Multikilogram Synthesis of a Potent Dual Bcl-2/Bcl-xL Antagonist. 2. Manufacture of the 1,3-Diamine Moiety and Improvement of the Final Coupling Reaction

Article abstract of DOI:10.1021/acs.oprd.9b00367

This paper describes the synthesis of kilogram quantities of the sulfonamide moiety 3 involved in a coupling reaction with acid moiety 2 to provide batches of drug candidate 1 for preclinical studies and first-in-human clinical trials. A first approach relying on a chiral separation furnished the desired enantiomer of 1,3-diamine 20, precursor of sulfonamide 3. An enantiomeric synthesis of 20 using the Ellman's chiral auxiliary coupled with an aza-Reformatsky reaction to control the stereochemistry is also discussed. Coupling conditions of the final step involving EDCI to provide 1 under a cGMP process are detailed. An alternative approach using N-(1-methanesulfonyl)benzotriazole is also presented.

Full text of DOI:10.1021/acs.oprd.9b00367

Subscriber access provided by Karolinska Institutet, University Library
Full Paper
Multikilogram synthesis of a potent Bcl-2 selective
antagonist. Part II. Manufacture of the 1,3-diamine
moiety and improvement of the final coupling reaction.
Christophe HARDOUIN, Sandrine Baillard, François Barière, Anthony Craquelin, Mathieu
Grandjean, Solenn Janvier, Stéphane Le Roux, Christine Penloup, and Olivier Russo
Org. Process Res. Dev., Just Accepted Manuscript • DOI: 10.1021/acs.oprd.9b00367 • Publication Date (Web): 19 Nov 2019
Downloaded from pubs.acs.org on November 19, 2019
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 46
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Multikilogram synthesis of a potent Bcl-2 selective
antagonist. Part II. Manufacture of the 1,3-diamine
moiety and improvement of the final coupling
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
Christophe Hardouin,* Sandrine Baillard, François Barière, Anthony Craquelin, Mathieu
Grandjean, Solenn Janvier, Stéphane Le Roux, Christine Penloup, Olivier Russo
Industrial Research Centre, Oril Industrie, 13 rue Desgenétais, 76210 Bolbec, France.
hardouin.christophe@yahoo.com
ACS Paragon Plus Environment

Organic Process Research & Development
Page 2 of 46
1
2
3
4
Table of Contents Graphic
5
6
7
8
9
23, EDCI, DMAP,
DMF reflux
Racemic
approach
O
Cl
O
O
O
cGMP process
Batch of up to 5.7 kg
Cl
-
O
O
S
NO2
NH
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
S
NO2
H2
N
Na+
N
NH
Cl
N
S
Me
23, BtSO2Me, Et3N,
N
3
Enantioselective
approach
S
Ph
NH2
1
OEt
OEt
NaOH, MeTHF, 50 °C
Me
Br
On 15 g scale
Chemical purity >99 area %
ACS Paragon Plus Environment

Page 3 of 46
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
ABSTRACT: This paper describes the synthesis of kilogram quantities of the
sulfonamide moiety 3 involved in a coupling reaction with acid moiety 2 to provide
batches of drug candidate 1 for preclinical studies and first-in-human clinical trials. A
first approach relying on a chiral separation furnished the desired enantiomer of 1,3-
diamine 20, precursor of sulfonamide 3. An enantiomeric synthesis of 20 using the
Ellman’s chiral auxiliary coupled with an aza-Reformatsky reaction to control the
stereochemistry is also discussed. Coupling conditions of the final step involving EDCI
to provide 1 under a cGMP process are detailed. An alternative approach using N-(1-
methanesulfonyl)benzotriazole is also presented.
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: Bcl proteins, acylsulfonamide inhibitors, enantiopure 1,3-diamine,
telescoped process.
ACS Paragon Plus Environment

Organic Process Research & Development
Page 4 of 46
1
2
3
4
5
6
Introduction
7
8
9
Finding innovative treatments to cure cancer has been a major challenge for the
scientific community over the past decades. The discovery that B-cell lymphoma 2 (Bcl-
2), a protein overexpressed in many types of cancer cells, promotes cells survival but not
cells proliferation demonstrated that impaired apoptosis is a critical step in tumor
development.1 Finding inhibitors to antagonize the activities of Bcl-2 family proteins has
been explored by medicinal chemists and resulted in the synthesis of small molecules
like ABT-737, ABT-263 and Venetoclax, the latter being approved by the FDA in 2016
(Figure 1).2 Despite encouraging results in terms of activity, ABT-737 suffered a
moderate bioavailability and ABT-263 some dose-limiting toxicity. To overcome those
issues, chemists at Servier identified compound 1 as promising drug candidate for the
treatment of Bcl-2 dependent malignancies such as leukemias and lymphomas.3 In a
previous article (Multikilogram synthesis of a potent Bcl-2 selective antagonist. Part I.),
we described the synthesis of the acid moiety that would ultimately be coupled with
sulfonamide 3.
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
This contribution describes the synthesis of sulfonamide moiety 3 following two
different paths to prepare enantiopure 1,3-diamine 20. Coupling conditions of the final
step were evaluated and improved to manufacture kilogram quantities of API 1 under a
cGMP process.
ACS Paragon Plus Environment

