Process Development and Good Manufacturing Practice Production of a Tyrosinase Inhibitor via Titanium-Mediated Coupling between Unprotected Resorcinols and Ketones

Article abstract of DOI:10.1021/acs.oprd.7b00036

A concise and economically attractive process for the synthesis of a novel tyrosinase inhibitor has been developed and implemented on a multikilogram scale under GMP. A major achievement to the success of the process is the development of a direct coupling between free resorcinol and ketone. First developed under basic conditions, this coupling has been turned to a novel titanium(IV) mediated process allowing good selectivity, easy isolation, and high atom efficiency. Other key steps feature an alkene reduction by palladium catalyzed transfer hydrogenation and a urea formation using N,N′-disuccinimidyl carbonate as the carbonyl source. This route allowed us to produce kilogram batches of the candidate to support preclinical and clinical studies.

Full text of DOI:10.1021/acs.oprd.7b00036

Subscriber access provided by UNIV OF CALIFORNIA SAN DIEGO LIBRARIES
Full Paper
Process Development and GMP Production of a Tyrosinase Inhibitor via
Titanium Mediated Coupling Between Unprotected Resorcinols and Ketones
Thibaud Gerfaud, Cedric Martin, Karinne Bouquet, Sandrine Talano, Corinne
Millois-Barbuis, Branislav Musicki, Jean-Guy Boiteau, and Isabelle Cardinaud
Org. Process Res. Dev., Just Accepted Manuscript • DOI: 10.1021/acs.oprd.7b00036 • Publication Date (Web): 20 Mar 2017
Downloaded from http://pubs.acs.org on March 20, 2017
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Organic Process Research & Development is published by the American Chemical
Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 30
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Process Development and GMP Production of a
Tyrosinase Inhibitor via Titanium Mediated
Coupling Between Unprotected Resorcinols and
Ketones
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
Thibaud Gerfaud*, Cédric Martin, Karinne Bouquet, Sandrine Talano, Corinne MilloisꢀBarbuis,
Branislav Musicki, JeanꢀGuy Boiteau* and Isabelle Cardinaud
Nestlé Skin Health R&D, 2400 Route des colles BP 87, 06902 SophiaꢀAntipolis Cedex, France
ACS Paragon Plus Environment
1

Organic Process Research & Development
Page 2 of 30
1
2
3
4
5
6
7
TABLE OF CONTENTS:
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment
2

Page 3 of 30
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
ABSTRACT: A concise and economically attractive process for the synthesis of a novel
tyrosinase inhibitor has been developed and implemented on a multiꢀkilogram scale under GMP.
Major achievement to the success of the process is the development of a direct coupling between
free resorcinol and ketone. First developed under basic conditions, this coupling has been turned
to a novel titanium (IV) mediated process allowing good selectivity, easy isolation and high atom
efficiency. Other key steps feature an alkene reduction by palladium catalyzed transfer
hydrogenation and a urea formation using N,N’ꢀDisuccinimidyl carbonate as the carbonyl source.
This route allowed us to produce kilogram batches of the candidate to support preclinical and
clinical studies.
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: Tyrosinase Inhibitors, Resorcinol, Titanium mediated coupling
INTRODUCTION: Hyperpigmentation disorders such as melasma are characterized by an
increase in melanin synthesis which accumulates in the epidermis and is responsible for a
darkening of the skin.1 Melanogenesis occurs in the basal layer of the epidermis into specific
organelles of the melanocytes called melanosomes.2 A detailed analysis of the biosynthetic
pathway reveals that tyrosinase is a key enzyme in melanogenesis3 and is responsible for the
oxidation of tyrosine into DOPA (3,4ꢀdihydroxyphenylalanine) and DOPAquinone.4 1 (Figure 1)
is a melanogenesis inhibitor working through inhibition of tyrosinase (IC50 = 0.1 µM on normal
ACS Paragon Plus Environment
3

Organic Process Research & Development
Page 4 of 30
1
2
3
4
5
6
human epidermal melanocytes) currently under development at Nestlé Skin Health R&D for the
topical treatment of hyperpigmentation disorders.5
7
8
9
Figure 1. Structure of Tyrosinase Inhibitor Clinical Candidate 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
Clinical candidate 1 is a small molecule (MW = 340.42 g.molꢀ1) designed against the following
parameters: activity, skin penetration properties, topical formulability, low systemic exposure
and high safety. Key structural features are a resorcinol moeity functionalized at position 4, a
urea and one chiral center.
Discovery chemistry synthesis6 (Scheme 1) started from 4ꢀbromoresorcinol, an expensive raw
material, difficult to source on large scale. Double benzyl protection followed by lithiumꢀhalogen
exchange under cryogenic conditions (ꢀ70 °C) and consecutive addition on 1ꢀcarbethoxyꢀ4ꢀ
piperidone yielded the tertiary alcohol 3 (62% yield after recrystallization). Hydrogenation of 3
led to the reduced and deprotected intermediate 4: reꢀprotection of the 2 phenol moieties was
necessary to avoid further side reactions and was accomplished with benzyl bromide to give 5.
Carbamate hydrolysis, followed by condensation with (S)ꢀ(ꢀ)ꢀαꢀmethylbenzyl isocyanate and
final benzyl deprotection gave 1 in 16% overall yield and 7 chemical steps.
Scheme 1. Discovery Chemistry Route to 1
ACS Paragon Plus Environment
4

