Scale-Up Synthesis of IID572: A New β-Lactamase Inhibitor

Article abstract of DOI:10.1021/acs.oprd.0c00069

The new potentially best-in-class β-lactamase inhibitor IID572 was discovered by a late-stage functionalization approach. An alternative synthesis was developed to satisfy the short-term material need for toxicological studies in animals. The new synthetic strategy was built on two key features, an intramolecular azomethine ylide [3 + 2] cycloaddition that allowed the efficient formation of molecular complexity from readily available starting materials and an enzymatic resolution that resulted in high optical purity of a key intermediate.

Full text of DOI:10.1021/acs.oprd.0c00069

Subscriber access provided by Uppsala universitetsbibliotek
Full Paper
Scale up synthesis of IID572: A new #-lactamase inhibitor
Markus Furegati, Sandro Nocito, Folkert Reck, Anthony
Casarez, Robert Simmons, Heiner Schuetz, and Guido Koch
Org. Process Res. Dev., Just Accepted Manuscript • DOI: 10.1021/acs.oprd.0c00069 • Publication Date (Web): 14 May 2020
Downloaded from pubs.acs.org on May 14, 2020
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 47
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Scale up synthesis of IID572: A new β-lactamase
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
inhibitor
Markus Furegati†*, Sandro Nocito†, Folkert Reck‡, Anthony Casarez‡, Robert Simmons‡,
Heiner Schuetz†, Guido Koch†
†
Synthesis and Technologies Group, Novartis Institutes for BioMedical Research,
Klybeckstrasse 141, Basel 4057, Switzerland
‡
Novartis Institutes for BioMedical Research, 5300 Chiron Way, Emeryville, California
94608, United States
For Table of Contents Only:
HN
H
O
H
N
(S)
8 steps
0.9% yield
19 steps
O
OH
O
HO
1.7% yield
(S)
(R)
N
BnHN
discovery
route
scale up
route
OSO3Na
N
N
OBn
limited access
O
O
IID572
readily available
ACS Paragon Plus Environment
1

Organic Process Research & Development
Page 2 of 47
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
ABSTRACT
The new potentially best-in-class β-lactamase inhibitor IID572 was discovered by a late
stage functionalization approach. An alternative synthesis was developed to satisfy the
short-term material need for toxicological studies in animals. The new synthetic strategy
was built on two key features, an intramolecular azomethine ylide [3+2] cycloaddition that
allowed the efficient formation of molecular complexity from readily available starting
materials and an enzymatic resolution that resulted in high optical purity of a key
intermediate.
KEYWORDS
β-lactamase inhibitors, diazabicyclooctane, DBOs, scale up, enzymatic resolution, [3+2]
cycloaddition
ACS Paragon Plus Environment
2

Page 3 of 47
Organic Process Research & Development
1
2
3
4
5
INTRODUCTION
6
7
8
9
Our research program on new β-lactamase inhibitors of the diazabicyclooctane (DBO)
class revealed IID572 (1) to be a potent, potentially best in class, broad spectrum β-
lactamase inhibitor that lacks antibacterial activity. In discovery chemistry, a late stage
functionalization (LSF) approach was applied to access the unique structural framework
of IID572.1 The starting point of the research synthesis was intermediate 4 of the
Avibactam synthesis (scheme 1), which can be obtained in 48% yield over 6 steps.2, 3 The
key step in our LSF route was a photochemical Giese-type radical addition to the Michael
acceptor 6 that resulted in >80% yield a mixture of 7 and three other diastereoisomers
(not shown).4 After the isolation of the desired isomer 7 and deprotection with TFA the
primary amine spontaneously cyclized under basic conditions to provide compound 8.
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 1. Synthesis of compound 1 by the addition of a photochemically generated
radical.a Copyright 2019 American Chemical Society.
ACS Paragon Plus Environment
3

Organic Process Research & Development
Page 4 of 47
1
2
3
4
5
6
O
O
O
H
H
H
a-e
f
g
h, i
(S)
OEt
EtO
HO
MeO
(S)
N
(S)
N
(S)
N
O
N
(R)
(R)
(R)
48%
87%
24%
O
Boc
N
N
N
OBn
OBn
OBn
O
O
O
7
2
3
4
5
starting material for
research activities
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
H
N
Boc
HN
HN
H
O
O
O
H
(S)
(S)
(S)
O
O
MeO
j
MeO
k
l-n
(S)
N
(S)
(S)
N
H
(R)
(R)
(R)
(R)
N
N
10%
63%
65%
N
N
N
N
OBn
OBn
OBn
OSO3Na
O
O
O
6
7
8
1
a Reagents and conditions: (a) Me3SOI, KOtBu, THF, DMSO (b) BnONH2·HCl, EtOAc (c)
MsOH, EtOAc then aq. KHCO3 (d) NaBH4, propanoic acid, H2SO4, EtOAc then oxalic acid
(e) triphosgene, picoline, NaHCO3, DCM/H2O (f) Novozyme 435 (yield given as 65% for
the in situ carboxamide formation) (g) MeOH, EDC, HOBt, DIPEA, DCM (h) LDA, PhSeCl,
THF (i) H2O2, AcOH, THF, H2O (j) Boc-Gly-OH, Ir[(dF(CF3)ppy)2(dtbbpy)]PF6, K2HPO4,
8W UVA fluorescence tube (k) TFA, then NEt3 (l) H2, Pd/C, MeOH (m) SO3·py, Bu4NHSO4
(n) sodium 2-ethyl hexanoate.
This sequence, although relatively short and therefore ideal for LSF, suffers from three
major drawbacks with respect to scale up. Firstly, PhSeCl, a toxic and expensive reagent
was required to introduce the radical acceptor olefin and yielded 6 in only 24%. Secondly,
the formation of a diastereoisomeric mixture after the radical addition resulted in formation
ACS Paragon Plus Environment
4