Page 5 of 46
Organic Process Research & Development
1
2
3
4
5
6
7
Figure 1. Structures of Bcl inhibitors
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
O
O
O
O
O
O
S
SO2CF3
NH
S
NO2
NH
N
H
N
H
N
N
Cl
N
Cl
N
S
S
Me
N
N
ABT-263
ABT-737
O
Me
O
O
O
O
O
O
O
-
S
NO2
NH
S
NO2
NH
N
Na+
N
H
N
N
Cl
N
Cl
N
S
Me
N
O
N
1
Venetoclax
Me
N
H
Results and Discussion
The strategy for the construction of compound 1 is outlined in the retrosynthesis shown
in Scheme 1. The acylsulfonamide motif could be prepared by coupling acid 2 and
sulfonamide 3. The C-N bond could be installed by reacting commercially available
halogenosulfonamides 21 or 22 with enantiopure 1,3-diamine 20, the latter focusing
most of our synthetic efforts. We wish to report herein the strategies developed to
synthesize sulfonamide 3 and to couple this key intermediate with carboxylic acid 2.
Scheme 1. Retrosynthetic analysis to access 1
ACS Paragon Plus Environment

Organic Process Research & Development
Page 6 of 46
1
2
3
4
5
6
7
8
O
O
O
-
S
NO2
NH
N
Na+
N
Cl
N
9
S
Me
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
Me
1
O
O
O
S
NO2
NH
H2 N
OH
N
Cl
N
S
Me
2
N
Me
3
O
O
NH2
S
NO2
X
H2 N
S
Me
N
Me
21 X= F
22 X= Cl
20
At the beginning of this project, precedents in the literature described the access to sulfonamide
3. For example a seven-step synthesis was developed using N-protected D-aspartic acid 4 as
chiral pool (Scheme 2).4 Another six-step approach was also reported starting with Fmoc-Asp(Ot-
Bu)-OH 5.5 However we faced some difficulty when attempting to reproduce the conditions
described in those two papers, especially during the scale up. Besides, the price of the starting
materials (unnatural amino acids) could be a burden when one thinks about developing a
commercial route. This prompted us to investigate alternative approaches allowing us to
manufacture intermediate 3 on large scale.
ACS Paragon Plus Environment

Page 7 of 46
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
1,3-Diamines are important chiral building blocks in the synthesis of bioactive compounds and
ligands.6 However, in contrary to vicinal diamines, limited methods for the direct synthesis of this
motif have been documented.7 We considered several alternative methods for the synthesis of 20
(Scheme 2). A first strategy was explored relying on the enantioselective alkylation of glycine
derivative 6 by phase transfer catalysis.8 Reacting 6 with KOH, n-Bu4NBr and an excess of
bromide derivative provided the desired intermediate 7 when R = Cl (degradation observed with
other corresponding N-derivatives). However when 7 was exposed to HNMe2, degradation
occurred. A second approach was investigated taking advantage of a proline-catalyzed
asymmetric -hydrazination.9 4-Bromo-1-butanal 8 was prepared starting from the corresponding
acid in a two-step sequence (BH3.THF reduction in THF followed by oxidation of the resulting
alcohol with TEMPO and bleach) and was treated with di-benzyl azodicarboxylate (DBAD) and
L-proline to provide intermediate 9 after subsequent reduction of the aldehyde with NaBH4.
Although this approach seemed appealing, the overall yield to synthesize 9 was low (15%),
starting from the commercially available acid precursor of 8. Functionalization of intermediate 9
or of the resulting amine prepared by hydrogenolysis of the N-N bond proved to be challenging
and forced us to envision a third approach. Mannich-type conditions were evaluated on ketone 10
using a mixture of paraformaldehyde, dimethylamine hydrochloride in acidic medium and
provided complex reaction mixtures that were not valuable. Finally acid chloride 12 was selected
as starting material; and we embarked on developing an alternative synthesis of key intermediate
3.
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 2. Existing and alternative approaches
ACS Paragon Plus Environment

Organic Process Research & Development
Page 8 of 46
1
2
3
Ph
N
CO2t-Bu
4
Ph
N
CO2t-Bu
Ph
6
5
6
7
8
Ph
+
X
X
7
Cl
R
Br
Cbz
7 steps
H N
HO2C
R = Cl, NHBoc, NMe2, NMe2.HBr
O
CO2H
9
4
H
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
Cbz
N
Cbz
O
O
S
NO2
NH
H2
N
OH
Br
Br
8
9
3
S
O
O
Ph
NH2
PhS
PhS
X
NMe2
11
10
Fmoc
6 steps
H N
New
approach
CO2t-Bu
O
HO2C
Cl
5
Cl
12
The synthesis of compound 3 starting from 12 is depicted in Scheme 3. Treatment of acid
chloride 12 with MeONHMe under Schotten-Baumann conditions provided Weinreb amide 13
with a better purity than commercially available batches.10 Nucleophilic substitution was carried
out with sodium thiophenolate in THF and yielded to 14 as an oil after evaporation to dryness. A
two-step sequence was then required to convert 14 into ketone 16: Grignard addition and
subsequent 1,4-addition of dimethylamine on 15. Despite many efforts, both yield and purity
stayed moderate and low temperature was required to limit chemical degradation. Besides, we
found that ketone 16 was not stable on storage (due to a putative retro-Michael reaction) and had
to be engaged in the next step quickly (loss of 8 area % by HPLC after 4 weeks at 5 °C).
Reductive amination was initially evaluated to install the amine function. However condensation
of -methylbenzylamine or NH3 in the presence of NaBH4, NaBH(OAc)3, NaBH3CN or H2 failed
to give the desired product. When condensation with MeONH2.HCl was performed in MeTHF,
oxime 17 was obtained after basic aqueous wash as an E/Z: 6/4 mixture of isomers. Because of
the moderate stability of 17, this step was then telescoped with a BH3.Me2S-mediated reduction
ACS Paragon Plus Environment