Page 5 of 30
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
While well suited for library synthesis with an advanced intermediate 6 that could be easily
functionalized at the N position, this synthesis required modifications to be amenable to large
scale GMP production. Not only chromatographic purifications nor cryogenic conditions had to
be avoided but material costs and atom efficiency were major issues to address for further
development. Our longꢀterm retrosynthetic strategy was thus driven mainly by the use of cheap
and easily available reagents (Scheme 2).
Scheme 2. Retrosynthetic Analysis of 1
Route selection
Given some important time constraints and relatively small amounts of drug substance
required for early toxicological studies (100 g), preliminary investigations focused on reducing
the overall number of steps and limiting protection/deprotection sequences. Doubly Oꢀ
benzylated bromoresorcinol 2 was kept as first intermediate and efforts were turned toward
coupling of the piperidine moeity and formation of the urea (Scheme 3). Chirality was set up
using cheap and readily available (S)ꢀ(ꢀ)ꢀαꢀmethylbenzylamine 8 and urea formation
ACS Paragon Plus Environment
5

Organic Process Research & Development
Page 6 of 30
1
2
3
4
5
6
accomplished by preꢀactivating 8 with N,N′ꢀcarbonyldiimidazole (CDI), avoiding the use of the
unstable isocyanate on scale.
7
8
9
Suzuki coupling on bromoresorcinol 2 was considered as an alternative to ketone 1,2ꢀaddition
and various options were investigated, most relevant ones being depicted in Scheme 3. Enol
triflate formation from NꢀBocꢀ4ꢀpiperidone 9 followed by Miyaura borylation with
bis(pinacolato)diboron (B2pin2) furnished vinylboronate 10 (57% yield on 10 g scale, isolated as
a solid). Subsequent Suzuki coupling with 2 gave 11 in good yield and good overall purity (oil).
Boc deprotection, urea formation and final double bond reduction/benzyl deprotection furnished
1 in 28% overall yield. An alternative approach using commercially available 4ꢀpyridineboronic
acid gave pyridine 14 (63% yield on 100 g scale, isolated as a solid), which was hydrogenated
using a mixture of palladium and platinum oxide catalyst with concomitant benzyl deprotection
to deliver amine 15-AcOH as the acetate salt. Interestingly urea formation with 13 proceeded
smoothly and delivered 1 in 20% overall yield with very low amounts of Oꢀacylation byproduct
observed. Other approaches involving direct Suzuki couplings on unprotected 4ꢀbromoresorcinol
were unsuccessful.7
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. Alternative Strategies Featuring a Suzuki Coupling
ACS Paragon Plus Environment
6

Page 7 of 30
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
While with these routes overall yield were improved and protection/deprotection sequences
reduced, major limitations still remained for a successful scaleꢀup campaign: prices of raw
materials (4ꢀbromoresorcinol, 4ꢀpyridineboronic acid), cryogenic temperature (enol formation),
harsh conditions (high catalyst loading, catalyst poisoning during pyridine hydrogenation)… It
was anticipated that even with strong optimization efforts, final material costs of 1 would remain
high using these synthetic pathways. The high selectivity observed when conducting the urea
formation with activated benzylamine 13 on fully unprotected amine 15 prompted us to
investigate original synthetic pathways to prepare this intermediate.
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
Resorcinol and other polyphenols are good nucleophiles toward Cꢀalkylation, especially under
FriedelꢀCraft conditions but unprotected resorcinols (free OH) remain challenging substrates
because Oꢀalkylation can be predominant with certain electrophiles (e.g. acyl chlorides). Cꢀ
alkylation can be achieved but selectivity and over alkylation issues are often encountered. We
were intrigued by the possibility to form the requisite CꢀC bond directly from resorcinol (or one
of its cheap derivatives) and preliminary investigations started with dimethoxy resorcinol which
condensed with piperidone in acetic acid and conc. HCl to yield alkene intermediate 16 (Scheme
4).8 Subsequent reduction furnished 17 in 39% yield over 2 steps and demethylation was
accomplished under harsh conditions (PyrꢀHCl, 160 °C) yielding targeted 15-HCl in 90% yield.
This approach afforded 1 in only 5 steps and 23% overall yield using inexpensive raw material.
Scheme 4. Alternative Approach with Dimethoxyresorcinol
ACS Paragon Plus Environment
7

Organic Process Research & Development
Page 8 of 30
1
2
3
4
5
6
7
8
Unfortunately, high amounts of impurities were formed during the first step and were
impossible to separate from 16 by nonꢀchromatographic methods. The very harsh conditions
required for the demethylation were another important drawback of this approach. To overcome
these issues, we decided to focus on the development of a method to couple unprotected
resorcinol with NꢀBocꢀ4ꢀpiperidone, both starting materials being cheap and readily available.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
RESULTS AND DISCUSSION:
Development of coupling conditions with unprotected resorcinol
Under acidic conditions, unprotected resorcinols are known to undergo polycondensations with
aldehydes and ketones providing resorcinarenes and polymeric species (with aldehydes)9 or
chroman derivatives (with ketones)10 (Scheme 5).
Scheme 5. Known Direct Condensation of Unprotected Resorcinol with Ketones11
With piperidone derivatives however, aldolization has been reported to be much slower
enabling a direct 1,2ꢀaddition on ketone and isolation of the corresponding alcohol.11 These
conditions were evaluated using NꢀBoc or Nꢀbenzyl piperidone leading in our case to the
formation of mixtures of alcohols and corresponding alkenes albeit always in low yields and
accompanied by significant amounts of bisꢀaddition sideꢀproducts, chroman and other impurities
ACS Paragon Plus Environment
8