Page 5 of 47
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
of only 10% of the desired isomer 7. The ratio of the four isomers was not influenced by
varying the reaction conditions e.g. solvent, catalyst or temperature. Thirdly, even if this
sequence would have worked starting from the ethyl ester 3 instead of the methyl ester 5
the overall yield was calculated to be 0.48% over 12 steps. The estimated time to produce
the required quantitity of the key intermediate 8 led us to investigate alternative routes
with the goal to synthesize 5 g 1 as quickly as possible to enable preclinical investigation
of the new development candidate.
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
In an exploratory evaluation project we investigated a) An Aza-Diels-Alder route (Scheme
2a):5 the cycloaddition, however, was a challenging key step resulting in a complex
mixture of products including pyridines. b) the asparagine cyclization (Scheme 2b):6
starting from known amino acid derived 9.7 The N-alkylation did not work well and the
subsequent cyclization failed. c) Pyridine hydrogenation from known 11 (Scheme 2c):8
Although this was a promising approach the synthesis of 11 starting from expensive
methyl 3-methylpyridine-2-carboxylate (10) was not reproduced in a satisfactory yield; in
ACS Paragon Plus Environment
5

Organic Process Research & Development
Page 6 of 47
1
2
3
4
5
6
7
8
9
addition, the ammonolysis also did not work well. A similar approach was later
successfully applied in an alternative Avibactam synthesis.9 All three initial approaches
were abandoned.
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. Attempted routes
a)
O
1) aza-DA, ACN, 100°C
2) aq. HCl
3) aq. NaHCO3
HN
N
H
N
O
N
O
O
O
N
N
O
b)
1) TFA
2) (R)-glycidyl-
TBS ether,
Mg(OTf)2
O
CN
O
CN
O
CN
steps
H
OTBS
OH
BnO
BnO
BnO
(S)
(S)
(S)
(S)
(R)
HN
HN
NHBoc
OH
9
c)
1) ammonolysis
2) ether cleavage
3) hydrogenation
rac
OH
HN
H
3 steps
MeO
O
Br
HN
ca 4%
N
CO2Me
10
N
CO2Me
11
12
Instead, we decided to focus on a route that started from simple and easily available
starting materials and allowed us to quickly build up complexity via intramolecular
azomethine ylide [3+2] cycloaddition (scheme 3). The three step synthesis of 15 was
described in the literature.10 We found that Boc protection of Bn-Gly-OH11 worked in
ACS Paragon Plus Environment
6

Page 7 of 47
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
quantitative yield in water/THF in the presence of NaHCO3. T3P proved in our hands to
be the best coupling reagent for the amide formation with commercial allylbenzylamine
and deprotection with TFA gave over 3 steps an overall yield of 97% (31 g). When we
outsourced this sequence the external partner delivered 1.5 kg 15 (57% yield over 3 steps
using different reagents). A similar cyclization of N-allyl-2-(methylamino)-N-
phenylacetamide and butyl glyoxylate was described with 30% yield.12 We were pleased
to find that slow addition of the ethyl glyoxylate significantly improved the yield of
cycloaddition product rac-16 up to 76%. The major by-product was a diastereoisomer that
was difficult to remove by chromatography.13 Reduction of the ethyl ester rac-16 with
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
14
LiBH4 to the alcohol rac-17 gave the starting material for the second key step, the ring
16
expansion reaction.15,
The in situ formed trifluoroacetate was displaced by the
benzylamine in an intramolecular fashion to form an aziridinium intermediate that was re-
opened at the methylene carbon (α to the ammonium) by trifluoroacetate to form the exo
product (5-membered ring resembling the starting material) or at the higher substituted
carbon to access the 6-membered ring target compound. The reaction mixture was
quenched with aq. NaOH (saponification of the TFA ester) and rac-18 was isolated
ACS Paragon Plus Environment
7

Organic Process Research & Development
Page 8 of 47
1
2
3
4
5
6
7
8
9
after chromatography over silica gel.17A long reaction time was crucial for a high
yield, attempts to shorten the reaction time were not successful. Allowing the reaction to
perform at higher temperature under microwave conditions resulted in higher amounts of
the undesired 5-membered ring isomer and ultimately a low conversion. A single
experiment applying the non-equilibrium, irreversible conditions (MsCl/Et3N followed by
addition of AgOAc) gave no desired product.15, 16 In order to avoid chiral chromatography
we screened 21 lipases for stereospecific enzymatic esterification with vinyl acetate of
alcohol rac-18 and found that QLM lipase selectively provided both, the acetate and the
alcohol in high enantiomeric purity (both >95% ee, E factor of 80). We completed the
synthesis with one enantiomer before starting the main batch and confirmed that acetate
19 had the desired 3R configuration; the same as determined for the target product
originally prepared by the LSF approach. While waiting for the large batch of the enzyme,
we successfully reduced the amount by factor 10. The reaction proceeded smoothly and
after six days the crude product was isolated and purified by the first chromatography in
this sequence. Having the desired isomer at the acetate stage, an additional hydrolysis
step was required to prepare key intermediate (–)-18 (scheme 4).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment
8

Page 9 of 47
Organic Process Research & Development
1
2
3
4
5
Scheme 3. Azomethine ylide route part 1: Core formation and enzymatic resolution
6
7
8
9
allylbenzylamine,
HATU, DIPEA, DMF
rt, 16 h
Boc2O, NaHCO3
THF/H2O 1:1
rt, 16 h
Bn
O
OH
TFA/DCM 5:1
r.t., 16 h
O
N
O
OH
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
BnN
BnN
yield: 92% (crude)
BnHN
yield: 62% (crude)
yield: (quant.)
Boc
Boc
Bn-Gly-OH
13
14
EtO
O
50% in toluene
added within 15 h
O
Bn
BnN
H
BnN
O
N
LiBH4, THF
r.t., 6 d
Et3N, MgSO4,
toluene, reflux, another 6 h
O
O
H
BnN
BnN
BnHN
yield: 76% (crude)
purity: 70% (HPLC)
(+ 13% diastereoisomer)
yield: 98% (crude)
purity: 74% (HPLC)
(+6% diastereoisomer)
OEt
OH
15
16
17
rac-
rac-
O
i) TFAA, Et3N,
reflux, 135 h
ii) NaOH, r.t., 15 min
BnN
BnN
BnN
Lipase QLM, vinyl acetate
TBME, 30°C, 6 d
(S)
(R)
O
O
O
(S)
(R)
(R)
+
H
H
H
(S)
BnN
BnN
BnN
chromatography
yield: 86% (crude)
purity: 76% (HPLC)
(+ 11% rac-
OH
OAc
OH
18
19
18
(+)-
rac-
11
)
yield: 41% (chrom.)
purity: 75% (HPLC)
ee: 98.2%
yield: 44% (chrom.)
purity: 87% (HPLC)
ee: 95.7%
The end game of the synthesis dealt with protecting group manipulations, inversion of the
stereo center at the 3-position, introduction of the hydroxylamine motif, urea and salt
formation (scheme 4). The saponification of 19 smoothly proceeded with NaOH in
THF/MeOH 2:1 in quantitative yield. The double debenzylation turned out to be more
difficult than expected. After considerable optimization efforts, we chose Birch
conditions18 using Li and EtOH to selectively remove the benzyl group at the lactam
ACS Paragon Plus Environment
9