Page 9 of 46
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
and yielded to racemic amine 18 after concentration to dryness.11 To efficiently purify this
intermediate, treatment with fumaric acid in a mixture of EtOAc and EtOH provided the
difumarate salt with >99 area % purity by HPLC and with a 28% yield over three steps.12 Basic
treatment was required before chiral chromatography which furnished desired amine 20 with a
satisfactory enantiomeric excess of 98.2%. Aromatic nucleophilic substitution on compound 21
was then achieved in DMSO with the presence of DIEA and provided desired 3.
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 this sequence allowed us to manufacture several kg of 3, we identified few issues that
could hamper a future commercial route. Handling BH3.Me2S on large scale in the pilot plant was
challenging in terms of health and safety since it disturbed some workers (foul smell). Also
several crops were needed to isolate the whole amount of difumarate 19. DSC studies on
compound 21 showed that a suspected autocatalytic decomposition occurred at 160 °C with a
heat flow of 2004 J/g corresponding to a Tad of 1000 °C (see Supporting Information). Finally,
the isolation process was unsuitable for larger scale synthesis and furnished 3 with a moderate
yield. Improvement was highly desirable and we started to search for alternative conditions.
Scheme 3. First generation synthesis of 3
O
O
O
O
MeONHMe, K2CO3
MgCl
PhSNa
Cl
Cl
Me
PhS
Me
PhS
Cl
N
N
MTBE, H2O, 0 °C
82%
THF, 0 °C
Quant.
THF, -50 °C
OMe
OMe
12
13
14
15
Low temperature
required
OMe
O
N
NH2
HNMe2
MeONH2.HCl
MeTHF, 20 °C
BH3.Me2S
Reflux
20 °C
68%
S
S
S
Ph
NMe2
Ph
NMe2
Ph
NMe2
16
17
18
Stench
Moderate purity
Low stality on storage
ACS Paragon Plus Environment

Organic Process Research & Development
Page 10 of 46
1
2
3
NH2
NH2
NH2
CO2H
Chiral
4
5
6
2.
Fumaric acid
EtOAc, EtOH
NaOH, water
chromatography
EtOAc, 20 °C
97%
47%
S
S
S
CO2H
Ph
NMe2
Ph
NMe2
Ph
NMe2
28%
7
over 3 steps
19
18
20
8
Several crops necessary
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
O
O
S
NO2
NH
H2 N
O
O
21, DIPEA
S
NO2
F
H2 N
DMSO, 95 °C
60%
3
S
21
Ph
NMe2
Safety issues
Isolation process not satisfactory
A quick survey was carried out on an E/Z mixture of 17 and conversion to 18 was evaluated with
reducing agents such as BH3.THF, a mixture NaBH4/ZrCl4 and Dibal-H.13 BH3.THF complex
was not selected because of safety concerns on large scale and the couple NaBH4/ZrCl4 as well
(in that case, we noticed a major exotherm when mixing ZrCl4 and NaBH4 together and also
when quenching the reaction). When 3.2 equiv of Dibal-H were added (incomplete conversion if
less), encouraging results were obtained and this process was evaluated in a reaction calorimeter
(RC1) for safety reasons. As shown on Figure 2, a massive heat flow occurred at the beginning of
Dibal-H addition with a maximum of 216 W/kg of reaction mass that corresponded to a Tad of
100 °C. Extended addition time was therefore recommended (2.5 hours instead of 1.5 proposed
initially) so that the energy released could be overpowered by the feeding of the reactant in order
to match with our cooling capacities. Concomitant gas evolution was also observed (46 L/kg of
17) especially during the first 45 minutes of Dibal-H addition. DSC of the reaction mixture
showed a first degradation between 61 and 218 °C and a second one between 297 and 391 °C,
both of them with low energy (Figure 3). Hydrolysis was also evaluated in a RC1 prior to the
pilot-plant batch (Figure 4). Here again the heat flow was high (maximum of 96 W/kg of reaction
ACS Paragon Plus Environment

Page 11 of 46
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
mass that corresponded to a Tad of 49 °C) associated with a low accumulation (4% observed
after addition of 17 wt % of Dibal-H). It was assumed that the heterogeneous nature of the
reaction mixture was responsible for the shape of the heat flow curve. Extended time was
therefore recommended (2 hours instead of 1) when the reaction mixture was added to aqueous
NaOH. Gas evolution was linear and could be controlled by dosing.
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 2. Heat flow profile of Dibal-H addition
Figure 3. DSC profile of the reaction mixture after completion of Dibal-H addition
ACS Paragon Plus Environment

Organic Process Research & Development
Page 12 of 46
1
2
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
Figure 4. Heat flow profile of hydrolysis
Since treatment with Dibal-H in toluene resulted in a complete conversion, we envisioned to
telescope oxime formation, hydride reduction and salt formation in that solvent (see scheme 4).
Gratifyingly condensation of ketone 16 with MeONH2.HCl in toluene (5 mL/g of 16) gave a
better purity than what was obtained in MeTHF (89 area % by HPLC instead of 85). At this
stage, aqueous washes were necessary to insure a satisfactory conversion later on with Dibal-H.
ACS Paragon Plus Environment