Page 9 of 30
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
(Scheme 6). A rapid screening of the reaction conditions regarding to resorcinol stoichiometry,
acid source (AcOH, H2SO4, pꢀTSA), solvent and temperature did not permit to optimize yield
and ratio of targeted products 18 and 19 vs bisꢀaddition sideꢀproducts to a level sufficient for
scaleꢀup.
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 6. Products Formed During Direct Resorcinol Alkyation Under Acidic Conditions
Given the poor preliminary results obtained under acidic conditions, we decided to evaluate
basic ones, aiming that deprotonation of resorcinol would enhance its nucleophilicity toward Cꢀ
alkylation and allow it to add on NꢀBoc piperidone. Interestingly, first experiments (LiOH in
THF/water mixture) yielded the desired alcohol intermediate 18 in 35% isolated yield along with
27% of bisꢀaddition adduct 20 (HPLC A%). An optimization of these conditions was performed
against the following parameters: type of base (NaOH, MeONa, EtONa, tꢀBuOK), solvent,
resorcinol stoichiometry, temperature… but minimization of the amount of bisꢀaddition adduct
formed during the reaction and isolation of the desired product free of this impurity proved to be
a significant challenge12. The best balance was obtained using sodium ethoxide in ethanol at 23
ACS Paragon Plus Environment
9

Organic Process Research & Development
Page 10 of 30
1
2
3
4
5
6
7
8
9
°C, with 4 equiv of resorcinol and a slow addition of NꢀBoc piperidone over the deprotonated
resorcinol (Scheme 7). In this case, only 8ꢀ10% of 20 were formed. Direct isolation of 18 out of
the reaction mixture was not possible, probably because the large excess of resorcinol used
prevented its crystallization. Upon completion of the reaction, pH of the solution was adjusted to
6.3 using NaH2PO4 and the reaction mixture warmed to reflux to dehydrate the alcohol into
corresponding alkene 19. Subsequent solvent exchange from EtOH to water led to crystallization
of 19 in 40% yield contaminated by 4ꢀ5% of impurity 22. This procedure was scaled up to
deliver >300 g of 19. Unfortunately, despite an extensive screening of recrystallization
conditions, a process able to fully remove the bisꢀaddition impurity at this stage was not found.13
Best compromise proved to be a trituration of the solid in 3 vol of toluene which improved the
impurity profile, reducing the amount of 22 to less than 3%. Further scaleꢀup of this approach
was impossible since product crystallization was capricious leading alternatively to gummy
residues or solid material with variable particle size, sometimes very difficult to drain out of the
reaction vessel. Lack of robustness of the crystallization was attributed to the large excess of
resorcinol used in the process. First 100 g batches of 1 were produced using this route but bisꢀ
addition impurity was not depleted in the downstream chemistry (alkene reduction, Boc
deprotection and coupling with activated urea). Triturations and chromatographic purifications of
intermediates were required to control the level of bisꢀaddition impurity below 1% in the final
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
Scheme 7. Optimized Base Promoted Condensation of Unprotected Resorcinol with N-Boc
piperidone
ACS Paragon Plus Environment
10

Page 11 of 30
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Intrigued by the difficulty encountered in the optimization of the coupling process (regarding
resorcinol stoichiometry and amount of bisꢀaddition impurity), the product of the condensation
(pure alcohol 18) was isolated by chromatography and reꢀsubmitted to the reaction conditions
(Table 1). Surprisingly, after 12 h only 50% (HPLC A%) of 18 remained in solution along with
35% resorcinol and 14% of bisꢀaddition impurity 20. These results clearly indicated that a retroꢀ
addition occurred with cleavage of a CꢀC bond and thus that an equilibrium was reached under
the reaction conditions.
Table 1. Reversibility Observed for the Base Promoted Processa
Time (h)
18 (A%)b
100
Resorcinol (A%)b
20 (A%)b
0
0
0
24c
50
35
14
aReaction conducted with 3.2 mmol of 18 in 7 vol of EtOH and 4 equiv of EtONa (2.7 M
solution in EtOH) at 23 °C for 24 h; bConversion measured by HPLC (A%); cNꢀBocꢀ4ꢀpiperidone
formation was confirmed by TLC analysis (DCM/EtOAc 9:1).
In light of this new information, it became clear that further attempts to reduce the amount of
resorcinol and bisꢀaddition byꢀproduct would be useless. To solve these issues (reversibility,
resorcinol amount, robustness) we reꢀexamined the acidic conditions aiming to find an oxophilic
ACS Paragon Plus Environment
11