Organic Process Research & Development
Page 10 of 47
1
2
3
4
5
6
7
8
9
nitrogen first, followed by hydrogenation under standard conditions for the second benzyl
cleavage. Interestingly, the two reactions also worked in the reverse order and turned out
to be completely orthogonal to each other. In situ Boc protection during the hydrogenation
step resulted in 77% yield of 21 for the sequential double debenzylation sequence. The
protecting group swap from Bn to Boc was necessary for two reasons: firstly, to avoid
interaction of the tertiary amine during the following Mitsunobu reaction and secondly,
because hydrogenation was not appropriate after the introduction of the hydroxylamine
motif. The inversion of the stereo center at the 3-position of 21 was accomplished by
formation of the 4-nitrobenzoic ester under Mitsunobu conditions19 followed by
saponification with K2CO3 in MeOH; chromatography increased the purity to 94% by
HPLC and gave 23 in 68% yield for the inversion. Mitsunobu condition19 with N-
(benzyloxy)-2-nitrobenzenesulfonamide (24) resulted in the formation of 25, which after
Boc deprotection and chromatography yielded 26 in 51% yield (over 2 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
Scheme 4. Azomethine ylide route part 2: stereo adjustments, urea and salt formation
ACS Paragon Plus Environment
10

Page 11 of 47
Organic Process Research & Development
1
2
3
4
5
6
NH3 (l), Li,
EtOH, THF
-78 to -63°C
H2 (atm.), Pd/C,
Boc2O, MeOH
22-25°C, 89 h
BnN
H
BnN
HN
NaOH,THF/MeOH 2:1
25-40°C, 65 h
(S)
(S)
(S)
O
O
O
(S)
(R)
(S)
(R)
(S)
(R)
H
BnN
H
BnN
BnN
yield: 79% (cryst.)
purity: 86% (HPLC)
yield: quant. (crude)
purity: 75% (HPLC)
yield: 98% (crude)
purity: 78% (HPLC)
OAc
OH
OH
7
19
18
20
(-)-
8
9
DIAD, PPh3,
O-Bn-2-nitrobenzene-
sulfonamide , THF
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
PPh3, DIAD, THF
4-nitrobenzoic acid
-10°C to 20°C, 15 h
HN
K2CO3, MeOH
20°C, 1 h
(S)
24
O
HN
HN
10°C, 94 h
used crude
OH
(S)
(S)
(S)
(S)
H
BocN
O
O
(S)
(R)
(S)
H
BocN
H
BocN
O
yield: 68% (over 2 steps,
chrom. + cryst.)
purity: 94% (HPLC)
(S)
used crude
OH
O
21
22
23
NO2
HN
HN
H
(S)
(S)
O
O
PhSH, K2CO3,
MeOH
22 h, r.t.
COCl2, Et3N,
ACN, 30 mM
5°C, 4 h
HN
(S)
(R)
(S)
(R)
H
TFA, DCM
75 min, r.t.
(S)
BocN
HN
O
NOBn
NOBn
(S)
(R)
H
O
S
O
O S O
HN
yield: 84% (chrom.)
NO2
purity: 95% (NMR)
yield: 51% (chrom.)
purity: 93%
NHOBn
yield: 81%
NO2
(chrom. + cryst.)
purity: 99.6% /HPLC)
ee: 100%
25
26
27
(over 2 steps)
i) SO3*py, pyridine, r.t., 18 h
ii) tetrabutylammonium-
hydrogensulfate
aq. NaH2PO4, DCM,
r.t., 30 min
sodium 2-ethyl-
hexanoate
isobutanol/water
40°C, 1 h
HN
HN
HN
H
H2 (atm.), Pd/C 10%
DCM/MeOH 1:1
r.t., 90 min
(S)
(S)
(S)
O
+NBu4
O
O
(S)
H
(S)
(S)
H
(R)
N
O
S
1
(R)
(R)
N
N
N
O
O-
yield: 73% (crude)
purity: 95% (NMR)
yield: quant. (crude)
yield: 85% (chrom.)
N
OBn
N OH
O
O
O
O
28
29
30
Nosyl deprotection under standard condition20 resulted in the formation of 27 in 84% yield.
The formation of the cyclic urea was performed by slow addition of a phosgene solution
(syringe pump) into a 30 mM solution of 27 and resulted in the key intermediate 28 in
81% yield after chromatography and crystallization. The chemical purity after this step
was increased to 99.6% by HPLC and the ee to 100%. This material was identical with
the one obtained by the original photochemical LSF approach starting from an Avibactam
ACS Paragon Plus Environment
11

Organic Process Research & Development
Page 12 of 47
1
2
3
4
5
6
7
8
9
precursor. In addition, an X-ray crystal structure determination of 28 confirmed its
absolute stereochemistry. Debenzylation with H2 on Pd/C gave 29 in quant. yield.
Treatment with sulfurtrioxide pyridine complex followed by the addition of
tetrabutylammonium hydrogensulfate in aq. sodium dihydrogenphosphate solution
provided 30 in 85% yield after chromatography. Cation exchange was accomplished by
treatment of 30 with sodium 2-ethylhexanoate that facilitated the undesired ion pair (e.g.
the tetrabutylammonium 2-ethylhexanoate) to precipitate.21 The target compound 1 was
isolated in 73% yield and with a purity of >95% by NMR. The enantiopurity was assumed
to be identical to the one measured for 28, since we did not observe any epimerization
during the transformation to 1.
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
CONCLUSIONS
In order to ensure material supply for preclinical toxicological studies, we developed a de-
novo synthesis route for the novel β-lactamase inhibitor IID572 (1). This was required due
to a rather low yielding discovery route and unavailability of key starting materials within
the necessary timelines. The herein described synthesis started from affordable,
ACS Paragon Plus Environment
12

