Convergent Synthesis of the Renin Inhibitor Aliskiren Based on C5-C6 Disconnection and CO2H-NH2 Equivalence

Article abstract of DOI:10.1021/acs.oprd.5b00396

A novel synthesis of the renin inhibitor aliskiren based on an unprecedented disconnection between C5 and C6 was developed, in which the C5 carbon acts as a nucleophile and the amino group is introduced by a Curtius rearrangement, which follows a simultaneous stereocontrolled generation of the C4 and C5 stereogenic centers by an asymmetric hydrogenation. Operational simplicity, step economy, and a good overall yield makes this synthesis amenable to manufacture on scale.

Full text of DOI:10.1021/acs.oprd.5b00396

Subscriber access provided by UNIV OF NEBRASKA - LINCOLN
Full Paper
Convergent Synthesis of Renin Inhibitor Aliskiren Based
on C5-C6 Disconnection And CO2H - NH2 Equivalence
Elena Cini, Luca Banfi, Giuseppe Barreca, Luca Carcone, Luciana Malpezzi,
Fabrizio Manetti, Giovanni Marras, Marcello Rasparini, Renata Riva, Stephen
Roseblade, Adele Russo, Maurizio Taddei, Romina Vitale, and Antonio Zanotti-Gerosa
Org. Process Res. Dev., Just Accepted Manuscript • DOI: 10.1021/acs.oprd.5b00396 • Publication Date (Web): 05 Jan 2016
Downloaded from http://pubs.acs.org on January 6, 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 45
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Convergent Synthesis of Renin Inhibitor Aliskiren
Based on C5ꢀC6 Disconnection And CO2H − NH2
Equivalence
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
Elena Cini,† Luca Banfi,¢ Giuseppe Barreca††, Luca Carcone††, Luciana Malpezzi,§ Fabrizio
Manetti†, Giovanni Marras,†† Marcello Rasparini,††,±,* Renata Riva,¢ Stephen Roseblade,¶ Adele
Russo,† Maurizio Taddei,†* Romina Vitale,¢ Antonio ZanottiꢀGerosa¶
†Dipartimento di Biotecnologie, Chimica e Farmacia, Università degli Studi di Siena, Via A.
Moro 2, 53100 Siena, Italy
††Chemessentia srl, Via Bovio 6, 28100 Novara, Italy
¢Dipartimento di Chimica e Chimica Industriale, Università degli Studi di Genova, Via
Dodecaneso 31, 16146 GENOVA
§Dipartimento di Chimica, Materiali e Ingegneria Chimica “G. Natta”, Politecnico di Milano, Via
Mancinelli 7, 20131 Milano, Italy
¶Johnson Matthey Catalysis and Chiral Technologies, 28 Cambridge Science Park, Milton Road,
Cambridge, CB4 0FP, United Kingdom
±Present address: Janssen Pharmaceutica PR&D, Turnhoutseweg 30, Bꢀ2340 Beerse (Belgium)
ACS Paragon Plus Environment
1

Organic Process Research & Development
Page 2 of 45
1
2
3
4
Toc graphic
5
6
7
8
9
Enzyme resolution
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
Catalytic enantioselective hydrogenation
Curtius rearrangement
Kg
ALISKIREN
ACS Paragon Plus Environment
2

Page 3 of 45
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
ABSTRACT. A novel synthesis of renin inhibitor Aliskiren was developed, based on an
unprecedented disconnection between C5 and C6, where the C5 carbon acts as nucleophile and
the amino group is introduced by Curtius rearrangement. Followed by simultaneous
stereocontrolled generation of C4 and C5 stereogenic centres by an asymmetric hydrogenation.
Operational simplicity, step economy and a good overall yield makes this synthesis amenable to
manufacture on 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
Keywords: Aliskiren, asymmetric catalysis, asymmetric hydrogenation, generic APIs
INTRODUCTION
The reninꢀangiotensin system (RAS) is the primary player of a complex process responsible
for the rapid increase of blood pressure, inflammation, kidney fibrosis, heart hypertrophy and
vasoconstriction.1ꢀ3 The classic therapeutic intervention is based on ACE (Angiotensin
Converting Enzyme) inhibitors to block the conversion of angiotensin I into angiotensin II or on
angiotensin II receptor blockers, and several drugs belonging to these classes are nowadays
available on the market.4 5ꢀ7 However, the RAS breakdown leads to a feedback increase of renin
production with the activation of a “nonꢀACE mediated” mechanism to produce angiotensin II.
Thus, longꢀterm treatment with ACE inhibitors may result in a general increase of plasmatic
renin activity with uncontrolled increasing blood pressure and other side effects.8 As renin is the
protease enrolled in the angiotensinogen cleavage to generate the angiotensin cascade, its
inhibition appeared as the most direct way to suppress the uncontrolled plasmatic renin increase
and prevents the problem related to the use of ACE inhibitors. However, although research
towards renin inhibitors started as early as in the late '70s of the last century, formidable
ACS Paragon Plus Environment
3

Organic Process Research & Development
Page 4 of 45
1
2
3
4
5
6
7
8
9
pharmacokinetic issues slowed down the discovery of efficient oral drugs. Only when the drug
candidates moved away from the typical peptideꢀlike systems were these problems solved, at
least in part, leading to Aliskiren 1, the first successful renin inhibitor.9,10 Marketing of Aliskiren
was also slowed down by the complexity of its structure, which makes the total synthesis rather
challenging. Contrary to most peptidomimetics, Aliskirien is characterized by a nearly allꢀcarbon
chain, apart from the amide group at its right terminus, and encompasses four stereogenic
centres. Poor bioavailability necessitates a relatively high dosage (typically 150 mg/day),
compared to ACE inhibitors (Ramipril: 5 mg/day) or angiotensin receptor blockers (Losartan: 50
mg/day), and thus the development of a costꢀeffective synthesis was particularly important since
a high yearly tonnage was expected.
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
In 2007, Aliskiren was licensed by Novartis in the US with the brand name Tekturna and in
Europe with the name Rasilez. The introduction of this new hypertension therapy raised
expectations for a pathology affecting more than 25% people worldwide, with a population of
patients expected to rise to > 1.5 billion people in 2025.11
Starting from the first NovartisꢀSpeedel synthesis,12ꢀ17 several research groups, both from
academy and industry, have reported brilliant solutions to the problem of Aliskiren synthesis.18ꢀ33
At the outset of this work, in view of an industrial scale preparation suitable for sale as a
generic drug, we looked for a proprietary, original approach endowed with these features: a)
good overall yield (> 10%); b) stepꢀeconomy (< 15 steps); c) amenable for large scale synthesis;
d) costꢀeffective; e) using, as much as possible, catalytic methods for the control of the
stereogenic centres (or allowing efficient recycling of any chiral auxiliary); f) avoidance of the
use of stoichiometric organometallic reagents; g) (and last but not least) non infringing previous
patents. The latter was probably the most challenging issue, considering the many previous
ACS Paragon Plus Environment
4

Page 5 of 45
Organic Process Research & Development
1
2
3
4
5
6
patent applications regarding Aliskiren synthesis, both from Novartis and competitor generic
producers.
7
8
9
Partial reports on our efforts have been already published.34,35 Here we give a full account of
the successful accomplishment of a synthesis that completely satisfies our initial objectives.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment
5