Organic Process Research & Development
Page 12 of 30
1
2
3
4
5
6
7
8
9
Lewis acid able to play multiple roles: 1) activate the ketone with a decent turnꢀover, 2) orientate
resorcinol attack to minimize bisꢀaddition adducts, 3) chelate hydroxyl group generated to avoid
retroꢀaddition reaction and 4) keep the Boc protection untouched. An extensive screening of
Lewis acids and solvents was performed14 and amongst all acids tested we were pleased to find
that only titanium alkoxides were able to promote the reaction to complete conversion with
limited amounts of sideꢀproducts under mild conditions. Various solvents were compatible with
the transformation and 2ꢀMeTHF15 was selected for its process compatibility. At 23 °C a clean
reaction profile was observed with Ti(OiPr)4, providing a promising starting point for further
optimization (Table 2, entry 1). At higher temperature, higher levels of bisꢀaddition adducts were
observed. Lowering the reaction temperature to 0 °C or the amount of resorcinol to one
equivalent had a detrimental impact on reaction’s rate or levels of bisꢀaddition impurity (Table 2,
entry 2 and 4). We postulated that the reaction proceeds first through a ligand exchange between
resorcinol and titanium (IV) alkoxide (a precipitate was observed at the beginning of the Ti
addition), followed by activation of the ketone leading to titanium complex 24 and CꢀC bond
formation as shown in Scheme 8.16 Formation of 24 was deemed to be responsible for the fast
reaction rate as it would bring the 2 reagents into close spatial proximity, leading to fast
intramolecular addition.
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 8. Proposed Mechanism for the Ti(IV)-Mediated Condensation
ACS Paragon Plus Environment
12

Page 13 of 30
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
This assumption was reinforced by the fact that running the same reaction with 3ꢀmethoxyphenol
gave only one isomer (28) with no evidence of bisꢀaddition impurity (Scheme 9).
Scheme 9. Ti(IV)-Mediated Condensation with 3-methoxyphenola
aReaction conducted on 10 mmol scale with 2 equiv of 3ꢀmethoxyphenol and 0.5 equiv of
Ti(OiPr)4 in 7 vol of solvent
We hypothesized that formation of the bisꢀaddition impurity proceeds via complexes 26 and/or
27 and that a key role is played by the ligands surrounding the Ti atom. Indeed, higher levels of
bisꢀaddition impurity 22 and longer reaction time were observed with less hindered Ti(OEt)4
(Table 2, entry 6). In this case, a thick and difficult to stir precipitate17 formed and remained in
suspension whereas with Ti(OiPr)4 the precipitate dissolved rapidly (15ꢀ20 minutes) and less
than 4 h were necessary to reach full conversion (Table 2, entry 5). With more hindered Ti(Otꢀ
Bu)4, levels of bisꢀaddition impurity could be reduced down to 3ꢀ4% with a shorter reaction time
and no precipitate was observed (Table 2, entry 7). We assume that formation of complex 23 (or
ACS Paragon Plus Environment
13

Organic Process Research & Development
Page 14 of 30
1
2
3
4
5
6
7
8
9
a structurally close polymeric species) is responsible of the observed precipitate. With small
ethoxide ligand, the Ti atom can accommodate more resorcinol ligands (x≥ 2) leading to
precipitation of 23 whereas with more hindered isopropoxide and tertꢀbutoxide ligands, complex
23 is more soluble (x = 1) and less stable, explaining the higher reactivity and the faster
dissolution of the precipitate. On the other hand, complex 27 is probably less prone to form with
bulkier ligands, explaining the lower amounts of impurity 22 observed with Ti(OtꢀBu)4.
Table 2. Conditions Screening for the Ti(IV)-Mediated Couplinga
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
entr
y
1
Ti
equiv
1.0
product
yieldc
79%
R
conditions
conversionb
100%
22b
iPr
2ꢀMeTHF, 23 °C, 1 h
6%
2
3d
4
iPr
iPr
iPr
iPr
Et
0.5
0.5
0.5
0.5
0.5
0.5
0.3
0.5
2ꢀMeTHF, 0 °C, >6 h
2ꢀMeTHF, 23 °C, 3 h
<30%
100%
100%
100%
100%
100%
100%
n.d.
n.d.
n.d.
78%
68%
85%
n.d.
n.d.
4%d
10%
13%
6%
2ꢀMeTHF, 60 °C, 45 min
2ꢀMeTHF, 23 °C, 3 h
5
6e
7
2ꢀMeTHF, 23 °C, 36 h
2ꢀMeTHF, 23 °C, 2 h
9%
tꢀBu
iPr
iPr
4%
8
2ꢀMeTHF, 23 °C, 8 h
10%
n.d.
9
2ꢀMeTHF:tꢀBuOH 1/1, 23 °C, 8 h
<
20%
b
aReactions conducted on 25 mmol scale in 7 vol of solvent; Ratio determined by HPLC
d
analysis (A%); cIsolated yield after flash chromatography; Only 1 equiv of resorcinol was used;
eA very thick precipitate remained in solution after 6 h, amount of bisꢀaddition impurity
measured after 6 h of reaction.
ACS Paragon Plus Environment
14