Page 13 of 47
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
commercially available starting materials and gave the target compound 1 between 1.7%
and 2.9% overall yield in an 19-step linear sequence (total number of steps: 20, including
the one step synthesis of 24). The elegance of this synthesis is the rapid and efficient
buildup of complexity from simple precursors and the application of an economic
enzymatic resolution to provide the desired enantiomer in high ee. Despite some synthetic
challenges with respect to functional group interconversions (scheme 4), the synthesis
was successfully performed on a 8 mol scale of Bn-Gly-OH. This allowed us to deliver 10
g of 1 and significant amounts of various intermediates for further SAR optimization.
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 reagents were purchased and used as received unless otherwise noted. The
larger amount of 24 was prepared according to the literature in one step.22 Palladium on
charcoal (10%) from BASF 549823. Lipase QLM (from Alcaligenes sp) from Meito-Sangyo
Co. Ldt. Melting ranges were determined on a Leitz Biomed microscope with integrated
hot plate, thermometer and an applied heat rate of 2°C/min. HPLC methods 1-3: XDB-
C18 column, 4.6 mm × 50 mm, 1.8 μm, using ACN and water as eluent (both containing
ACS Paragon Plus Environment
13

Organic Process Research & Development
Page 14 of 47
1
2
3
4
5
6
7
8
9
0.05% TFA), column temperature of 35°C, flow rate of 1.0 mL/min, detection at 215 nm,
purities were characterized with area%; method 1: gradient from 40 to 100% ACN over 6
min, 100% ACN for 1.5 min, followed by 100 to 40% ACN over 0.5 min; method 2: gradient
from 5 to 100% ACN over 6 min, 100% ACN for 1.5 min, followed by 100 to 5% ACN over
0.5 min; method 3: gradient from 30 to 100% ACN over 6 min, 100% ACN for 1.5 min,
followed by 100 to 30% ACN over 0.5 min. Chiral HPLC method 1: Chiralpak IC, 4.6 mm
× 250 mm, 5 μm, heptane/EtOH/diethylamine 92:8:0.05, column temperature: rt, flow 1
mL/min, detection at 220 nm; method 2: Chiralpak IC, 4.6 mm × 250 mm, 5 μm,
heptane/EtOH 9:1, column temperature: rt, flow 1 mL/min, detection at 210 nm; method
3: Chiralpak AD-H, 4.6 mm × 250 mm, 5 μm, heptane/EtOH 1:1, column temperature: rt,
flow 1 mL/min, detection at 220 nm. LCMS methods 1 and 2: Acquity HSS T3 1.8 μm 2.1
× 50 mm column at 50°C. Eluent A: water + 0.05% formic acid + 3.75 mM ammonium
acetate; eluent B: ACN + 0.04% formic acid, detection at 210-450 nm, purities were
characterized with area%; method 1: gradient from 2 to 98% B in 1.4 min with a flow rate
of 1.0 mL/min; method 2: gradient from 1 to 98% B (concave) in 1.4 min with a flow rate
of 1.2 mL/min; method 3: Water Acquity SDS, Kinetex C18 2.6 μm 2.1 × 50 mm column
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment
14

Page 15 of 47
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
at 50°C. Eluent A: water + 0.1% TFA; eluent B: ACN + 0.1% TFA, detection at 220 nm,
purities were characterized with area%; gradient from 2 to 88% B in 1.29 min and from
88% B to 95% B in 0.16 min, flow rate: 1.2 mL/min. LCMS method 4: Acquity HSS T3 1.8
μm 2.1 × 50 mm column at 60°C. Eluent A: water + 0.05% formic acid + 3.75 mM
ammonium acetate; eluent B: ACN + 0.04% formic acid, detection at 210-450 nm, purities
were characterized with area%; gradient from 5 to 98% B in 1.4 min with a flow rate of
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
1.0 mL/min. NMR spectra were recorded on a Bruker BioSpin machine. H shifts were
referenced to DMSO-d6 at 2.49 ppm and CDCl3 at 7.26 ppm. 13C shifts were referenced
to DMSO-d6 at 39.52 ppm. LC-HRMS: The analyses was performed by using electrospray
ionization in positive ion mode after separation by liquid chromatography (Vanquish,
Thermo). The elemental composition was derived from the mass spectra acquired at the
high resolution of about 240’000 on an Orbitrap Fusion Lumos mass spectrometer
(Thermo Scientific). The high mass accuracy below <1 ppm was obtained by using an
Internal Calibrant (IC). Optical rotations were measured for 18 on a Anton Paar MCP100
and for 1 on a Autopol IV automatic (Rudolph Research Analytical) polarimeter. A 30 L
Buechi hastelloy reactor CR30 equipped with Huber thermostat 1015W, Flexy ALR with
ACS Paragon Plus Environment
15

Organic Process Research & Development
Page 16 of 47
1
2
3
4
5
6
7
8
9
automated temperature control, nitrogen inlet and dosage control. 10L, 20 L and 30 L
triple jacketed AMSI Glas reactors equipped with a Huber Unistat 390W, automated
temperature control, condenser and nitrogen inlet. 60 L Buechi reactor CR60 equipped
with Huber thermostat 390W, Flexy ALR with automated temperature and dosage control,
nitrogen inlet and condenser. For inertization, the reactors were evacuated twice to 0.3
bar and refilled with nitrogen (or argon) before adding flammable solvents.
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
rac-(2S*,3aS*,6aS*)-Ethyl
1,5-dibenzyl-6-oxooctahydropyrroloα3,4-b]pyrrole-2-
carboxylate (rac-16) A 60 L reactor was loaded with 15 (1.46 kg, 4.81 mol) and toluene
(20 L). To the yellow solution magnesium sulfate (2.32 kg, 19.24 mol, 4 equiv), and
triethylamine (0.87 L, 6.25 mol, 1.3 equiv) was added. The reaction mixture (pale yellow
suspension) was heated to reflux (internal temperature ca 107°C) within 1 h. To the
refluxing reaction mixture an ethyl glyoxylate solution 50% in toluene (1.18 kg, 5.77 mol,
1.2 equiv) was added within 15 h via a dosage pump. The reaction mixture (pale yellow
suspension) was stirred for another 6 h at reflux (internal temperature ca 109°C). HPLC
of the reaction mixture showed full conversion. The reaction mixture was cooled to 15°C
ACS Paragon Plus Environment
16