Page 13 of 46
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Salt formation was also carried out in toluene with addition of an alcohol. In preliminary testing,
MeOH provided a better yield than EtOH and was selected for further improvement. Using a
Crystal 16 apparatus, a 65/35 (v/v) mixture of toluene/MeOH at 40 g/L (or 25 mL/g of 19) was
identified and a range between 50 and 60 °C was determined for seeding. To avoid any
degradation, 50 °C was preferred. When we applied this sequence on 8 kg scale of 16 having a
moderate purity (62 area % by HPLC), we were pleased to isolate 19 with 47% overall yield
without any material left in the mother liquors. This process was also very efficient to remove
impurities and provided 19 with a high purity (99.5 area %). The free base was obtained after
aqueous NaOH treatment and enantiomers were separated by chiral chromatography to furnish
enantiopure 20.
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 5. Determination of the metastable zone of 19 with Crystal 16.
Scheme 4. New telescoped process to manufacture 19
ACS Paragon Plus Environment

Organic Process Research & Development
Page 14 of 46
1
2
3
4
5
6
7
8
OMe
O
NH2
NH2
N
CO2H
MeONH2.HCl
Dibal-H
Fumaric acid
MeOH
2
Toluene, 20 °C
S
S
S
S
Ph
NMe2
CO2H
Ph
NMe2
Ph
NMe2
Ph
NMe2
17
18
16
47%
over 3 steps
19
One crop
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
Since we had concerns with potential safety issues associated with fluoro-nitrophenyl compound
21, conditions of aromatic nucleophilic substitution were revisited using chloride derivative 22.
In terms of safety, DSC studies showed that compound 22 was a better partner than 21 because
decomposition started at 225 °C with a maximum reached at 306 °C (see Supporting
Information). Aromatic nucleophilic substitution with amine 20 was consequently evaluated in 50
mL glass reactor with a reaction temperature limited at 90 °C to ensure process safety. This was
performed using diisopropylethylamine (DIPEA) in various solvents and the results are
summarized in Table 1. Formation of the desired product was observed in all solvents and DMSO
proved to be by far the best option (entry 6 and 7). The reaction became sluggish when conducted
at a lower temperature (entry 7). Thus 90 °C was found to be an acceptable compromise between
kinetics and safety (entry 6). A sequential acid-base workup applied on crude 3 allowed removal
of both excess of DIPEA and reactant 20. Compound 3 was initially isolated by filtration after a
solvent switch from EtOAc to MCH (methylcyclohexane). This process produced large hard balls
that were difficult to drain and could cause damage to our vessels upon scale-up. Replacing MCH
with i-PrOH was beneficial in terms of processability. A reslurry with MTBE was incorporated to
remove traces of remaining 20 and to facilitate drying of the cake. Upon scale-up (3.6 kg of 20),
3 was isolated with 74% yield and with a satisfactory purity of >99%.
Table 1. Screening of solvents on gram scale to synthesize 3a
ACS Paragon Plus Environment

Page 15 of 46
Organic Process Research & Development
1
2
3
O
O
4
5
6
7
8
9
NO2
NH
S
H2 N
NH2
O
O
Solvent
DIPEA
NO2
Cl
S
H2 N
+
S
Ph
NMe2
20
22
3
S
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
NMe2
Entry
Solvent
20 (%)b 22 (%)b 3 (%)b
1
2
AcOEt
i-PrOH
Toluene
DMF
11
14
11.6
44
3.5
0
25.8
35
57.2
48.2
51.4
32
3
23.1
5
4
5
NMP
19.2
3.6
8
57.6
81.7
73
6c
7d
DMSO
DMSO
4
b
aConditions: 20 (1.1 equiv), 22 (1.0 equiv), DIEA (4.5 equiv), 24 hours, 90°C. Mesured by
HPLC analysis of a sample of reaction mixture. cIsolated yield of 56%. d40 h, 60 °C.
End-game chemistry
Acylsulfonamides are an important class of compounds in medicinal chemistry as they serve as
carboxylic acid bioisosters. They can be prepared by condensing carboxylic acids with
sulfonamides (RSONH2) using dehydrating agents such as EDC (N-(3-dimethylaminopropyl)-N′-
ethylcarbodiimide) for example. Addition of an excess of dimethylaminopyridine (DMAP) is
usually required due to the low nucleophilicity of the sulfonamide. At the end of the reaction, the
acylsulfonamide is obtained after an acidic workup to protonate the acylsulfonamide-DMAP
salt.14 Originally, coupling of partners 23 and 3 proceeded smoothly at 20 °C in DMF15 with 2.5
equiv of EDC and 3 equiv of DMAP (see scheme of Table 2). Isolation proved to be a major
issue since acylsulfonamide 24 was sparingly soluble in only few solvents (DMF, MeOH, EtOH,
ACS Paragon Plus Environment