Page 15 of 30
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Capitalizing on what we had learned in the optimization of the basic route, we were able to
quickly identify a process offering the best balance between all the parameters. While Ti(OtꢀBu)4
gave better yields and lower amounts of bisꢀaddition impurity, we decided to focus on Ti(OiPr)4,
economically more attractive.18 Use of tꢀBuOH as a coꢀsolvent proved to be detrimental to the
reaction’s rate with <20% conversion observed after 4 h (Table 2, entry 9). Amount of Ti(OiPr)4
could be reduced down to 0.5 equiv with no impact over bisꢀaddition impurity levels but further
reduction had a negative impact on reaction’s rate. Ketone stability was also evaluated: no
significant degradation was observed (TLC analysis) after 18 hours under the reaction’s
conditions (Ti(OiPr)4, 0.5 equiv, 2ꢀMeTHF, 23 °C) but only 88% of the ketone was recovered
after this time. It was thus decided to keep the amount of Ti at 0.5 equiv to ensure a fast reaction
time of 2ꢀ3 hours. Addition of 0.5 equiv of Ti(OiPr)4 to a solution of resorcinol (2 equiv) and Nꢀ
Bocꢀ4ꢀpiperidone (1 equiv) in 2ꢀMeTHF at 23 °C led to complete formation of the desired
product (75:25 alcohol to alkene ratio) with <6% of bisꢀaddition impurity in 3 h. Reaction
quenching by 1N HCl, followed by phase separation enabled an easy removal of the Ti residues.
ICPꢀOES analysis of the organic phase showed levels of Ti consistently <40 ppm. Solvent switch
from 2ꢀMeTHF to iPrOH under reflux with a trace amount of AcOH was accompanied by
complete elimination of the alcohol to the corresponding alkene 19. Addition of water under
reflux followed by crystallization gave crude 19 in 69% yield contaminated with <3% of bisꢀ
addition impurity 22. A final slurry of this material in iPrOH/water (1:1) reduced the amount of
22 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
Manufacture of the GMP batch
The first GMP batch was made using this new route for the preparation of the starting
materials. The Ti(IV)ꢀmediated coupling was scaled up to 7 kg of NꢀBoc piperidone, delivering
ACS Paragon Plus Environment
15

Organic Process Research & Development
Page 16 of 30
1
2
3
4
5
6
7
8
9
4.7 kg of 19 in 46% yield and 99.1% HPLC A% purity (Scheme 10). The absence of large scale
hydrogenation equipment (batch or flow) in our Kilo Lab prompted us to focus on transfer
hydrogenation conditions19: addition of an aqueous solution of sodium formate to a suspension
of 19 and 5% Pd/C in a mixture of isopropanol and acetic acid led to the clean reduction of 19
into 30 which was isolated by crystallization after Pd/C filtration and isopropanol removal. This
process was scaled up to deliver >4 Kg of 30 (starting material of the GMP steps) in 92% yield
and 99.2% HPLC A% purity. In all batches of 30 prepared, Ti and Pd content were <1 ppm
(ICPꢀOES analysis).20
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 10. Starting Material Synthesis
GMP manufacture of 1 is depicted in Scheme 11 and started with Boc deprotection of 30,
readily achieved using 5ꢀ6 N hydrogen chloride solution in iPrOH followed by crystallization in
EtOAc to furnish 15-HCl (95% yield, >3 Kg, 98 HPLC 97.9 A% purity). In the second GMP
step, (S)ꢀ(ꢀ)ꢀαꢀmethylbenzylamine 8 was activated with N,N′ꢀDisuccinimidyl carbonate21 (N,N’ꢀ
DSC) in acetonitrile to deliver 31. CDI was originally used for the urea formation but activated
carbamate 13 could not be isolated cleanly and unstability upon storage was problematic. A
telescoped procedure featuring activation of 8 followed by coupling with 15-HCl was devised
but suffered from lack of reproducibility and unacceptable levels of symmetrical urea impurity
ACS Paragon Plus Environment
16

Page 17 of 30
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
32 present in the final API batches. On the contrary, adding a solution of 8 in acetonitrile to a
suspension of N,N’ꢀDSC in acetonitrile at ꢀ5 °C led to the clean formation of the activated
carbamate 31, isolated in 80% yield by crystallization upon water addition with low (<0.7%)
levels of urea impurity 32. Final urea formation was accomplished by treating a solution of 31
and 15-HCl in 3.8 vol of 2ꢀMeTHF and 1 vol of DMF at 50 °C by DIPEA. After an acidic
aqueous workup, 2ꢀMeTHF was swapped under reflux for iPrOH and crude 1 was isolated by
crystallization upon water addition (78% yield, 97.7% HPLC purity).
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 11. Optimized GMP Manufacturing Process of 1
A polymorph study was performed on 1 and only one crystal phase was obtained from multiple
solvents systems22. This opened a wide range of options for the final recrystallization and a 1:1
mixture of iPrOH and water was selected for its volumetric and purification efficiency. Main
issue arose from the color of the isolated API which turned out to be slightly pink after
recrystallization with color enhancement observed upon storage. This behavior was attributed to
the presence of ppm levels of oxidation quinone 33 in the product (Scheme 12).
ACS Paragon Plus Environment
17