Page 17 of 47
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
internal temperature, then water (20 L) was added (exotherm, the temperature raised to
32°C). The reaction mixture was stirred for 15 min, then transferred to a twin-70 L
separation vessel and the phases were separated (good separation). The organic phase
was washed with water (15 L) and brine (15 L). The aqueous layers were extracted twice
with TBME (2 × 10 L). The second TBME phase (contained a lot of watery mud) was
filtered through celite and washed with TBME. The water of the filtrate was separated and
the combined organic phases concentrated in vacuo at 45°C to a volume of ca 6 L. This
solution was dried over anhydrous sodium sulfate, filtered, concentrated in vacuo at 50°C
and dried overnight at 50°C and 10 mbar to obtain 1.97 kg of a turbid brown oil rac-16
(yield: 76% based on 70% HPLC-purity). The aqueous phases did not contain any product
and were discarded. HPLC (method 1): tR = 3.22 min (69.5%). LCMS (method 1): tR =
1.21 min (64.6%, product), m/z 379.3 [M+H]+; 1.16 min (12.4%, diastereoisomer), m/z
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
379.3 [M+H]+; 1.05 min (toluene H NMR (600 MHz, DMSO-d6) δ 7.43 – 7.22 (m,
16+5.6H), 4.43 (d, J = 14.9 Hz, 1H), 4.38 – 4.31 (m, 2H), 4.13 – 4.03 (m, 2+0.9H), 3.88
(d, J = 13.2 Hz, 1H), 3.74 (d, J = 8.9 Hz, 1H), 3.46 – 3.40 (m, 2H), 3.01 (ddd, J = 16.9,
8.4, 3.4 Hz, 1H), 2.96 (dd, J = 10.0, 2.8 Hz, 1H), 2.31 (s, 0.28H toluene), 2.13 (ddd, J =
ACS Paragon Plus Environment
17

Organic Process Research & Development
Page 18 of 47
1
2
3
4
5
6
7
8
9
13.0, 9.0, 2.4 Hz, 1H), 1.81 (dt, J = 13.1, 7.7 Hz, 1H), 1.66 (ddd, J = 12.6, 8.5, 6.7 Hz,
0.17H other diastereoisomer), 1.18 (t, J = 7.1 Hz, 4H), 1.04 (t, J = 7.1 Hz, 0.58H other
diastereoisomer). LC-HRMS: calcd for C23H27N2O3 [M+H]+ 379.20162, found 379.20169.
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
rac-(2S*,3aS*,6aS*)-1,5-Dibenzyl-2-(hydroxymethyl)hexahydropyrrolo[3,4-b]pyrrol-
6(1H)-one (rac-17). A 30 L hastelloy reactor was loaded with crude rac-16 (1.97 kg, 3.64
mol, 70% purity) and THF (20 L). The brown solution was cooled to 0°C then LiBH4 (238
g, 10.4 mol, 2.9 equiv) was added in portions over a period of 10 min (slightly exotherm).
The reaction mixture was warmed to 23°C. After 4 d another portion of LiBH4 (25 g, 1.1
mol, 0.3 equiv) was added. After another day additional LiBH4 (17 g, 0.8 mol, 0.2 equiv)
was added. After 6 d the conversion was complete. The reaction mixture was cooled to –
10°C. 2 N hydrochloric acid (ca 8 L, 16 mol, 4.4 equiv) was added dropwise via dosage
pump within two hours until pH 3 was reached (CAUTION: strong gas and foam evolution
was observed!). The solution became temporarily thick and a yellow suspension was
formed. After the addition the reaction mixture was stirred for 30 min at 0°C then sat.
sodium bicarbonate solution (10 L) was added. The reaction mixture was transferred to
ACS Paragon Plus Environment
18

Page 19 of 47
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
the 70 L-separation vessels and extracted with TBME (1 × 20 L and 2 × 7 L) and water
(ca 8 L). The combined org. phases were washed twice with brine (2 × 10 L) and
concentrated in vacuo at 45°C to a volume of ca 8 L. The residue was dried over
anhydrous sodium sulfate (overnight). The suspension was filtered, the filter cake washed
with TBME and the filtrate concentrated in vacuo at 50°C. The residue was dissolved in
toluene (3 L), evaporated again and dried (ca 3 h) at 50°C and 10 mbar to obtain 1630 g
of a turbid yellow-brown oil rac-17 (yield: 98% based on 74% HPLC-purity). HPLC
(method 2): tR = 3.64 min (74.1%). LCMS (method 1): tR = 0.77 min (73.9%, product), m/z
337.3 [M+H]+; 0.80 min (6.7%, diastereoisomer), m/z 337.3 [M+H]+; 1.05 min (toluene).
The crude material contained about 7% of the diastereoisomer and 3% toluene and was
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
used without further purification. A small sample was purified for NMR: H NMR (600
MHz, DMSO-d6) δ 7.44 – 7.19 (m, 10H), 4.52 – 4.45 (m, 2H), 4.29 (dd, J = 36.5, 14.3 Hz,
2H), 4.08 (d, J = 13.7 Hz, 1H), 3.58 (d, J = 8.1 Hz, 1H), 3.45 (ddd, J = 10.9, 5.5, 4.0 Hz,
1H), 3.39 – 3.32 (m, 2H?+H2O signal), 2.92 (dd, J = 9.9, 2.3 Hz, 1H), 2.87 (qd, J = 6.3,
3.9 Hz, 1H), 2.76 (dtdd, J = 10.0, 7.6, 5.1, 2.2 Hz, 1H), 2.00 – 1.94 (m, 1H), 1.60 (ddd, J
= 12.6, 7.2, 5.1 Hz, 1H), 1.18 (t, J = 7.1 Hz, 5% EtOAc). LC-HRMS: calcd for C21H25N2O2
ACS Paragon Plus Environment
19