Organic Process Research & Development
Page 16 of 46
1
2
3
4
5
6
7
8
9
CH2Cl2, THF) and attempts to crystallize it furnished either gels or gums. Isolation as
hydrochloride salt was evaluated with aqueous HCl and provided the final product free of any
residual solvent (at this stage, lyophillization was required). An alternative encouraging
procedure was then developed and consisted in treating the reaction mixture with EtOAc and 1 N
NaOH (Scheme 5). It allowed both elimination of residual DMF in the aqueous layer and
formation of 1 (sodium salt of 24) that was soluble in the organic phase. Solvent switch to i-
PrOH furnished the desired API with 82% yield and 98.5% purity. In order to comply with
cGMP requirements, a polishing filtration had to be incorporated in the process. Gratifyingly
solubilization of 1 in 10 mL/g of refluxing acetone proceeded well and filtration at 40 °C was
beneficial to remove traces of 2. Solvent switch to i-PrOH and addition of MCH before filtration
furnished the desired API with 92% yield and 99.3 area % purity on 5 kg 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
Scheme 5. Multikilogram-scale synthesis of 1
O
O
O
O
O
O
S
NO2
NH
-
S
NO2
NH
OH
H2 N
N
Na+
1
N
N
+
1) EDCI, DMAP, DMF
2) EtOAc, 1N NaOH
3) i-PrOH
1) Acetone
Cl
N
Cl
N
HCl
2) Polishing filtration
3) i-PrOH
99.3% purity
S
Me
S
Me
N
N
82%
92%
1
23
3
Me
Me
cGMP conditions
on 5.47 kg
98.5% purity
On 3,67 kg
Although this step was successful to manufacture the first batches of 1, the following issues
needed to be addressed to increase the practicality for scale-up: use of EDC (sticky oil,
expensive, concerns about its chemical stability on storage), DMF as solvent in a late stage of the
synthesis, tedious isolation process. Development of a second-generation process was undertaken
and focused on identifying alternative coupling conditions. N-(3-dimethylaminopropyl)-N′-
ethylcarbodiimide hydrochloride (EDC-HCl), thionyl chloride, CDI, and propanephosphonic acid
ACS Paragon Plus Environment

Page 17 of 46
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
anhydride (T3P) were selected and reacted with 23 (HCl salt). After 18 hours at 50 °C; the
conversion was almost complete with EDC-HCl (entry 1, Table 2) and the product was isolated
as follows: quench with aqueous NaOH, organic phase extracted with EtOAc, solvent switch to i-
PrOH, reslurry with i-PrOH. However, the moderate purity coupled with the knowledge that
EDC-HCl was both expensive and a suspected sensitizer encouraged us to switch to other
coupling systems.16 Scouting experiments showed that no product was formed with SOCl2 (entry
2) and CDI (entry 3 and 4). With an excess of both Et3N and T3P, 14% of desired product was
observed at room temperature (entry 5) but reflux was deleterious since compound 3 degraded
(entry 6). We then decided to proceed stepwise by neutralizing the hydrochloride before adding
the coupling agent. Treatment with 2.1 equiv of Et3N in EtOAc provided the carboxylate
derivative in solution after filtration of the hydrochloride salts. This solution was then subjected
to several activating agents (see Table 3).
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
Table 2. Survey of coupling conditions applied on 23
O
O
O
O
O
O
S
NO2
NH
NO2
NH
S
OH
H2
N
N
H
N
N
Conditions
Cl
N
Cl
N
+
HCl
23
S
Me
S
Me
N
N
Me
Me
24
3
23
24
Entry
1b
Conditions
3 (0.9 equiv), EDC-HCl (1.3 equiv),
3 (%)a
(%)a
(%)a
5.1
0.6
81.3
DMAP (2.9 equiv), DMF, 18 h at 50 °C
2
3
SOCl2 (1.2 equiv), EtOAc, 3 (1.1 equiv),
16.6
13.5
74.0
25.1
-
-
6 h at reflux
CDI (1.2 equiv), EtOAc, 3 (1 equiv),
DBU (1 equiv), 6 h at 50 °C
ACS Paragon Plus Environment

Organic Process Research & Development
Page 18 of 46
1
2
3
4
5
6
7
8
9
4
5c
6
CDI (1.2 equiv), NMP, DBU (1 equiv), 3
12.8
29.4
0.5
-
13.9
-
(1 equiv), 3 h at 50 °C
3 (1.1 equiv), EtOAc, Et3N (5 equiv),
23.4
T3P (5 equiv), 12 h at 20 °C
3 (1.1 equiv), EtOAc, Et3N (5 equiv),
53.9
1.3
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
T3P (5 equiv), 3 h at reflux
aMesured by HPLC analysis of a sample of reaction mixture. Quench with aqueous NaOH,
organic phase extracted with EtOAc, solvent switch to i-PrOH, reslurred in i-PrOH. 65% yield,
96.2 area %purity. cNo isolation.
b
Table 3. Step-wise sequence to couple 23 with 1.1 equiv of 3
O
O
O
O
O
O
S
NO2
NH
NO2
NH
S
OH
H2
N
N
H
1) Et3N, AcOEt
2) Filtartion
N
N
Cl
N
Cl
N
+
3) Conditions
HCl
23
S
Me
S
Me
N
N
Me
Me
24
3
Entry
1
Conditions
23 (%)a 3 (%)a 24 (%)a Yield (%)c
SOCl2 (2.5 equiv), EtOAc, 6 h at reflux
-
-
-
-
-
2b
CDI (1.5 equiv), THF, DBU (1 equiv), 3 h
at reflux then 3 (1 equiv) and 20 h at reflux
12.2
25.4
10.8
3
4b
5
EtOAc, Et3N (2 equiv), T3P (5 equiv),
23 h at reflux
3.2
5.7
0.9
0.5
86.6
84.4
12.0
-
55
-
EtOAc, Et3N (2 equiv), T3P (5 equiv),
23 h at reflux
EtOAc, Et3N (2 equiv), T3P (2 equiv),
20 h at reflux
34.4
12.2
6
7
EtOAc, T3P (5 equiv), 24 h at reflux
EtOAc, T3P (5 equiv), 30 h at reflux
2.0
1.6
0.6
0.4
92.8
94.0
60
63
aMesured by HPLC analysis of a sample of reaction mixture. bWith 1.0 equiv of 3. cIsolated.
ACS Paragon Plus Environment