Organic Process Research & Development
Page 6 of 45
1
2
3
4
RESULTS AND DISCUSSION
5
6
7
8
9
Retrosynthesis. Our retrosynthetic plan is shown in Scheme 1. Most of the previous
approaches to the synthesis of Aliskiren are based on the aminolysis of lactone 2: we also
decided to aim at this advanced intermediate for two main reasons. First, from the available
literature, it was clear that 2 is one of the few crystalline Aliskiren late intermediates, thus
providing opportunity for efficient purification. Second, interception of Novartis synthesis at this
stage would be advantageous from a regulatory point of view, since the last steps of our
synthesis would coincide with those of the originator. For the preparation of 2, we chose to rely
on the synthetic equivalence between a carboxylic acid and an amine, which is granted by the
stereospecific Curtius rearrangement. The use of a carboxylate as masked amino group brings
about another advantage: the possibility to exploit ester enolates as nucleophiles in the crucial
carbonꢀcarbon bond formation that joins the two fragments to build up the eight carbon skeleton.
Having established Cꢀ5 as the nucleophilic carbon for this CꢀC bond forming event, two
disconnections in principle are possible: between Cꢀ4 and Cꢀ5 (route A, described by Novartis27)
or Cꢀ5 and Cꢀ6 (route B).
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
Therefore, looking for a practical and original approach, we decided to follow route B, relying
on an unprecedented C5ꢀC6 disconnection where C5 acts as the nucleophilic counterpart.36
Route B can be further divided in two branches: B1 and B2. They are differentiated by the way
the stereogenic centres at Cꢀ5 and Cꢀ4 are introduced, and by the timing of Curtius
rearrangement. In route B1 the two stereogenic centres are generated stepwise, by reduction of
the keto group and of the double bond on intermediate 10, whereas the Curtius rearrangement is
planned after formation of lactone 3. In route B2 the stereogenic centres are generated
ACS Paragon Plus Environment
6

Page 7 of 45
Organic Process Research & Development
1
2
3
4
simultaneously, by reduction of enol derivative 8 and the conversion of carboxylate into Boc
5
6
amine is carried out before lactonization.
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
Scheme 1. Retrosyntheses. R1 = MeO(CH2)3ꢀ
ACS Paragon Plus Environment
7

Organic Process Research & Development
Page 8 of 45
1
2
3
4
5
6
7
8
9
The key intermediates of the two routes are 9 and 10. The latter may derive from a
Knoevenagel condensation between ketoester 12 and aldehyde 11. On the other hand, 9, can
result from reduction of 10, or from an alkylation of 12 with iodide 13.
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 chiral building blocks 11 and 13. For the preparation of enantiopure aldehyde
11 and iodide 13 we envisioned as a common precursor alcohol 19, in turn derived from
reduction of acid 16 (Scheme 2). For the preparation of 1637ꢀ39 we first explored a catalytic
methodology, based on enzymatic kinetic resolution of esters 15.40,41 A series of esters were
initially produced by Fischer esterification of racemic 16 and a thorough study was carried out in
order to find the best conditions for this resolution. Initially, being concerned by the bulkiness of
our substrate, we chose methyl ester 15b (R2 = Me) because of the known higher reactivity in
enzyme mediated hydrolysis. A considerable number of lipases, proteases and esterases were
tested, but none of them was able to efficiently catalyse the reaction. Since traces of product
were detected with esterases from pig liver or from Mucor miehei, we concentrated on this class
of enzymes, eventually finding a good activity for Horse Liver Acetone Powder (HLAP)42,43
a
crude preparation, that was previously proved by us to be particularly efficient on sterically
encumbered substrates.44 However, conversion after 48 h was only 43.5% and the E factor45 (=
35) was too low for achieving a high ee of the formed acid 16, which, by comparison with
literature data, had the desired (R) configuration.38 Several other parameters were thus varied: a)
the nature of R2 group (Me, Et, CF3CH2, nPr, nBu); b) the addition of cosolvents; c) the type and
concentration of buffer; d) the substrate concentration; e) the temperature. Complete details of
this optimization may be found in the Supporting Information. We eventually found that with the
butyl ester 17a, a great increase in E value was observed and unexpectedly, also the reaction rate
increased. At a temperature of 40 °C, a substrate concentration of 0.043 M, with 0.3 M TRIS
ACS Paragon Plus Environment
8

Page 9 of 45
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
buffer and no organic cosolvent, an excellent E value of 147 was achieved. This allowed, with a
25 h reaction time, to obtain, at a 47.5% conversion, an ee of 96.2% for (R) 16. Separation of (R)
16 from unreacted (S) ester 15a was trivial and the latter could be racemized by treatment with
NaOH in EtOH, allowing a quantitative conversion of the unwanted enantiomer into racemic 16.
Esterification of the latter to 15a completed the full recycle of unwanted enantiomer. The ee of
acid 16 could be raised to 99.5% by neutralization with dry ammonia in isopropanol, filtration of
the solid and extractive recovery of the acid (95% 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
Scheme 2. Synthesis of building blocks 11 and 13. R1 = MeO(CH2)3ꢀ
Although in preliminary studies the racemic ester 17a was prepared from racemic acid 16 by
Fischer esterification, we later found that it could be directly obtained in good overall yield from
ACS Paragon Plus Environment
9

Organic Process Research & Development
Page 10 of 45
1
2
known aldehyde 1417, 18 by a sequence of aldol reaction with butyl isovalerate, acid promoted
dehydration, and hydrogenation.Enantiomerically pure acid (R)ꢀ16 was then reduced to the
alcohol 19, which was in turn oxidized to aldehyde 11 or converted into iodide 13 in excellent
overall yields.
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
Although the enzymatic resolution enabled us to produce the first gramꢀscale batches of 19,
which were crucial to prove the downstream chemistry, when we started to upscale this
enzymatic methodology, we found that its major drawbacks (low dilution, sourcing of the
enzyme and laborious workꢀup) made it unsuitable on multiꢀKg scale.
Therefore we draw our attention on an alternative route based on Evans' auxiliary. Thus,
Knoevenagel condensation of Evans' oxazolidinone 18 with aldehyde 14, followed by
hydrogenation and removal of the chiral auxiliary, gave acid (R)ꢀ16 in good overall yield.
However, the ee (80%) was too low for our purposes. Therefore we draw our attention on an
alternative route based on alkylation of oxazolidinone 18 with 4ꢀ(bromomethyl)ꢀ1ꢀmethoxyꢀ2ꢀ(3ꢀ
methoxypropoxy)benzene. This methodology was already known17,18 and would have been
freely operable at product patent expiry. Indeed, after the necessary fineꢀtuning to adapt it to our
plants, this route was eventually selected for the large scale preparation of acid (R)ꢀ16.
Synthesis of the C1ꢀC5 fragment. Enantiopure fragments 12 (Scheme 1; unambiguously
identified as 21 or 24ꢀ25 in Scheme 3) were prepared starting from the known isovaleryl
oxazolidinone (S)ꢀ18,46 derived from (S)ꢀphenyl alanine, that was alkylated with tertꢀ
butylbromoacetate in the presence of LDA at ꢀ78 °C. The reaction gave known product 2046ꢀ48 in
80% yield as a single diastereomer. The tertꢀbutyl protection was then removed with TFA and
the corresponding acid subjected to homologation with carbonyldiimidazole and magnesium
monoꢀethyl malonate49 to produce the βꢀketo ester 21 in 72% overall yield. In order to simplify
ACS Paragon Plus Environment
10

