014,
289, 6656ꢀ6668. c) Nagase, M. Clin Exp Nephrol. 2010, 14, 303ꢀ314.
2. a) Kolkhof, P.; Borden, S. A. Mol. Cell. Endocrinol. 2012, 350, 310ꢀ317. b) Luther, J. M.
Curr. Opin. Nephrol. Hy. 2014, 23 , 456ꢀ461.
3. a) Kowala, M. C. PCT Int. Appl. WO 2015/130568 A1. b) Gavardinas, K.; Jadhav, P. K.
U.S. Pat. Appl. US20090163472 A1.
ACS Paragon Plus Environment
29