Page 19 of 46
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Except with SOCl2 and CDI (entry 1 and 2), the two-step procedure provided encouraging results
with T3P. The reaction proceeded well with an excess of both base and T3P at reflux (entry 3).
Decreasing the amount of 3 resulted in a lower conversion (entry 4). The amount of T3P was also
critical (entry 5). Assuming that the dimethylamine moiety of 3 could act as a base, we showed
that Et3N removal was not detrimental (entry 6 and 7). Since the reaction was sluggish, an
extended time was beneficial in terms of conversion (entry 7). In that case, washing the reaction
mixture twice with 6 N NaOH provided 1 with a disappointing quality of 96.9 area % after
solvent switch to i-PrOH and subsequent crystallization. Since extra reslurries didn’t improve the
quality, the use of T3P was discontinued and we switched to N-(1-methanesulfonyl)benzotriazole
26 as coupling agent.
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
Synthesis of coupling agent 26
The preparation of N-acylbenzotriazoles has been previously reported by the Katritsky group.17
Direct treatment of benzotriazole 25 with methanesulfonyl chloride in the presence of pyridine
afforded desired N-(1-methanesulfonyl)benzotriazole 26 with 89% yield after concentration to
dryness and recrystallization from benzene. Since large quantities of this coupling agent was
anticipated, we thought there was room for improving this procedure and a quick survey of
alternative conditions was undertaken (Table 4). Original conditions were reproduced and
afforded the desired product with 48% yield after addition of water, concentration of the organic
phase and recrystallization in toluene (entry 1). Replacing pyridine with Et3N improved the yield
(entry 2) but formation of benzotriazole was observed when the reaction mixture was heated to
reflux for several hours. In MeTHF, Et3N gave the best results in terms of kinetics (1 h compared
to 6 h with 2,6-collidine and 22 h with pyridine) and conversion (entry 3, 4 and 5). However
keeping 26 in MeTHF in order to telescope the reaction with 23 was precluded since some
ACS Paragon Plus Environment

Organic Process Research & Development
Page 20 of 46
1
2
3
4
5
6
7
8
9
degradation was observed on storage (loss of 10% purity after 72 hours at 4 °C).18 Issues were
overcome when 26 was prepared in THF using Et3N as a base: addition of a solution of
methanesulfonyl chloride in THF to the reaction mixture was required to control an exotherm.
The reaction was complete after 1 hour and crystallization occurred after addition of water (entry
7). On 500 g scale, we were pleased to isolate 26 after filtration with 90% yield and >99 area %
purity as a white powder stable on storage. This isolation process was by far more efficient than
filtering Et3N hydrochloride salts from the reaction mixture before adding water and separation of
the phases (entry 6).
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
Table 4. Alternative conditions to prepare 26
N
N
N
Conditions
N
N
N
H
SO2Me
25
26
25
26
Yield
(%)h
Purity
(%)
Entrya
1b
Solvent, base (equiv), MesCl (equiv)
Toluene, pyridine (1.6), (1.2)
Toluene, Et3N (1.6), (1.2)
MeTHF, pyridine (1.5), (1.3)
MeTHF, 2,6-collidine (1.6), (1.4)
MeTHF, Et3N (1.5), (1.3)
THF, Et3N (1.3), (1.2)
(%)g
(%)g
5.8
1.4
2.7
0.6
0.1
0.5
0.6
74.0
89.9
98.2
86.0
99.4
99.2
99.2
48
70
-
99.1
99.1
86.5
90.9
98.8
96.6
99.7
2b
3c
4d
-
5c
-
6e
57
87
7f
THF, Et3N (1.5), (1.3)
aAll reactions were run in round bottom flasks at 20 °C. bAddition of water, concentration of the
c
organic phase and recrystallization in toluene. Filtration of the reaction mixture, 26 obtained in
d
e
solution. Addition of water then phase separation. 26 obtained in solution. Filtration of the
f
reaction mixture, concentration to dryness, recrystallization in toluene. Addition of water,
filtration of the slurry. gMesured by HPLC analysis of a sample of reaction mixture. hIsolated.
ACS Paragon Plus Environment

Page 21 of 46
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
N-acetyl-benzotriazol 27 was then synthesized by the reaction of acid 23 with BtSO2Me in the
presence of an excess of Et3N (3.5 equiv). THF and MeTHF were selected as solvents and
provided the active species in solution after filtration of the hydrochloride salts of Et3N with a
similar chemical purity (>90 area %; remaining 2 was <0.5 area %). Those solutions were then
reacted with the anion of sulfonamide 3 prepared by reacting 3 with various bases as described in
Table 5.19
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
Table 5. Survey of coupling conditions between 27 and 3 in the presence of a base
O
O
O
O
O
O
-
S
NO2
NH
N
S
NO2
NH
H2
N
N
N
N
Na+
Base
N
N
+
Cl
N
Cl
N
Solvent
S
Me
S
Me
N
1
N
27
Me
Me
3
Yield
(%)c
Purity
(%)d
Entry
Conditionsa
NaH (2.5 equiv), THF, 1 h
1 (%)b 3 (%)b
1
2
69.4
58.2
63.0
73.2
75.7
8.5
5.9
5.2
1.8
1.0
70
59
65
78
81
98.5
98.9
98.1
98.6
98.8
t-BuONa (2.1 equiv), THF, 3 h
NaOH (1.9 equiv), THF, 22 h
NaOH (1.9 equiv), MeTHF, 2 h
NaOH (1.5 equiv), MeTHF, 2 h
3e
4e
5f
b
aAll reactions run at reflux in round bottom flasks. Mesured by HPLC analysis of a sample of
c
d
e
reaction mixture. Isolated. After three reslurries in i-PrOH:MCH 87.5:12.5. NaOH prills.
fNaOH flakes.
Primary screening showed that the nature of the cation of the base was crucial to provide the
desired product. For example, coupling reaction mediated with n-BuLi furnished the lithium salt
of 1 and switching to the sodium analogue proved to be impraticable. This prompted us to choose
bases constituted with sodium such as NaH, t-BuONa or NaOH. When 2.5 equiv of NaH 60% in
ACS Paragon Plus Environment