Organic Process Research & Development
Page 18 of 30
1
2
3
4
Scheme 12. Postulated Oxidation of 1 under Air
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
The postulated quinone 33 was synthesized from an analogue of 1 fluorinated at position 56
which was treated with aqueous NaOH under air. Deep red 33 could be isolated by
chromatography but turned out to degrade rapidly upon storage.23
We hypothesized that slightly acidic conditions would fully protonate the phenol moieties of 1
and prevent oxidation. Tests were conducted from freshly recrystallized 1 which was dissolved
in a mixture of iPrOH and water and placed under various pH conditions under open air or under
a nitrogen atmosphere (Picture 1). Under open air at pH 12, solution of 1 turned deep red rapidly
with a very intense darkening over time, indicating that oxidation to the quinone under air was
promoted under basic conditions. Under nitrogen, this phenomenon was minimized but still
present while at pH 5 under air, no color change was observed.
Picture 1. Color change observed for solutions of 1 at basic and acidic pH
ACS Paragon Plus Environment
18

Page 19 of 30
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
a) at t = 30 min
b) at t = 8 h
c) at t = 24 h
Solutions of 1 in iPrOH/Water (4:1) at pH = 5 (AcOH) under open air (left), at pH = 12 (1N
aqueous NaOH) under open air (middle), at pH = 12 (1N aqueous NaOH) under N2
atmosphere (right).
Recrystallization solvents were thus buffered with acetic acid (0.025 vol), leading to a white
solid with no color change during storage. This optimized recrystallization procedure was
applied to crude 1 to deliver 2.85 Kg of GMP 1 in 82% yield and >98.5% assay purity.
CONCLUSION
A GMP manufacturing process was developed for the synthesis of 1, a novel and potent
tyrosinase inhibitor. This 5 steps process was carried out to deliver >2 Kg of the final API with
greater than 97.5% chemical purity in 26% overall yield. A direct and efficient route was
devised, featuring an operationally simple Ti(IV)ꢀmediated coupling between free resorcinol and
NꢀBocꢀ4ꢀpiperidone, an alkene reduction by transfer hydrogenation and a urea formation
involving N,N’ꢀDSC as the carbonyl source. This short sequence from cheap and easily available
starting materials reduces significantly the material cost of the final API.
EXPERIMENTAL SECTION
ACS Paragon Plus Environment
19

Organic Process Research & Development
Page 20 of 30
1
2
3
4
5
6
7
8
9
General. All reagents and solvents were purchased commercially and used without further
purification. HPLC/MS were recorded on a Waters UPLC Acquity system. Column: Acquity
CSH C18 1.7 µm 2.1*50 mm; flow rate = 0.8 mL/min; 2.0 min runs; Solvent system:
MeCN+0.1% HCOOH and H2O+0.1% HCOOH. Gradient: 10ꢀ98% MeCN from 0 to 0.8 min,
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
13
98% MeCN from 0.8 to 1.5 min and 98 to 10% MeCN from 1.5 to 2.0 min. H NMR and C
NMR were recorded on a Bruker Avance 400 MHz, 100 M. Chemical shifts are expressed in
parts per million (ppm) downfield from residual solvent peaks, and coupling constants are
reported in hertz (Hz). Splitting patterns are indicated as follows: br, broad; s, singlet; d, doublet;
dd, doublet of doublet; t, triplet; q, quartet; p, quintet; m, multiplet. ICPꢀOES analyses were
performed on a ThermoFisher iCAP 6000 Series spectrometer.
tert-butyl 4-(2,4-Dihydroxyphenyl)-3,6-dihydropyridine-1(2H)-carboxylate (19). In a 100
L double jacketed reactor were charged resorcinol (7.7 kg, 70.2 mol, 2.0 equiv), NꢀBocꢀ4ꢀ
piperidone (7.0 kg, 35.1 mol, 1.0 equiv) and 2ꢀMeTHF (49 L, 7 vol). To this clear solution was
added at 23 °C under vigorous stirring Ti(OiPr)4 (4.9 kg, 17.6 mol, 0.5 equiv). The temperature
of the solution rose from 23 °C to 24.5 °C and a precipitate appeared and dissolved gradually.
The red solution was stirred at 23 °C for 1.5 h. HPLC IPC after this time showed 18.7%
remaining ketone. Extra Ti(OiPr)4 (0.49 kg, 1.76 mol, 0.05 equiv) was added to the solution and
stirring pursued at 23 °C for 1 h. HPLC IPC after this time showed 10% remaining ketone. Extra
Ti(OiPr)4 (0.49 kg, 1.76 mol, 0.05 equiv) was added to the solution and stirring pursued at 23 °C
for 45 min. HPLC IPC after this time showed ≤5% remaining ketone. The mixture was quenched
by addition of a 1 M aqueous solution of HCl (24.5 L) over 15 min at ≤30 °C. To avoid phase
inversion, 2ꢀMeTHF (V = 24.5 L, 3.5 vol) was added to the reaction mixture. Phases were
separated and the organic layer was washed with water (3 X 24.5 L, 3 X 3.5 vol), pH of the final
ACS Paragon Plus Environment
20