Page 11 of 45
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
the recovery of the chiral auxiliary, we also studied the conversion of 20 into an appropriate
ester. Using lithium benzyloxide, the benzyl ester 23 was obtained in high yield. Moreover,
under these reaction conditions, the recovery of the chiral auxiliary (89% yield) was
exceptionally efficient: a simple filtration was needed. As above, selective removal of the tertꢀ
butyl protection followed by the homologation with Mg monoꢀethyl or monoꢀtertꢀbutyl malonate
worked well giving βꢀketoesters 24 and 25 in 72% and 82% yields.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 3. Synthesis of C1ꢀC5 fragment and union of C1ꢀC5 and C6ꢀC8 fragments. R1 =
MeO(CH2)3ꢀ
Union of C1ꢀC5 and C6ꢀC8 fragments. For exploitation of the retrosynthetic route B1 we
needed to perform a Knoevenagel reaction between βꢀketoesters 21, 24 or 25 and aldehyde 11
(Scheme 3). This reaction was first tested on compound 21. Unexpectedly, classic conditions
ACS Paragon Plus Environment
11

Organic Process Research & Development
Page 12 of 45
1
2
3
4
5
6
7
8
9
using piperidine in DMF at 60 °C afforded the anticipated adduct 22a only in very low yields.
The major product was 22b, where piperidine had displaced the Evans auxiliary (70% overall
yield and 8:1 ratio between 22b and 22a). Both adducts were obtained as E:Z mixtures. Although
compound 22b could in principle, act as an effective intermediate for our plan, we were
concerned about the possibility of a challenging lactonization reaction, since previous Aliskiren
syntheses have shown that the lactone ꢀ amide equilibrium is normally shifted towards the amide.
Therefore we carried out a thorough investigation in order to find optimal conditions to obtain
22a. Details of this optimization are reported in the SI. The optimal solution was a Lewis acid
promoted Knoevenagel reaction: using TiCl4/pyridine in equimolar amount (1.2 eq respect to
aldehyde 11),50 compound 22a could be obtained in 46% yield, as a 60/40 mixture of E and Z
isomers.
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
Our optimized Knoevenagel reaction of benzyl ester 24 with aldehyde 11 produced the desired
26, however, in unsatisfactory yields and with low purity due to extensive TiCl4 catalyzed
transesterification. The unsuccessful Knoevenagel reaction prompted the search for a different
approach, using iodide 13 for the alkylation of βꢀketoester 24. Unfortunately, this reaction also
proved to be troublesome. With K2CO3 in acetone, the expected compound 27 was obtained
together with another product having the same mass that was attributed to the corresponding Oꢀ
alkylated derivative. An extensive optimization of the reaction conditions were carried out in
order to prevent the formation of the Oꢀalkylated impurity. After a long optimization process, we
discovered that the product distribution was highly influenced by the polarity of the solvent and
by the counterion of the base. We used the counterion effect and solvent dependence to our
advantage and selected LiH in DMA at 60 °C for 18 h to provide a highly efficient C alkylation
with complete retention of the original fragment configuration and produced exclusively the Cꢀ
ACS Paragon Plus Environment
12

Page 13 of 45
Organic Process Research & Development
1
2
3
4
5
6
alkylated product 27 in 71% isolated yield. Under these optimised conditions, the corresponding
tertꢀbutyl ester 28 was also obtained in 80% yield.
7
8
9
Investigation of B1 route. Having secured derivatives 22a and 22b, enabled us to explore the
feasibility of route B1 (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
Carbonyl reduction of 22b with NaBH4 in the presence of CeCl3 proceeded with complete
stereoselection and compound 29 was isolated in 54% yield after column chromatography.
Further double bond hydrogenation under standard conditions gave 30 as a 4:1 diastereomeric
mixture (nonꢀoptimized conditions).
Scheme 4. Attempted synthesis through retrosynthetic route B1. R1 = MeO(CH2)3ꢀ
To our surprise, working on compound 30, we noted that lactonization to give 31 occurred
very easily, upon treatment with even mild acids or bases. Therefore, it was not possible to
selectively hydrolyze the ester without promoting lactonization.
ACS Paragon Plus Environment
13

Organic Process Research & Development
Page 14 of 45
1
2
3
4
5
6
7
8
9
At first we were pleased by this result, that demonstrated that our concerns about difficult
lactonization of the piperidine derivative were not substantiated, and that 22b could be a valid
intermediate. However, after separation of the desired major stereoisomer from the minor one,
we were unable to remove the ethyl ester without affecting the lactone fragment. This outcome
made it impossible to carry out a Curtius rearrangement selectively on the carboxylic acid
appended to Cꢀ5 without touching also the Cꢀ1 carboxy group. The facile lactonization of amide
30, which prevented all attempts to protect the hydroxy group, was unexpected. The equilibrium
between esters and amides is normally shifted towards the latter, although notable exception are
reported in the literature in the case of tertiary amides.51 We therefore reasoned that with a
secondary amide the equilibrium position could be reversed, and installed the amidic part of the
final target at this level, with the goal of converting the CO2Et group in 32 into the free
carboxylic acid. However, when lactone 31 was treated with 3ꢀaminoꢀ2,2ꢀdimethylpropanamide
in the presence of AlMe3, although ES/MS of the crude reaction mixture showed the formation
of the expected compound 32, all attempts to purify it by column chromatography generated
again lactone 31 that was isolated as the main component of the reaction mixture. An analogous
behavior was observed also when we tried to prepare 33 by reaction of lactone 31 with
benzylamine/AlMe3. This strong propensity to lactonize of compounds 30, 32, and 33 depends
on the presence of the carboxylic ester in position 4, since no example of spontaneous
lactonization was reported with other intermediates used in Aliskiren synthesis, having a Boc
protected amino group in place of the CO2Et group at Cꢀ4.
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
To understand the effect of the CO2Et group, a systematic Monte Carlo Multiple Minimum
(MCMM) conformational search was performed. The problem was subjected to MacroModel
software,52 by using the OPLS_2005 force field53 and the PolakꢀRibiere conjugate gradient
ACS Paragon Plus Environment
14

Page 15 of 45
Organic Process Research & Development
1
2
(PRCG) minimizer on simplified models (with R1 = Me) of 30, 32, 33, and of the analogue of 30
having a NHBoc instead of a CO2Et group. Details on these computations are reported in the SI.
One of the decisive steps for the lactonization reaction is the adoption of an optimal
conformation, where the distance between the carbonyl atoms of the amide moiety is reduced to
approximately 3 Å from the oxygen atom of the hydroxy group. Moreover, according to the
BürgiꢀDunitz trajectory hypothesis,54 the angle between the approaching nucleophile and the
carbonyl group (OꢀꢀC=O) should be about 105 degrees. An analysis of conformer population
resulting from the calculations shows that the global minimum of the simplified model of benzyl
derivative 33 possesses both these optimal geometric requirements for a nucleophilic attack onto
the amide carbonyl. In fact, the distance between the hydroxyl oxygen and the carbonyl carbon
atoms is about 3.1 Å and the HOꢀꢀꢀC=O angle is 108 degrees (Figure 1).
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 1. Graphical representation of the minimum energy conformer of the benzyl derivative 33
(R1 = Me). The molecule is arranged in a folded conformation stabilized by a hydrogen bond
(upper black line) between the hydroxyl group and the carboxyl oxygen of the ester moiety, as
well as by a parallel stacked and displaced πꢀπ interaction between the two terminal aromatic
rings. Distance between the hydroxyl oxygen (nucleophile) and the carbonyl carbon atom is 3.1
Å, while the OhydroxylꢀꢀꢀC=O angle is 108 degrees.
ACS Paragon Plus Environment
15