Organic Process Research & Development
Page 20 of 47
1
2
3
4
5
[M+H]+ 337.19105, found 337.19104.
6
7
8
9
rac-(3R*,4aS*,7aS*)-1,6-Dibenzyl-3-hydroxyoctahydro-7H-pyrrolo[3,4-b]pyridin-7-one
(rac-18). A 30 L hastelloy reactor was loaded with crude rac-17 (1.63 kg, 3.59 mol, 74%
purity), molecular sieves 4 Å (2.5 kg) and THF (23 L). The resulting brown solution was
cooled to –5°C, then trifluoroacetic anhydride (0.820 L, 5.81 mol, 1.6 equiv) was added
dropwise within 35 min. The reaction mixture was stirred for 1.5 h while warming to rt. To
the reaction mixture was added triethylamine (3.37 L, 24.2 mol, 6.7 equiv) within 10 min,
then heated at reflux (jacket temperature 75°C) for 135 h. During this time the ratio of the
product (after hydrolysis) and its 5-membered ring isomer changed from 55:45 (after 15
h) to 90:10 (equilibrium). For the IPCs a samples of the reaction mixture was quenched
with 0.5 N NaOH and checked with LCMS. The reaction mixture was cooled to rt and
poured into a 70 L vessel containing 1 N aq. NaOH solution (24 L, 24 mol, 6.7 equiv) and
stirred for 15 min. To the resulting brown suspension was added celite (3 kg) and the
suspension stirred for 15 min after which it was filtered over a pad of celite to remove of
the molecular sieves and washed with TBME (filtration took ca 5 h). To the filtrate was
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
20

Page 21 of 47
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
added TBME (20 L) and after extraction and phase separation the aq. phase was
extracted with TBME (2 × 7 L). The combined org. phases were washed with sat. sodium
bicarbonate solution (10 L) and brine (15 L) and concentrated in vacuo at 45°C to a
volume of ca 8 L. After drying with anhydrous sodium sulfate (2 kg) the suspension was
filtered through silica gel (40-63 μm, 1 kg) to remove baseline impurities. The silica gel
bed was washed with EtOAc (4 × 2 L) until no product eluted. The filtrate was
concentrated in vacuo at 45°C and dried 3 h at 50°C and 15 mbar to obtain 1366 g rac-
18 as a dark brown oil, which was used crude for the next step (yield: 86% based on 76%
HPLC-purity (the rest was mostly sm)). HPLC (method 2): tR = 3.70 min (75.8%). LCMS
(method 1): tR = 0.84 min (65.9%, product), m/z 337.3 [M+H]+; 0.77 min (9.9%, starting
material) ES+ m/z 337.3 [M+H]+; 0.86 min (7.7%, diastereoisomer), m/z 337.3 [M+H]+. A
small sample was purified for NMR: 1H NMR (600 MHz, DMSO-d6) δ 7.39 – 7.21 (m, 10H),
4.67 – 4.59 (m, 2H), 4.43 (d, J = 13.9 Hz, 1H), 4.25 (d, J = 15.1 Hz, 1H), 3.54 (tq, J = 9.1,
4.3 Hz, 1H), 3.41 (d, J = 13.9 Hz, 1H), 3.18 (d, J = 8.2 Hz, 2H), 2.85 (d, J = 5.7 Hz, 1H),
2.70 (ddd, J = 10.9, 3.9, 1.4 Hz, 1H), 2.61 (ddt, J = 8.5, 5.8, 2.8 Hz, 1H), 1.78 – 1.68 (m,
2H), 1.34 (ddd, J = 13.4, 10.0, 6.1 Hz, 1H). LC-HRMS: calcd for C21H25N2O2 [M+H]+
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
21

Organic Process Research & Development
Page 22 of 47
1
2
3
4
5
337.19105, found 337.19113.
6
7
8
9
(3R,4aS,7aS)-1,6-Dibenzyl-7-oxooctahydro-1H-pyrrolo[3,4-b]pyridin-3-yl acetate (19). A
30 L glass reactor was loaded with crude rac-18 (1.36 kg, 3.05 mol, 76% purity), TBME
(21 L), vinyl acetate (4.20 L, 45.6 mol, 15 equiv) and lipase QLM (25 g, activity: 101400
U/g). The suspension was stirred (200 rpm) at 30°C internal temperature. After 143 h
HPLC the reaction mixture showed a 50:50 mixture of alcohol and acetate. The reaction
mixture was cooled to 20°C and filtered over celite (0.5 kg). The filtrate was concentrated
in vacuo at 35°C to a volume of 3 L then toluene (1 L) was added and concentrated in
vacuo first at 35°C then at 50°C to dryness. 1689 g of a dark brown honey was obtained.
This crude product was dissolved in TBME / heptane 2:1 (4.4 L) and purified in 14 portions
(12 + 2 mixed fractions) by column chromatography. The same batch of silica gel (7 kg,
25-40 μm, 60Å, Macherey-Nagel) in an axially compressed chromatography column was
used for the entire purification; a ternary gradient (heptane/EtOAc/MeOH) was applied,
followed by a wash program (1/2 column volume MeOH) and re-conditioning, flow rate 1
L/min. The purification resulted in 626 g brown honey 19 (yield: 41% based on 75%
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
22

Page 23 of 47
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
HPLC-purity). HPLC (method 3): tR = 2.47 min (75.0%). Chiral HPLC (method 1):
98.2%ee, tR (3S,4aR,7aR-enantiomer) = 23.95 min, tR (3R,4aS,7aS-enantiomer 2) =
33.75 min. LCMS (method 1): tR = 1.16 min (75.5%, product), m/z 379.3 [M+H]+; 1.15 min
(15.6%, 5-membered ring isomer (signal at 3.62 ppm)), m/z 379.3 [M+H]+. 1H NMR (600
MHz, DMSO-d6) δ 7.47 – 7.15 (m, 10+4H), 4.72 (qd, J = 6.0, 3.1 Hz, 1H), 4.55 (d, J =
14.9 Hz, 1H), 4.32 (dd, J = 38.6, 14.5 Hz, 2H), 3.75 (d, J = 14.3 Hz, 1H), 3.29 – 3.21 (m,
2H), 3.00 (dd, J = 9.6, 5.8 Hz, 1H), 2.71 (dd, J = 11.6, 3.2 Hz, 1H), 2.62 (h, J = 6.2 Hz,
1H), 2.23 (dd, J = 11.6, 7.1 Hz, 1H), 1.97 (s, 3H), 1.66 (t, J = 6.1 Hz, 2H). LC-HRMS:
calcd for C23H27N2O3 [M+H]+ 379.20162, found 379.20172.
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
(3S,4aR,7aR)-1,6-Dibenzyl-3-hydroxyoctahydro-7H-pyrrolo[3,4-b]pyridin-7-one ((+)-18)
From the previously described chromatography 510 g of the undesired enantiomer
(alcohol form) as a dark brown honey was collected from chromatography runs 1-12 (+)-
25
18 (yield: 44% based on 87% HPLC-purity). [α]D +59.6 (c=1.0 w/v%, CHCl3). HPLC
(method 3): tR = 1.58 min (86.7%). LCMS (method 1): tR =0.85 min (85.6%, product), m/z
337.3 [M+H]+; 0.87 min (9.7%, 5-membered ring isomer), m/z 337.3 [M+H]+. Chiral HPLC
ACS Paragon Plus Environment
23