Page 21 of 30
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
aqueous wash = 5.5. 2ꢀMeTHF was distilled under atmospheric pressure and stripped by iPrOH
until refractive index of the distillate indicates <10% 2ꢀMeTHF with a total volume in the reactor
of ~35 L. AcOH (3.5 L, 0.5 vol) was added to the mixture and stirring under reflux was pursued
for 1 h. HPLC IPC after this time showed ≤1 % remaining alcohol intermediate 18. At 80 °C was
added water (35 L, 5 vol.) over 20 min: internal temperature dropped to 60 °C and after the end
of the addition the mixture was warmed to 80 °C, stirred for 30 min and allowed to cool to 23 °C
over 18 h. The mixture was filtered through a 6 µm filter medium, the cake washed with a 1:1
mixture of iPrOH and water (7.0 L) and dried under a N2 flux for 1 h. IPC analysis of the crude
solid revealed 3.5% of bisꢀaddition impurity 22. The solid was triturated in 8 L (1.2 vol) of
iPrOH for 1 h at 23 °C, collected by filtration and dried under a N2 flux for 1 h. Content in bisꢀ
addition impurity was lowered to 2.2%. The solid was triturated in 3.5 L (0.6 vol) of iPrOH and
3.5 L (0.6 vol) of water for 1 h at 23 °C, collected by filtration and dried under a N2 flux for 1 h.
Content in bisꢀaddition impurity was lowered to 1.1%. The solid was triturated in 3.5 L (0.6 vol)
of iPrOH and 3.5 L (0.6 vol) of water for 1 h at 23 °C, collected by filtration and dried under
vacuum at 50 °C for 48 h to yield 19 (4.70 kg, 46%, 99.1 A% purity) containing 0.9% of bisꢀ
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
addition impurity 22. H NMR (400 MHz, DMSOꢀd6) δ 9.20 (br, 2H), 6.85 (d, J = 8.3 Hz, 1H),
6.28 (d, J = 2.4 Hz, 1H), 6.17 (dd, J = 8.3, 2.4 Hz, 1H), 5.72 (s, 1H), 3.92 (br, J = 3.6 Hz, 2H),
13
3.46 (t, J = 5.7 Hz, 2H), 2.43ꢀ2.32 (m, 2H), 1.43 (s, 9H); C NMR (101 MHz, DMSOꢀd6) δ
157.4, 155.4, 153.9, 134.7, 129.1, 120.6, 119.9, 106.2, 102.8, 78.7, 43.5, 40.9, 28.5, 28.2; HRMS
(m/z, ESꢀ) for C16H20NO4 (MꢀH)ꢀ calcd. 290.1392, measd. 290.1400.
tert-butyl 4-(2,4-Dihydroxyphenyl)piperidine-1-carboxylate (30). In a 100 L double
jacketed reactor were charged 19 (4.65 kg, 15.96 mol, 1.0 equiv), iPrOH (37.2 L, 8 vol) and
AcOH (2.74 L, 3 vol). The suspension was heated to 55 °C and turned to a solution in ~30 min.
ACS Paragon Plus Environment
21

Organic Process Research & Development
Page 22 of 30
1
2
3
4
5
6
7
8
9
To this solution was added a suspension of palladium 5% w/w on activated carbon, Degussa
Type E105CA/W 50% wet (0.46 kg, 10% w/P) in iPrOH (0.92 L). A solution of sodium formate
(2.71 kg, 39.90 mol, 2.5 equiv) in water (37.2 L, 8 vol) was prepared separately. This solution
was added to the reaction mixture at 50 °C over 1 h. HPLC IPC after 2 h at 50 °C showed
complete conversion with ≤1 % of remaining 19. The reaction medium was cooled down to 23
°C and filtered through a 6 µm filter medium. The filtrate was transferred in a clean reactor and
iPrOH was distilled under atmospheric pressure. When distillate volume reached 28 L, the
reaction mixture was allowed to cool down to 23 °C over 3 h and maintained at this temperature
for 18 h. The reaction medium was filtered through a 25 µm filter medium, the cake washed with
water (5 L) and the solid collected and dried under vacuum at 50 °C for 48 h to yield 30 (4.30 kg,
92%, 99.2 A% purity). 1H NMR (400 MHz, DMSOꢀd6) δ 9.09 (br, 2H), 6.80 (d, J = 8.3 Hz, 1H),
6.30 (d, J = 2.4 Hz, 1H), 6.16 (dd, J = 8.3, 2.4 Hz, 1H), 4.06 (d, J = 12.7 Hz, 2H), 2.92ꢀ2.79 (m,
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
13
1H), 2.75 (br, 2H), 1.70 – 1.61 (m, 2H), 1.41 (s, 9H); C NMR (101 MHz, DMSOꢀd6) δ 156.2,
155.2, 153.9, 126.8, 122.3, 106.1, 102.4, 78.4, 44.5, 34.5, 31.7, 28.2; HRMS (m/z, ESꢀ) for
C16H22NO4 (MꢀH)ꢀ calcd. 292.1549, measd. 292.1556.
4-(Piperidin-4-yl)-benzene-1,3-diol hydrochloride (15-HCl). In a 100 L double jacketed
reactor were charged 30 (4.23 kg, 14.41 mol, 1.0 equiv) and iPrOH (34.0 L, 8.0 vol). The
suspension was heated to 50 °C and turned to a solution in ~1 h. To this solution was added
hydrogen chloride solution 5ꢀ6 N in iPrOH (7.18 L, 36.03 mol, 2.5 equiv) and the reaction
mixture was stirred at 50 °C for 22 h. HPLC IPC after this time showed complete conversion
with ≤1 % of 30. EtOAc (34 L, 8.1 vol) was added and the reaction mixture was cooled down to
23 °C over 1 h. The reaction medium was filtered through a 25 µm filter medium, the cake
washed with EtOAc (4.2 L) and the solid collected and dried under vacuum at 50 °C for 48 h to
ACS Paragon Plus Environment
22