Organic Process Research & Development
Page 16 of 45
1
2
3
4
5
6
7
8
9
This conformation is stabilized by a hydrogen bond between the hydroxyl group and the
carbonyl oxygen of the ester moiety. Moreover, πꢀπ interaction (parallel stacked and displaced)
is also found between the two terminal aromatic rings. Although, for the simplified models of 30
and 32, the global minimum does not possess the above cited optimal requirements, a significant
number of conformations with energy close to the minimum endowed with these features could
be found. On the contrary, replacement of the ester side chain with a NHBoc group led to a
significant modification of conformational behaviour, due to a hydrogen bond between the
secondary amide NH and the Boc carbonyl. This bond causes the global minimum to be
characterized by a distance of 3.8 Ǻ and an angle of about 131 degrees. None of the other
conformations with energy close to the global minimum satisfy the geometric requirements for
cyclization.
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
Investigation of B2 route and accomplishment of total synthesis of Aliskiren.
From the above described outcomes it was clear that in order to prevent lactonization, a
protected alcohol must be generated before the carboxylic acid/amine transformation. Therefore,
the alternative retrosynthetic route B2, depicted in Scheme 1, was pursued. According to this
strategy, the two stereogenic centres at Cꢀ4 and Cꢀ5 would be simultaneously generated by
hydrogenation of an enol derivative, providing an intermediate already protected at the hydroxy
group, and thus unable to lactonize. The transformation into enol acetates 34ꢀ35 (Scheme 5) was
of paramount importance, since the correct setting of the relative stereochemistry can be obtained
only from an Eꢀenol ether, in view of the fact that both heterogeneous and homogeneous
hydrogenations occur largely as syn additions. With regard to the absolute stereochemistry of
positions 4 and 5, we hoped to capitalize on the stereochemical bias created by the two isopropyl
groups flanking the C=C bond to be reduced.
ACS Paragon Plus Environment
16

Page 17 of 45
Organic Process Research & Development
1
2
3
4
5
6
We investigated this approach first on the ethyl ester 27 (Scheme 5). After conversion into enol
acetate 34, it was submitted, without purification, to a panel of hydrogenation conditions.34
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
Scheme 5. Completion of Aliskiren synthesis along retrosynthetic route B2. R1 = MeO(CH2)3ꢀ
Good conversion into product 34 was observed using [Rh(S)ꢀPhanephos(cod)]BF4 and
[Rh(dipf)(cod)]BF4 but a significant amount of the other diastereomers, as well as of byꢀproducts
was still evident by HPLC and no improvements in the dr was observed modifying the reaction
ACS Paragon Plus Environment
17

Organic Process Research & Development
Page 18 of 45
1
2
3
4
5
6
7
8
9
conditions. In order to minimize the amount of byꢀproducts and increase the stereoselectivity
through a stronger coordinative center located on the substrate, the free acid 36 was prepared
starting from tertꢀbutyl ester 28. The possibility of working with a free acid also has the
advantage to avoid the problem of ester hydrolysis in the presence of the acetate. In this case
enolacetate 35 was isolated and we could verify that it was a single diastereomer.55 On
subsequent upscaling we found out that enol acetate 35 could be prepared in 80% overall yield
from βꢀketoester 25, without isolation of the alkylated product 28. Removal of the tꢀbutyl
protection with TFA in DCM at 0 °C gave free acid 36, that was directly hydrogenated, without
any purification, using [Rh(S)ꢀPhanephos(cod)]BF4 (10 mol %) in CH2Cl2 at 60 °C and 30 bar H2
over 18 hours. Unfortunately, only moderate conversion to product 38 was observed, in contrast
to the hydrogenation of the corresponding ester 34, which was more reactive under these
conditions. However, the addition of 0.7 equiv. of NEt3 dramatically increased the rate,
permitting high conversion into 38 (99% yields) with excellent diastereoselectivity (99:1:0:0; d.r.
= 98%). Further extensive optimization during scaleꢀup allowed decreasing the catalyst loading
to 0.05% mol working on 1.00 Kg of 36. On this scale, complete conversion was observed in 24
hours (TON = 2115, TOF = 88 hꢀ1) with no erosion in the diastereoselectivity.
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
Acid 38 has all the characteristics for a correct application of the Curtius rearrangement
without lactonization risk: a free CO2H and the contiguous protected OH. Thus, 38 was reacted
with DPPA in the presence of tꢀBuOH or benzyl alcohol. Only in the second case the reaction
proceeded giving the NHCbz derivative 39 in good yields. Acidic removal of the acetate resulted
in spontaneous lactonization to compound 40. This was the first crystalline intermediate of the
synthetic sequence following the alkylated βꢀketo ester 25, and its structure was determined by
Xꢀray crystallography, confirming the correct stereochemistry postulated after the hydrogenation
ACS Paragon Plus Environment
18

Page 19 of 45
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
process (see SI). Unfortunately, the Cbz group was unstable under the further aminolysis
conditions, giving rise to the corresponding mixed urea, which proved to be very difficult to
purge by crystallization. A Cbz/Boc exchange (benzyl hydrogenolysis in the presence of Boc2O)
was therefore carried out to give known NꢀBoc lactone 41.56 Aminolysis with 3ꢀaminoꢀ2,2ꢀ
dimethylpropanamide in the presence of etylhexanoic acid as the catalyst was greatly improved
with respect to the prior art56 shifting the equilibrium towards the product by crystallizing it
while it is formed,57 by adding heptane as solvent in aminolysis reaction in Scheme 5, providing
42 in 92 % yield and with the same [α]D value (− 19.45, c 1.45 in MeOH) described in the
literature for NꢀBoc Aliskiren.30,56 Finally, Boc removal and crystallization with fumaric acid
gave the active principle 1 as the hemifumarate (m.p. = 110ꢀ112 °C).
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
The synthesis of Aliskiren presented herein was designed around two conditions: 1) the
chemistry needed to be scalable in common multiꢀpurpose plants and 2) must not infringe
product and process patents from other parties. Our synthesis was based on the unprecedented
C5ꢀC6 disconnection, where C5 acts as the nucleophilic counterpart, and was carried out on a
multigram scale in 9 steps58 and 20% overall yield starting from enantiomerically pure known
compound 20. Our synthesis is intended for production of a generic API, and the route
capitalized on known intermediates whenever intellectual property issues allowed, and where it
was convenient from a process chemistry or regulatory perspective. The key points of the
synthesis are: the use of the CO2H ꢀ NH2 synthetic equivalence, that permitted the exploitation of
the typical ketoester enolate reactivity in the formation of CꢀC bonds under mild conditions; the
Cꢀselective alkylation of the βꢀketo ester 25; and the highly enantioselective catalytic
hydrogenation of enol acetate 36, that allowed the full control of relative and absolute
ACS Paragon Plus Environment
19