Organic Process Research & Development
Page 24 of 47
1
2
3
4
5
6
7
8
9
1
(method 2): 95.7%ee, tR (3R,4aS,7aS) = 15.98 min, tR (3S,4aR,7aR) = 16.94 min. H
NMR (600 MHz, DMSO-d6) δ ppm 1.34 (ddd, J=13.43, 9.95, 6.14 Hz, 1 H) 1.69 - 1.79
(m, 2 H) 2.56 - 2.65 (m, 1 H) 2.70 (dd, J=10.36, 3.03 Hz, 1 H) 2.86 (d, J=5.87 Hz, 1 H)
3.18 (d, J=8.25 Hz, 2 H) 3.42 (d, J=13.75 Hz, 1 H) 3.50 - 3.60 (m, 1 H) 4.21 - 4.30 (m, 1
H) 4.43 (d, J=13.94 Hz, 1 H) 4.57 - 4.65 (m, 2 H) 7.16 - 7.44 (m, 12 H) contained 2%
residual solvent. LC-HRMS: calcd for C21H25N2O2 [M+H]+ 337.19105, found 337.19107.
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
(3R,4aS,7aS)-1,6-Dibenzyl-3-hydroxyoctahydro-7H-pyrrolo[3,4-b]pyridin-7-one ((–)-18).
A 20 L flask of a rotavapor was evacuated twice to 0.3 bar and re-filled with nitrogen and
loaded with 19 (616 g, 1.22 mol, 75% purity), THF (4 L) and aq. 2 N NaOH (3.97 L, 7.94
mol, 6.5 equiv). The biphasic mixture was vigorously stirred for 18 h at 25°C and for
another 7 h at 40°C in the rotavapor water bath. IPC (HPLC) showed 19% sm. MeOH (2
L) was added and the reaction mixture was stirred for 18 h at 25°C, 7 h at 40°C and
another 15 h at rt, then HPLC indicated full conversion. The reaction mixture was
extracted with TBME (1 × 6 L and 3 × 3 L). The combined organic phases were washed
with brine (4 L) and concentrated in vacuo at 45°C to a volume of 5 L. Anhydrous sodium
ACS Paragon Plus Environment
24

Page 25 of 47
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
sulfate (1 kg) was added, the suspension filtered and the filtrate concentrated in vacuo at
45°C. The residue was re-dissolved in toluene (3 L) and concentrated again and dried for
2 h at 60°C and 20 mbar to obtain 555 g of a brown honey (–)-18 (yield: quant. based on
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
25
76% HPLC-purity). [α]D –56.1 (c=1.0 w/v%, CHCl3). HPLC (method 3): tR = 1.58 min
(76.4%). LCMS (method 1): tR = 0.78 min (14.0%, 5-membered ring isomer), ES+ m/z
337.3 [M+H]+; 0.84 min (74.3%, product), ES+ m/z 337.3 [M+H]+. H NMR (600 MHz,
1
DMSO-d6) δ 7.39 – 7.19 (m, 13H), 4.65 – 4.59 (m, 2H), 4.42 (d, J = 13.9 Hz, 1H), 4.32 (d,
J = 13.7 Hz, 0.28H (main impurity)), 4.25 (dd, J = 15.0, 3.0 Hz, 1H), 3.54 (tq, J = 9.1, 4.4
Hz, 1H), 3.42 (d, J = 13.9 Hz, 1H), 3.18 (d, J = 8.2 Hz, 2H), 2.85 (d, J = 5.8 Hz, 1H), 2.70
(ddd, J = 10.8, 3.8, 1.4 Hz, 1H), 2.65 – 2.56 (m, 1H), 1.78 – 1.70 (m, 2H), 1.34 (ddd, J =
13.4, 9.9, 6.1 Hz, 1H). LC-HRMS: calcd for C21H25N2O2 [M+H]+ 337.19105, found
337.19104.
(3R,4aS,7aS)-1-Benzyl-3-hydroxyoctahydro-7H-pyrrolo[3,4-b]pyridin-7-one
(20)
Equipment: 30 L hastelloy reactor with an ammonia inlet and an argon inlet attached to a
gas scrubber filled with 100 L aq. 30% sulfuric acid solution. The reactor was cooled to –
ACS Paragon Plus Environment
25

Organic Process Research & Development
Page 26 of 47
1
2
3
4
5
6
7
8
9
80°C and loaded with liquid ammonia (12.7 L, 587 mol) (note: the gas cylinders were pre-
cooled in dry ice for about 10 min), a solution of (–)-18 (543 g, 1.21 mol, purity 76%) in
THF (1.5 L) and anhydrous EtOH (236 mL, 4.04 mol, 3.3 equiv). Finally, granular lithium
(99% trace metals basis, 44.8 g, 6.46 mol, 5.3 equiv) was added portion wise within 15
min (temperature raised from –72°C to –63°C). After 5 min the color changed from yellow
to gray to dark blue. The reaction mixture (suspension) was stirred at –65°C internal
temperature. After 10 min, the color changed back to gray. 1st IPC (HPLC) indicated 45%
remaining starting material. To the reaction mixture lithium (22.4 g, 3.23 mol, 2.7 equiv)
were added at –63°C (temperature raised to –56°C, the color changed immediately to
dark blue). To the reaction mixture anhydrous EtOH (94 mL, 1.62 mol, 1.3 equiv) was
added and stirred at –60°C. The color changed to gray after 30 min. 2nd IPC (HPLC)
indicated 20% remaining starting material. Lithium (11.2 g, 1.62 mol, 1.3 equiv) and
anhydrous EtOH (47 mL, 0.81 mol, 0.7 equiv) was added at –60°C (blue color). After 45
min anhydrous EtOH (47 mL, 0.81 mol, 0.7 equiv) was added at –60°C and after further
10 min the color turned gray. 3rd IPC (HPLC) indicated 10% remaining starting material.
Lithium (11.2 g, 1.62 mol, 1.3 equiv) was added, the color remained blue after stirring
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
26