Organic Process Research & Development
Page 22 of 46
1
2
3
4
5
6
7
8
9
oil was used, deprotonation of sulfonamide 3 proceeded almost instantly at 20 °C (dark red
color).20 After 30 minutes, the mixture was heated to 40 °C and a solution of 27 was added
slowly before heating at reflux for 1 hour. Quenching the reaction with water provided an organic
phase containing 69.4 area % of 1. The purity was improved to 94.4 % when the crude was
reslurred in a mixture of i-PrOH:MCH 87.5:12.5 and even 98.9% after two more re-slurries (see
entry 1). Despite the use of MCH to eliminate oil residues, we decided to evaluate other bases
because we had concerns about our ability to detect and quantify those contaminants in the API.
The same procedure was applied with t-BuONa and furnished the desired product with 59% yield
and a satisfactory purity (entry 2).21 However we observed a batch-to-batch variability when we
repeated this reaction resulting in the formation of 10 to 22% of carboxylic acid 2. This lack of
reproducibility prompted us to switch to NaOH.
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
We noticed that the nature of both NaOH and solvent had a great impact on the outcome of the
reaction. The rate of conversion was greatly improved using prills of NaOH when THF was
replaced with MeTHF (entry 3 and 4).22 Switching to NaOH flakes was also beneficial (entry 5)
in terms of both equivalents (1.5 instead of 1.9) and conversion assuming it offered a better
surface area than prills. In that case, 5% of carboxylic acid 2 was observed in the reaction
mixture. Quenching with an aqueous solution of NaOH eliminated most of benzotriazole released
during the reaction. Solvent switch to i-PrOH under vacuum at 50 °C was preferred rather than
atmospheric pressure at 80 °C since less degradation was observed. Filtration of the slurry was
facilitated by addition of a small amount of MCH and provided 1 with 98.5 area %. At this stage,
compound 2 (initially 0.3 to 0.5%) was purged after solubilization of 1 in 10 mL/g of refluxing
acetone followed by filtration at 40 °C.23 Solvent switch to i-PrOH and addition of MCH to the
reaction mixture furnished the desired API with 92% yield and 98.9 area % purity (Scheme 6).
ACS Paragon Plus Environment

Page 23 of 46
Organic Process Research & Development
1
2
3
4
Scheme 6. Coupling with BtSO2Me and isolation process
5
6
7
8
O
O
O
S
NO2
NH
H2 N
OH
9
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
Cl
N
HCl
23
S
Me
N
Me
3
NaOH,
Et3N,
MeTHF, reflux
MeTHF, reflux
O
O
O
S
NO2
NH
N
NaHN
N
N
N
Cl
N
+
S
Me
N
Me
27 in MeTHF
28 in MeTHF
O
O
O
-
S
NO2
N
Na+
N
NH
Cl
N
S
Me
N
Me
1 in MeTHF
1) Solvent switch to i-PrOH, 82%
2) Acetone
3) Polish filtration
4) Solvent switch to i-PrOH, 92%
1
ACS Paragon Plus Environment

Organic Process Research & Development
Page 24 of 46
1
2
3
4
5
6
7
8
9
A closer look on the purity profile of the final product provided some information about the
nature of major impurities (Figure 6). One advantage implementing the benzotriazole chemistry
relied on the fact that impurity 29, arising from condensation of EDCI on 1 and difficult to purge,
no longer existed. Besides, we observed in early isolation studies the formation of 30 when
CH2Cl2 was used as solvent in the presence of NaOH. This prompted us to switch to i-PrOH to
avoid this side reaction. Compounds 31, 32 and 33 were considered as daughter impurities arising
from two side reactions of the Suzuki coupling that occurred during the synthesis of the biaryl
motif: de-chlorination mediated with Pd to give 31 (see Multikilogram synthesis of a potent Bcl-2
selective antagonist. Part I.) and multiple Suzuki couplings to give 32 and 33. Finally we
suspected that Me2NH contained traces of MeNH2 and provided the mono N-methylated
analogues of 3 to explain the formation of 33 and 34. Fortunately our isolation process managed
to purge those impurities to levels <0.10 %.
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
Figure 6. Identified process-related impurities.
ACS Paragon Plus Environment

Page 25 of 46
Organic Process Research & Development
1
2
3
4
5
O
O
O
O
O O
S
N
NO2
NH
S
NO2
NH
N
N
H
N
H N
N
6
Cl
N
Cl
N
Cl
7
8
S
30
S
N
Ph
NMe2
Ph
N+
29
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
31 R = Ph
32 R =
O
O
O
S
NO2
NH
N
H
Cl
N
N
Cl
33 R =
R
S
Ph
NMe2
O
O
O
O
O O
NO2
NH
S
NO2
NH
S
N
H
N
H
N
N
Cl
N
Cl
N
S
Me
N
35
S
Me
34
Ph
O
N+
Ph
N
N
O–
O
O
S
N
H N
O
O
N
N
Cl
Cl
Epilogue
Access to chiral diamine 20 has been the one of the major challenge of this project. As reported
above, efforts have been undertaken to develop an original synthesis to manufacture the first
batches of 1. They finally culminated with an asymmetric approach involving a diastereoselective
aza-Reformatsky reaction coupled with a chemoselective amination under Mitsunobu conditions
(Scheme 7).24 Our strategy relied on the preparation of chiral sulfinimine using Ellman’s N-tert-
butanesulfinamide 38 that was reported to be a powerful auxiliary for the synthesis of chiral
amines.25 Bromide 36 was selected as starting material and was easily converted to sulfinimine
ACS Paragon Plus Environment