Organic Process Research & Development
Page 20 of 45
1
2
3
4
5
6
7
8
9
stereochemistry of the internal 1,2ꢀsyn amino alcohol. In the final industrial synthesis, the
absolute configuration of two out of four stereogenic centers was controlled by catalytic
methods, whereas for the other two we used chiral auxiliaries. To improve the efficiency of the
process for building block 16, we had to move from a highly enantioselective, but not efficient
enough at the industrial scale, enzymatic resolution to a more robust chiral auxiliary based
preparation. However, as for 20, the chiral auxiliary was recovered in high yield. However, if we
consider the enzymatic kinetic resolution and the fact that catalytic enantioselective synthesis of
esters like 20 have been recently published,59ꢀ61 a fully catalytic synthesis is in principle possible.
Notwithstanding the several brilliant previously reported syntheses of Aliskiren, we believe that
the here described methodology compares well in terms of operational simplicity, avoidance of
harsh conditions or toxic reagents, greenꢀmetric values, and overall cost. This process was indeed
scaledꢀup in our development laboratories for the production of this active pharmaceutical
ingredient in several Kg batches, with an analytical profile adequate for use in a drug product.
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
General methods. All reagents were used as purchased from commercial suppliers without
further purification. The reactions were carried out in oven dried or flamed vessels and
performed under nitrogen. Solvents were dried and purified by conventional methods prior use.62
Flash column chromatography was performed with silica gel 60, 0.040ꢀ0.063 mm (230ꢀ400
mesh). Al backed plates preꢀcoated with silica gel 60 (UV254) were used for thin layer
chromatography and were visualized by staining with KMnO4. NMR spectra were recorded
under conditions that are specified for each spectrum (temperature 25 °C unless specified).
Splitting patterns are designated as s, singlet; d, doublet; t, triplet; q, quartet; m, multiplet; br,
ACS Paragon Plus Environment
20

Page 21 of 45
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
broad. Chemical shifts (δ) are given in ppm relative to the resonance of their respective residual
solvent peak, CHCl3 (7.27 ppm, 1H; 77.16 ppm, the middle peak, 13C). High and low resolution
mass spectroscopy analyses were recorded at 70 eV by electrospray ionization using a triple
quadrupole mass spectrometer. Specific rotations were measured with a 10 cm cell with a Na 589
nm filter: values are given in 10ꢀ1 deg.cm3.gꢀ1. Details of the molecular modeling are reported in
the SI.
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
General remark: flash chromatography was used during the route scouting studies for the
purpose of obtaining analytically pure materials for spectroscopic characterization.
In the scaled up synthesis the final building blocks (13) and (25) as well as all the intermediates
described in Scheme 5 were used as such, without any chromatographic purification.
Butyl 2ꢀ(4ꢀmethoxyꢀ3ꢀ(3ꢀmethoxypropoxy)benzyl)ꢀ3ꢀmethylbutanoate (15a). General procedure.
A solution of 2ꢀ(4ꢀmethoxyꢀ3ꢀ(3ꢀmethoxypropoxy)benzyl)ꢀ3ꢀmethylbutanoic acid (11.20 g,
36.08 mmol) in dry CH2Cl2 (120 ml) and DMF (34 µl, 439 µmol) is cooled to 0°C. Then oxalyl
chloride is added dropwise (4.3 ml, 50.5 mmol). The reaction is stirred at room temperature until
complete (usually 1.5 h). The solvent is evaporated under reduced pressure and the crude taken
up twice with CH2Cl2 and concentrated again. The acyl chloride is dissolved in 1ꢀbutanol (120
ml) and stirred overnight at room temperature. After concentration under reduced pressure, the
crude is partitioned between toluene and 5% aqueous NaHCO3 and extracted with toluene. The
organic layer is washed with brine, dried over Na2SO4 and concentrated to give 15 as an oil (9.24
g, 70%). 1H NMR (400 MHz, CDCl3): δ 6.77ꢀ6.67 (m, 3H), 4.08 (t, J = 6.5,2H), 3.94 (t, J = 6.6,
2H), 3.82 (s, 3H), 3.57 (t, J = 6.2, 2H), 3.35 (s, 3H), 2.78 (dꢀlike, J = 7.8, 2H), 2.42 (q, J =
7.5,1H), 2.09 (quintuplet, J = 6.3, 2H), 1.92 (octuplet, J = 6.9,1H), 1.41ꢀ1.51 (m, 2H), 1.22 (m,
13
2H), 1.02 (d, J = 6.9, 3H), 0.97 (d, J = 6.6, 3H), 0.85 (t, J = 7.2, 3H). C NMR (100 Mhz,
ACS Paragon Plus Environment
21

Organic Process Research & Development
Page 22 of 45
1
2
3
4
CDCl3):
δ
175.0, 148.1, 147.7, 132.5, 120.9, 113.9, 111.6, 69.3, 65.9, 63.7, 58.6, 55.9, 54.9,
5
6
7
8
35.5, 30.7, 30.6, 29.5, 20.4, 20.2, 19.0, 13.6. HRMS (EI) calcd for C21H34O5Na [M+Na]+
389.2304. Found 389.2301.
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
Methyl 2ꢀ(4ꢀmethoxyꢀ3ꢀ(3ꢀmethoxypropoxy)benzyl)ꢀ3ꢀmethylbutanoate (15b). (10.77 g, 92%
yield). 1H NMR (400 MHz, CDCl3, δ 6.78ꢀ6.668 (m, 3H), 4.08 (t, J = 6.6,2H), 3.82 (s, 3H), 3.57
(t, J = 6.2, 2H), 3.54 (s, 3H), 3.56 (s, 3H), 2.81 and 2.77 (AB part of an ABX system, JAB = 13.6,
JAX = 7.9, JBX = 0.8,2H), 2.44 (dt, Jd = 9.0, Jt = 6.8, 1H), 2.09 (m, 2H), 1.92 (m, 1H), 1.02 (d, J =
13
6.9, 3H), 0.96 (d, J = 6.6, 3H). C NMR (100 MHz, CDCl3):
δ, 175.4, 148.2, 147.8,. 132.5,
120.9, 114.0, 111.7, 69.4, 66.0, 58.6, 56.0, 54.9, 51.1, 35.5, 29.6, 20.5, 20.1. HRMS (EI) calcd
for C18H28O5Na [M+Na]+ 347.1835. Found 347.1832.
(2R)ꢀ2ꢀ{[4ꢀMethoxyꢀ3ꢀ(3ꢀmethoxypropoxy)phenyl]methyl}ꢀ3ꢀmethylbutyric
acid
(R)ꢀ16.
Racemic 15a (15.24 g, 41.58 mmol) is dispersed in 0.3 M TRIS buffer (500 ml, 83 mmol/ml)
and vigorously stirred for 5 min. Then HLAP (10.40 g, 250 mg/mmol respect to 15a, (enzyme
purchased from Sigma L9627) is added and the mixture stirred at 40 °C for 80 h. An aqueous
solution of citric acid (0.5 M) is added to adjust pH to 3ꢀ4 and the aqueous phase saturated with
solid NaCl. The crude is filtered over a Celite pad, washed wit Et2O/MeOH 9:1, brine and again
with Et2O/MeOH 9:1. This biphasic system is poured into a separatory funnel and 3 additional
extractions with Et2O (200 mL each) are performed. After drying on Na2SO4 and solvent removal
a red oil is obtained, which was purified by chromatography (AcOEt) to give known (R)ꢀ16 as a
white solid. The process gives (S)ꢀ15a (9.43 g, 62%) and (R)ꢀ30 (2.96 g, 23%) with 35%
conversion and 52.7% and 97.8% ee respectively; [α]D + 41.2 (c 1, EtOH). Lit63 [α]D +42.1 (c 1,
EtOH). The spectroscopic properties are analogues to those reported in the literature.64
ACS Paragon Plus Environment
22