Page 27 of 47
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
overnight at –60°C. Anhydrous EtOH (94 mL, 1.62 mol, 1.3 equiv) was added, the color
turned gray after 15 min. 4th IPC (HPLC) indicated 3-5% remaining starting material. The
reaction mixture was quenched by the addition of solid ammonium chloride (2003 g, 37.4
mol, 31 equiv) portion wise within 10 min (temperature raised to –42°C). The reaction
mixture was warmed (–28°C jacket temperature) and stirred overnight, then the jacked
temperature was raised to –2°C to allow the complete evaporation of ammonia that was
all neutralized in the gas scrubber. To the residue water (15 L) and TBME (8 L) was added
(pH 12-13), followed by aq HCl 32% until pH 9-10 was obtained (product is soluble in
water at pH 12-13) and extracted. The aqueous phase was washed with DCM (3 × 2 L).
The combined organic phases were washed with brine (5 L) and concentrated in vacuo
at 45°C to a volume of ca 3 L. The residue was dried over anhydrous sodium sulfate (1
kg). The suspension was filtered and the filtrate was concentrated in vacuo at 45°C. Then
dried for 2 h at 65°C and 20 mbar to deliver 373 g of a brown resin 20 (yield: 98% based
on 78% HPLC-purity). HPLC (method 2): tR = 2.17 min (78.5%). LCMS (method 2): tR =
0.75 min (ca 87%, product), ES+ m/z 247 [M+H]+. A small sample was purified for NMR:
1H NMR (600 MHz, DMSO-d6) δ 7.79 (s, 1H), 7.40 – 7.16 (m, 5H), 4.62 (d, J = 4.9 Hz,
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
27

Organic Process Research & Development
Page 28 of 47
1
2
3
4
5
6
7
8
9
1H), 4.32 (d, J = 13.9 Hz, 1H), 3.62 (tq, J = 9.2, 4.4 Hz, 1H), 3.43 (d, J = 13.9 Hz, 1H),
3.13 (d, J = 8.1 Hz, 2H), 2.69 (ddd, J = 10.7, 3.9, 1.4 Hz, 1H), 2.64 (d, J = 6.0 Hz, 1H),
2.57 (ttd, J = 8.5, 5.9, 3.2 Hz, 1H), 1.73 (ddd, J = 19.8, 9.7, 6.5 Hz, 2H), 1.34 (ddd, J =
13.3, 9.9, 6.1 Hz, 1H). LC-HRMS: calcd for C14H19N2O2 [M+H]+ 247.14410, found
247.14412.
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
tert-Butyl
(3R,4aS,7aS)-3-hydroxy-7-oxooctahydro-1H-pyrrolo[3,4-b]pyridine-1-
carboxylate (21) To a solution of 20 (372 g, 1.18 mol, purity 78%) and Boc2O (270 g, 1.24
mol, 1.05 equiv) in THF (4 L) was added palladium on charcoal 10% (15 g, 0.014 mol,
0.01 equiv) and the reaction mixture hydrogenated at 0.1 bar and 22-25°C in a shaking
duck reactor. After 18 h (57% hydrogen consumption) another portion of palladium on
charcoal 10% (15 g, 0.014 mol, 0.01 equiv) was added. After total 89 h complete
conversion was observed by HPLC. The reaction mixture was filtered through celite,
washed with THF and concentrated in vacuo to obtain 545 g (wet) of the crude product
as a pale brown solid. The crude product was suspended in EtOAc (1 L) and stirred well
for 1 h at 75°C bath temperature. To the pale brown suspension heptane (1.5 L) was
ACS Paragon Plus Environment
28

Page 29 of 47
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
slowly added at 75°C and the suspension stirred for 2 h at rt. The product was filtered off
and the solid was washed with heptane in several portions. The solid was dried in vacuo
at 45°C until constant weight was obtained: 278.5 g 21 as white crystals (yield: 79% based
on 86% HPLC-purity). Melting range: 172-174°C. HPLC (method 2): tR = 2.55 min
(86.1%). LCMS (method 2): tR = 0.86 min (ca 70%, product), ES+ m/z 257.3 [M+H]+. 1H
NMR (600 MHz, DMSO-d6) δ 7.72 (d, J = 21.2 Hz, 1H), 4.80 – 4.49 (m, 2H), 3.76 (d, J =
45.5 Hz, 2H), 3.37 (s, 1H), 2.72 (d, J = 9.8 Hz, 1H), 2.60 – 2.44 (m, 2H), 1.82 – 1.71 (m,
1H), 1.47 – 1.36 (m, 9H), 1.31 (q, J = 12.9, 11.1 Hz, 1H). LC-HRMS: calcd for C12H21N2O4
[M+H]+ 257.14958, found 257.14960.The mother liquor was concentrated in vacuo at
45°C to obtain 162 g as a brown oil that contained <5% product according to HPLC and
was discarded.
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
tert-Butyl
(3S,4aS,7aS)-3-((4-nitrobenzoyl)oxy)-7-oxooctahydro-1H-pyrrolo[3,4-
b]pyridine-1-carboxylate (22) Equipment: 20 L glass reactor. A pale brown suspension of
21 (270 g, 0.91 mol, purity 86%), in THF (13 L) was cooled to –10°C. 4-nitrobenzoic acid
(323 g, 1.90 mol, 2.1 equiv) and triphenylphosphine (524 g, 1.90 mol, 2.1 equiv) were
ACS Paragon Plus Environment
29