Organic Process Research & Development
Page 26 of 46
1
2
3
4
5
6
7
8
9
39 telescoping nucleophilic displacement with sodium thiophenolate, acetal deprotection into the
corresponding aldehyde and condensation with N-tert-butanesulfinamide 38. The aza-
Reformatsky reaction was then the main hurdle we encountered. Formation of the Reformatsky
reagent from activated zinc metal and -bromoester is known to be a challenging reaction
because it is highly exothermic after an unpredictable time of induction. To overcome this issue,
a zinc-activation procedure using diisobutylaluminium hydride (Dibal-H) has been reported
avoiding the need of reflux to activate zinc and limiting the induction period.26 For safety
reasons, the entire process was evaluated in RC1 on our substrates and the calorimetric profiles
are shown below. Scouting experiments demonstrated that an excess of zinc (3 equiv) and 1.7
equiv of methyl bromoacetate were needed to obtain a satisfactory conversion. The Reformatsky
reagent was prepared by suspending zinc in THF at 30 °C with vigorous stirring followed by
subsequent addition of 0.1 equiv of bromide and 0.1 equiv of Dibal-H. A moderate heat output of
27 kJ/kg of reaction mass was monitored corresponding to an adiabatic rise of 15 °C (see Figure
7). When 1.6 equiv of methyl bromoacetate were added in 60 minutes at 45 °C, some heat
accumulation was observed (15% estimated) followed by a sharp heat evolution. With a heat
output of 270 kJ/kg and a Tad of 130 °C, we decided to extend the addition rate to 2 hours and to
proceed step wise: add 15% of the total charge of methyl bromoacetate in 15 minutes (addition
stopped if the reaction didn’t start) and then remaining 85% in 105 minutes, controlling the
addition rate if necessary. DSC showed two events with onsets at respectively 74 and 100 °C
with low energy (Figure 8). A third event with an onset at 196 °C was also observed but was
spread on a wide window of temperature. Addition of the sulfinimine proceeded at 20 °C (50% of
the total charge added in 30 minutes followed by 50% in 15 minutes) with 4% of heat
accumulation (Figure 9). This step was easily manageable in terms of safety (heat output of 30
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

Page 27 of 46
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
kJ/kg associated with a Tad of 17 °C). DSC showed two events with onsets at 75 and 200 °C
associated with moderate to low energy. A quench procedure using a 10% aqueous solution of
citric acid was also evaluated and was finally replaced with a three-step procedure: addition of
brine followed by aqueous citric acid and by brine.27 To limit the heat release of the first addition
of brine, it was recommended to extend the addition rate (10% of the total charge added in 10
minutes followed by 90% in 10 minutes, see Figure 10). Addition of citric acid over 20 minutes
proceeded well with almost no accumulation and gave an acceptable heat evolution rate (Figure
11). It is worth mentioning that extra stirring was needed because the reaction mixture became
thicker at 30% of total charge of citric acid. At 60%, the reaction mixture became more fluid and
stirring more efficient. The second addition of brine required no specific attention. This
procedure allowed us to isolate 40 after column chromatography on silica gel with a
diastereoisomeric ratio of 94:6. Telescoping the reduction of the ester mediated with Red-Al and
the Mitsunobu reaction was possible using toluene to provide 42 with >99 area % purity after
purification on silica gel. Chiral diamine was obtained as difumarate 43 with >99% enantiomeric
purity after telescoping removal of the chiral auxiliary and salt formation.
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 7. Second-generation synthesis of enantiopure 20 isolated as difumarate salt 43
ACS Paragon Plus Environment

Organic Process Research & Development
Page 28 of 46
1
2
3
4
5
6
7
8
O
S
1) PhSNa, EtOH,
50 °C, 1 h
O
S
1) H2SO4, reflux, 1 h
OEt
OEt
OEt
OEt
N
t-Bu
Br
PhS
PhS
H2 N
t-Bu
2) Solvent switch
H
2) PPTS cat, 38,
reflux, 4 h
38
39
37
36
68% over 3 steps
9
O
O
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
S
S
Zn, DIBAL-H cat
methyl bromoacetate
H N
t-Bu
CO2Me
H N
t-Bu
OH
Red-Al, toluene
PhS
PhS
0 °C, 2 h
THF, 20 °C, 30 min
80% crude yield
40
41
80% crude yield
O
NH2
S
PhS
H N
t-Bu
NH2
PPh3, DIAD, Me2NH
Toluene, 20 °C, 1 h
1) MeOH, anh. HCl
2) Solvent switch
Fumaric acid
MeOH, reflux
NMe2
CO2H
PhS
PhS
NMe2
NMe2
HO2C
2.
42
64% after chromatography
20
78% over 2 steps
43
HPLC purity >99%
99.5% ee
Figure 7. Heat flow profile for addition of methyl bromoacetate
Figure 8. DSC thermogram of the organozinc mixture
ACS Paragon Plus Environment

Page 29 of 46
Organic Process Research & Development
1
2
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
Figure 9. Heat flow for imine addition
Figure 10. Heat flow profile for addition of brine
ACS Paragon Plus Environment