Page 23 of 45
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
General procedure for HPLC analysis. A solution from enzymatic resolution (typically 10 mg,
32.2 mol) in CH2Cl2 (1 ml) is treated dropwise, at 0 °C, with a CH2N2 solution (0.25 M in
Et2O) until the yellow color persists. After 30 min the excess of CH2N2 is quenched with few
drops of AcOH. The solution is concentrated in vacuum and the last traces of AcOH
azeotropically removed with heptane. The sample is dissolved in MeOH ad submitted to HPLC
analysis.
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
Regeneration of ester 15a A solution of (S)ꢀ15a(12.62 g, 37.29 mmol) in ethanol (95%, 50 ml)
was treated with NaOH (4.47 g, 111.9 mmol) previously dissolved in water (8 mL) and refluxed
overnight. After concentration under reduced pressure the residue was partitioned between water
and toluene and extracted twice. The aqueous phase was acidified with 1 M HCl (to pH 2) and
saturated with solid NaCl. Then the acid was extracted with toluene (or AcOEt). After drying on
Na2SO4 and solvent removal a red oil was obtained, which was purified by chromatography
(AcOEt) to give known racemic 2ꢀ(4ꢀmethoxyꢀ3ꢀ(3ꢀmethoxypropoxy)benzyl)ꢀ3ꢀmethylbutanoic
acid as a white solid (10.42 g). Rf 0.32 (hexane/AcOEt 8:2 + 3% AcOH,) The acid was further
transformed into the butyl ester 15a following the general procedure described above.
(R)ꢀ2ꢀ(4ꢀMethoxyꢀ3ꢀ(3ꢀmethoxypropoxy)benzyl)ꢀ3ꢀmethylbutanal (11). To a solution of (R)ꢀ2ꢀ(4ꢀ
methoxyꢀ3ꢀ(3ꢀmethoxypropoxy)benzyl)ꢀ3ꢀmethylbutanoic acid (R)ꢀ16 (2 g, 6.44 mmol) in THF
(20 mL), under nitrogen at 0 °C, LiAlH4 (0.488 g, 12.88 mmol) is added. The reaction mixture is
stirred vigorously at r.t. for 3 hours, then at 0 °C water is added. The salts are filtered through a
pad of Celite and the organic phase concentrated under reduced pressure to give the desired
alcohol (1.8 g, 5.63 mmol, 87% yield). The crude product is dissolved in CH2Cl2 (5 mL) and
aqueous solutions of NaHCO3 0.5M (2.5 mL) and K2CO3 0.5M (2.5 mL) added. Next, TEMPO
(0.016 g, 0.104 mmol), TBACl (0.029 g, 0.104 mmol) and NCS (0.2 g, 1.49 mmmol) are added.
ACS Paragon Plus Environment
23

Organic Process Research & Development
Page 24 of 45
1
2
3
4
5
6
7
8
9
The reddish mixture is stirred vigorously at rt for three hours, then the mixture is diluted with
dichloromethane (20 mL) and water (20 mL). The phases are separated and the organic phase
washed with brine, dried over sodium sulfate and concentrated to give the desired compound 11
(0.279 g, 91% yield) as a yellow oil. An analytical sample was obtained by column
chromatography on silica gel (eluent hexane: AcOEt 3: 4). The spectroscopic properties are
analogues to those reported in the literature.26 HRMS (EI) calcd for C17H27O4 [M+H]+ 295.1909,
found 295.1910.
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)ꢀ4ꢀ(2ꢀ(Iodomethyl)ꢀ3ꢀmethylbutyl)ꢀ1ꢀmethoxyꢀ2ꢀ(3ꢀmethoxypropoxy)benzene (13) To
a
solution of (R)ꢀ16(3.500 g, 11.27 mmol) in THF (35 mL), under nitrogen at 0 °C, LiAlH4
(0.854g, 22.50 mmol) was added. The reaction mixture was stirred vigorously at r.t. for 3 h,
water (0.85 mL) was added at 0 °C, followed by 15% NaOH (0.85 g) and water (2.55 mL), the
aluminum salts were filtered through a pad of Celite and the filtrate was concentrated under
reduced pressure. The residue was dissolved in toluene (15 mL), Et3N (1.64 mL, 11.77 mmol)
was added and, after cooling to 0 °C under nitrogen, mesyl chloride (0.87 mL, 11.2 mmol) was
added dropwise. The resulting solution was stirred for 1.5 h at 20 °C, then water was added (10
mL), the phases were separated, the organic phase was washed with brine, dried over Na2SO4
and concentrated to residue. The mesylate was taken up in acetonitrile (20 mL), NaI was added
(4.50 g, 30.00 mmol) and the solution was heated under reflux and under nitrogen for 12 h. Then
the solvent was evaporated and the residue was taken up in toluene (25 mL), the organic phase
was washed twice with water, dried over Na2SO4 and concentrated to residue. The product was
obtained as a white solid that did not require further purification (4.05 g, 9.97 mmol, 88% yield).
The spectroscopic properties are analogues to those reported in the literature.34 HRMS (EI) calcd
for C17H27IO3Na [M+Na]+ 429.0903, found 429.0905
ACS Paragon Plus Environment
24

Page 25 of 45
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
(S)ꢀtertꢀButyl 3ꢀ((S)ꢀ4ꢀbenzylꢀ2ꢀoxooxazolidineꢀ3ꢀcarbonyl)ꢀ4ꢀmethylpentanoate (20) A solution
of hexyllithium 2.3 M in hexane (18.3 mL, 42.1 mmol) is added at 0 °C and under nitrogen to a
solution of diisopropylamine (5.19 mL, 42.1 mmol) in anhydrous THF (10 mL). After 15
minutes the solution is cooled to ꢀ78 °C and a solution of (S)ꢀ4ꢀbenzylꢀ3ꢀ(3ꢀ
methylbutanoyl)oxazolidinꢀ2ꢀone (S)ꢀ18 (10 g, 38.3 mmol), in THF (5 mL) is added dropwise.
After 45 minutes at ꢀ78 °C, tꢀbutyl bromoacetate (10.7 mL, 72.7 mmol) is added. The
temperature is raised to 20 °C during 4 hours, then a saturated solution of ammonium chloride
(50 ml) is added, the THF evaporated and the suspension thus obtained is extracted twice with
ethyl acetate (2 x 100 mL). The combined organic phases are washed with 0.5 N HCl, brine and
dried over anhydrous sodium sulfate. After evaporating the solvent under reduced pressure, the
residue was purified by flash column chromatography (cyclohexane/ethyl acetate 8:2) to give 20
(11.5 g, 90% yield) as a white solid. [a]D25 = + 69.0 (c = 1 in CDCl3). Literature65 [a]D25 = + 69.9
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
1
(c= 1 in CH2Cl2). H NMR (300 MHz, CDCl3,) δ 7.37ꢀ7.20 (m, 5H), 4.70ꢀ4.60 (m, 1H), 4.20ꢀ
4.08 (m, 3H), 3.35ꢀ3.25 (dd, J = 13.5 Hz, 3.1 Hz,1H), 2.85ꢀ2.75 (dd, J = 28.9, 10.1 Hz, 1H),
2.75ꢀ2.65 (dd, J = 13.8, 10.11 Hz, 1H), 2.47ꢀ2.37 (dd, J = 16.8 Hz, 3.1 Hz, 1H), 2.01ꢀ1.91 (m,
1H), 1.45ꢀ1.35 (s, 9H),1.05ꢀ0.95 (d, J = 6.7 Hz, 3H), 0.95ꢀ0.85 (d, J = 6.7 Hz, 3H). 13C NMR (75
MHz, CDCl3) δ 176.9, 174.5, 172.2, 137.3, 130.7, 130.5, 128.5, 81.1, 66.9, 57.1, 45.4, 38.5,
34.5, 31.1, 29.3, 21.7, 19.5. HRMS (EI) calcd for C21H29NO5Na [M+Na]+ 398.1944, found
398.1948.
(S)ꢀEthyl 5ꢀ((S)ꢀ4ꢀbenzylꢀ2ꢀoxooxazolidineꢀ3ꢀcarbonyl)ꢀ6ꢀmethylꢀ3ꢀoxoheptanoate (21). A 100ꢀ
mL flask is charged with potassium ethyl malonate (7.99 g, 46.9 mmol), anhydrous THF (20
mL) and anhydrous magnesium chloride (2.24 g, 23.5 mmol). The suspension is heated under
ACS Paragon Plus Environment
25

Organic Process Research & Development
Page 26 of 45
1
2
3
4
5
6
7
8
9
reflux under nitrogen for 12 hours and then cooled to 20ꢀ22 °C. Compound 20 (5.87 g, 15.65
mmol) is dissolved into CH2Cl2 (10 mL) and the solution cooled to 0 °C. TFA (4.56 g, 40 mmol)
is slowly added and the solution stirred at 0°C for 12 h. The solvent is evaporated and the crude
dissolved in anhydrous THF (15 mL). At 0 °C, carbonyldiimidazole (2.79 g, 17.2 mmol) is
added to this solution and the mixture stirred at 20 °C for 2 hours. After cooling again to 0 °C,
this solution is added to the suspension containing the magnesium ethyl malonate. The resultant
mixture is stirred at 20ꢀ22 °C for 3 hours, then 10% w/w aqueous solution of HCl (about 80 mL)
is added to pH = 2. THF is evaporated at reduced pressure and the mixture extracted with toluene
(80 mL); the organic phase is washed twice with a saturated solution of NaHCO3 (80 mL), dried
over anhydrous sodium sulfate and concentrated. The product is obtained in the form of viscous
oil that does not require further purification (5.5 g, 14.1 mmol, 90% yield). An analytical sample
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
was obtained by column chromatography on silica gel (eluent hexane: AcOEt 2: 1). H NMR
(300 MHz, CDCl3) δ 7.30ꢀ7.10 (m, 5H), 4.62ꢀ4.52 (m, 1H), 4.15ꢀ4.00 (m, 5H), 3.48ꢀ3.32 (q, J =
13.20 Hz, 2H), 3.22ꢀ3.12 (dd, J = 13.2, 2.8 Hz ,1H), 3.12ꢀ3.02 (dd, J = 18.4, 11.6 Hz, 1H), 2.75ꢀ
2.65 (m, 2H), 1.97ꢀ1.87 (m, 1H), 1.22ꢀ1.15 (t, J = 6.7 Hz, 3H), 0.95ꢀ0.90 (d, J = 6.7 Hz, 3H),
13
0.83ꢀ0.78 (d, J = 6.7 Hz, 3H). C NMR (75 MHz, CDCl3) δ 202.6, 176.5, 168.2, 154.2, 137.2,
130.3, 128.5, 67.1, 57.0, 50.8, 44.9, 42.9, 38.8, 30.9, 22.1, 21.2, 15.3 HRMS (EI) calcd for
C21H27NO6Na [M+Na]+ 412.1736, found 412.1737.
(S)ꢀ1ꢀBenzyl 4ꢀtertꢀbutyl 2ꢀisopropylsuccinate (23) To a solution of 20 (25.0 g, 66.6 mmol) in
anhydrous THF (100 mL), at 0 °C, a solution of benzyl alcohol (9.36 g 86.6 mmol) in anhydrous
THF (40 mL) is added. The mixture is cooled to –10 °C and a 2.3 M solution of hexyllithium in
hexane (31.8 mL, 73.2 mmol) added dropwise. The resultant solution was stirred for 6 hours at 0
°C, then water (250 mL) and AcOEt (150 mL) added. The phases are separated, the organic
ACS Paragon Plus Environment
26

Page 27 of 45
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
phase washed with water, dried over sodium sulfate and concentrated under reduced pressure.
The residue is taken up in AcOEt (40 mL) and hexane (200 mL); the mixture is stirred for 30
minutes and the precipitated Evan’s auxiliary filtered (recovery 9.9 gr, 85% yield). The filtrate
solution is concentrated to residue and purified by flash column chromatography (eluent
hexane/AcOEt 7:3) to give the desired compound 23 (16.0 g, 52.25 mmol, 79% yield) as 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
1
colorless oil. HꢀNMR 300 MHz (CDCl3) δ (ppm): 7.35ꢀ7.23 (m, 5H); 5.09 ꢀ 5.14 (system AB,
2H); 2.79ꢀ2.60 (m, 2H); 2.38ꢀ2.28 (m, 1H); 1.97 (m, 1H); 1.40 (s, 9H); 0.90 (d, J = 7.4 Hz, 3H);
0.87 (d, J = 7.4 Hz, 3H). 13CꢀNMR (100 MHz, CDCl3) 173.8, 171.1, 135.6, 128.0, 127.8, 127.6,
80.1, 65.7, 47.2, 33.9, 29.7, 27.5, 19.6, 19.1. HRMS (EI) calcd for C18H26O4Na [M+Na]+
329.1729, found 329.1727.
(S)ꢀ1ꢀBenzylꢀ6ꢀethyl 2ꢀisopropylꢀ4ꢀoxohexanedioate (24) (S)ꢀ1ꢀBenzyl 4ꢀtertꢀbutyl 2ꢀ
isopropylsuccinate 23 (20.0 g, 65.3 mmol) is dissolved in 100 mL of a 1:2 v/v solution of TFA
and CH2Cl2 and the mixture stirred at room temperature overnight. The reaction mixture is
evaporated under reduced pressure to give the acid as a yellow oil. The acid is dissolved in
anhydrous THF and carbonyldiimidazole (10.50 g, 65.3 mmol) added at 0 °C. The solution is
added to a suspension of magnesium ethyl malonate, prepared from potassium ethyl malonate
(28.0 g, 165 mmol) and MgCl2 (7.85 g, 82.5 mmol) in THF stirred at 20 °C. The mixture is
refluxed 12 h, then stirred at 20 °C for other 3 h. A 10% w/w aqueous solution of HCl is then
added up to pH = 2 (80 mL). THF is evaporated under reduced pressure and the mixture
extracted with toluene (80 mL); the organic phase is washed twice with a saturated solution of
NaHCO3 (80 mL), dried over Na2SO4 and concentrated. The product is obtained as a viscous oil
that does not require further purification (15.04 g, 72% yield). An analytical sample is obtained
1
by flash column chromatography (heptane/AcOEt 1:2). HꢀNMR (400 MHz, CDCl3)
δ 7.34 (m,
ACS Paragon Plus Environment
27

Organic Process Research & Development
Page 28 of 45
1
2
3
4
5H), 5.15ꢀ5.05 (AB system, J = 12.2 Hz, 2H), 4.15 (q, J = 7.0 Hz, 2H), 3.52ꢀ3.37 (AB system, J
5
6
= 15.3 Hz, 2H), 3.02ꢀ2.86 (m, 2H), 2.59 (dd, J = 17.4, 3.0, 1H), 2.03 (m, 1H), 1.26 (t, J = 7.0,
7
8
9
13
3H), 0.90 (d, J = 7.4 Hz, 3H), 0.87 (d, J = 7.4 Hz, 3H). CꢀNMR (100 MHz, CDCl3)
δ 201.0,
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
176.2, 169.0, 137.8, 129.5, 128.4, 67.2, 59.4, 48.2, 39.8, 30.3, 20.6, 16.0. HRMS calcd for
C18H24O5Na [M+Na]+ 343.1522; found 343.1518.
(S)ꢀ1ꢀBenzylꢀ6ꢀtꢀbutyl 2ꢀisopropylꢀ4ꢀoxohexanedioate (25). This product is prepared following
the procedure already described for 24. The product is obtained as a viscous oil that does not
require further purification (18.3 g, 82% yield). An analytical sample is obtained by flash column
1
chromatography (heptane/AcOEt 1:1). H NMR (400 MHz, CDCl3) δ 7.30 (m, 5H), 5.08 (m,
2H), 3.40 – 3.25 (m, 2H), 2.95 (m, 2H), 2.60 (dd, J = 17.8, 3.3 Hz, 1H), 1.99 (dp, J = 13.1, 6.6
13
Hz, 1H), 1.44 (s, 9H), 0.87 (dd, J = 12.0, 6.9 Hz, 6H). CꢀNMR (100 MHz, CDCl3) 201.7,
172.2, 166.2, 136.6, 128.5, 127.9, 127.6, 82.1, 66.5, 51.3, 46.5, 41.1, 29.9, 28.4, 20.1, 19.3.
HRMS calcd for C20H28O5Na [M+Na]+ 371.1835. Found 371.1833.
(S)ꢀEthylꢀ5ꢀ(((S)ꢀ4ꢀbenzyloxazolidinꢀ2ꢀonꢀ3ꢀyl)carbonyl)ꢀ2ꢀ((R)ꢀ2ꢀ(4ꢀmethoxyꢀ3ꢀ(3ꢀ
methoxypropoxy)benzyl)ꢀ3ꢀmethylbutylidene)ꢀ6ꢀmethylꢀ3ꢀoxoheptanoate, E/Z mixture (22a). Dry
titanium tetrachloride (18.40 g, 9.7 mmol) is dissolved in dichloromethane (24 mL) and to this
solution, cooled to 0 °C, a solution containing aldehyde 11 (14.3 g, 48.5 mmol) and keto ester 21
(18.9 g, 48.5 mmol) in THF (67 mL) is added. A solution of pyridine (15.3 g, 194 mmol) in THF
(26 mL) is added, during 2 h, to the resultant solution, under vigorous stirring. In the further 14
hours of stirring, the reaction mixture is allowed to reach the temperature of 15 °C. Then it is
cooled again to 0 °C and water and methylꢀtꢀbutyl ether (200 mL) are added. The phases are
separated, the aqueous phase is extracted with methylꢀtꢀbutyl ether, the combined organic phases
are washed with brine, dried over sodium sulfate and concentrated under reduced pressure. The
ACS Paragon Plus Environment
28

Page 29 of 45
Organic Process Research & Development
1
2
3
4
residue is purified by flash column chromatography (petroleum ether/ethyl acetate 7:3) to give
5
6
1
the desired compound 22a (15.0 g, 46% yield) as a yellow oil. HꢀNMR (400 MHz, CDCl3) δ
7
8
9
7.28ꢀ7.19 (m, 5H), 6.79ꢀ6.67 (m, 3H), 6.59 (d, J = 16.8 Hz, 1H), 4.64ꢀ4.62 (m, 1H), 4.18ꢀ3.99
(m, 6H), 3.72 (s, 3H), 3.49 (t, J = 6 Hz, 2H); 3.27 (s, 3H); 3.1ꢀ3.0 (m, 2H); 2.77ꢀ2.68 (m, 2H);
2.55ꢀ2.30 (m, 2H); 2.02ꢀ1.97 (m, 2H); 1.88ꢀ1.77 (m, 1H); 1.72ꢀ1.55 (m, 1H); 1.27ꢀ1.17 (m, 4H);
0.96ꢀ0.77 (m, 13H). HRMS (EI) calcd for C38H51NO9Na [M+Na]+ 688.3462, found 688.3463.
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)ꢀEthyl 2ꢀ((R)ꢀ2ꢀ(4ꢀmethoxyꢀ3ꢀ(3ꢀmethoxypropoxy)benzyl)ꢀ3ꢀmethylbutyldene)ꢀ6ꢀmethylꢀ3ꢀoxoꢀ
5ꢀ(piperidineꢀ1ꢀcarbonyl)heptanoate (E/Z mixture) (22b). To a solution of (R)ꢀ2ꢀ(4ꢀmethoxyꢀ3ꢀ
(3ꢀmethoxypropoxy)benzyl)ꢀ3ꢀmethylbutanal 11 (0.304 g, 1.03 mmol) in DMF (3 mL), (S)ꢀ
ethylꢀ5ꢀ(((S)ꢀ4ꢀbenzyloxazolidinꢀ2ꢀonꢀ3ꢀyl)carbonyl)ꢀ6ꢀmethylꢀ3ꢀoxoheptanoate 21 (0.684 g,
1.758 mmol) and piperidine (0.509 mL, 5.15 mmol) are added. The resulting solution is stirred at
60 °C for two hours, then Et2O is added, the organic phase washed with water, dried over sodium
sulfate and concentrated under reduced pressure. The residue is purified by flash column
chromatography (petroleum ether/ethyl acetate 8:2) to give the desired compound 22b (0.413 g,
70% yield) as a pale yellow oil. 1HꢀNMR (400 MHz, CDCl3) δ 6.74ꢀ6.62 (m, 3H); 4.19ꢀ4.02 (m,
5H); 3.78 (s, 3H); 3.55ꢀ3.48 (m, 8H); 3.32 (s, 3H); 3.17ꢀ3.0 (m, 3H); 2.77ꢀ2.52 (m, 4H); 2.19ꢀ
2.03 (m, 3H); 1.84ꢀ1.42 (m, 12H); 1.27ꢀ1.18 (m, 6H); 0.94ꢀ0.78 (m, 15H). HRMS (EI) calcd for
C33H51NO7Na [M+Na]+ 596.3563, found 596.3562.
(2R,S)(5S)ꢀ6ꢀBenzyl 1ꢀethyl 5ꢀisopropylꢀ2ꢀ((R)ꢀ2ꢀ(4ꢀmethoxyꢀ3ꢀ(3ꢀmethoxypropoxy)ꢀbenzyl)ꢀ3ꢀ
methylbutyl)ꢀ3ꢀoxohexanedioate (27). Keto ester 24 (0.910 g, 2.841 mmol) is dissolved in dry
dimethylacetamide (1.8 mL) and LiH (21.6 mg, 2.717 mmol) added. The mixture is heated to 60
°C for 1 hour, then iodide 13 (1.050 g, 2.584 mmol)28 added and the mixture stirred at 60 °C for
48 hours, then diluted with toluene (30 mL) and washed with water. The organic phase is
ACS Paragon Plus Environment
29