Page 23 of 30
Organic Process Research & Development
1
2
1
3
4
yield 15-HCl (3.11 kg, 95%, 97.9 A% purity). H NMR (400 MHz, DMSOꢀd6) δ 9.34 (s, 1H),
5
6
9.09 (s, 1H), 8.99 (br, 2H), 6.78 (d, J = 8.3 Hz, 1H), 6.34 (d, J = 2.3 Hz, 1H), 6.19 (dd, J = 8.3,
7
8
9
13
2.4 Hz, 1H), 3.29 (dt, J = 12.7, 3.2 Hz, 2H), 3.07ꢀ2.49 (m, 3H), 1.97ꢀ1.60 (m, 4H); C NMR
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
(101 MHz, DMSOꢀd6) δ 156.4, 155.2, 126.4, 121.1, 106.1, 102.5, 43.7, 32.3, 28.3; HRMS (m/z,
ES+) for C11H16NO2 (M+H)+ calcd. 194.1181, measd. 194.1178.
2,5-dioxopyrrolidin-1-yl-(S)-(1-Phenylethyl)carbamate (31). In a 100 L double jacketed
reactor were charged N,N’ꢀDSC (7.35 kg, 28.90 mol, 1.0 equiv) and MeCN (9.0 L, 2.6 vol). The
suspension was cooled to ꢀ10 °C. To this suspension was added a solution of 8 (3.50 kg, 28.90
mol, 1.0 equiv) in MeCN (7.0 L, 2.0 vol) over 25 min keeping the internal temperature below ꢀ4
°C. HPLC IPC after 1 h showed complete conversion with ≤1 % of remaining 8. Water (52.0 L,
14.9 vol) was charged over the reaction mixture over 25 min and the reaction medium was
allowed to warm to 23 °C. The suspension was stirred at 23 °C for 2 h, filtered through a 25 µm
filter medium, the cake was washed with water (4.1 L) and the solid collected and dried under
vacuum at 45 °C for 48 h to yield 31 (6.02 kg, 79%, 94.2 A% purity). 1H NMR (400 MHz,
DMSOꢀd6) δ 8.86 (d, J = 7.9 Hz, 1H), 7.47ꢀ7.08 (m, 5H), 4.68 (p, J = 7.1 Hz, 1H), 2.75 (s, 4H),
1.43 (d, J = 7.0 Hz, 3H); 13C NMR (101 MHz, DMSOꢀd6) δ 170.8 , 151.1 , 143.5 , 128.4 , 127.0 ,
125.8 , 51.1 , 25.2 , 22.4; HRMS (m/z, ESꢀ) for C13H13N2O4 (MꢀH)ꢀ calcd. 261.0875, measd.
261.0883.
(S)-4-(2,4-Dihydroxyphenyl)-N-(1-Phenylethyl)piperidine-1-carboxamide (1). In a 100 L
double jacketed reactor were charged 15-HCl (3.05 kg, 13.31 mol, 1.0 equiv), 31 (3.51 kg, 13.38
mol, 1.0 equiv) followed by DMF (3.0 L, 0.98 vol) and 2ꢀMeTHF (11.5 L, 3.80 vol). The
suspension was heated to 50 °C. To this suspension was added DIPEA (2.57 kg, 19.87 mol, 1.49
equiv) over 25 min. The suspension turned gradually to a solution. HPLC IPC after 3 h showed
ACS Paragon Plus Environment
23

Organic Process Research & Development
Page 24 of 30
1
2
3
4
5
6
7
8
9
complete conversion with ≤1 % of remaining 15 and ≤5 % of remaining 31. The reaction
medium was diluted with 2ꢀMeTHF (10.0 L, 3.28 vol) and the reaction quenched by addition of
1M aqueous HCl (37.0 L, 12.0 vol) over 10 min. Phases were separated at 40 °C and the organic
layer washed with water (6.0 L, 2.0 vol). Phases were separated and the organic layer was
filtered through a 6 µm filter medium to remove mechanical impurities. The filtrate was charged
in a clean reactor and 2ꢀMeTHF was distilled under atmospheric pressure and stripped by iPrOH
until refractive index of the distillate indicates <10% 2ꢀMeTHF. Distillation of iPrOH was
pursed to reach a final volume of 9.1 L (3.0 vol) of iPrOH in the reactor. Under reflux was added
water (15.0 L, 5.0 vol) over 30 min maintaining the internal temperature above 60 °C. The
reaction medium was then allowed to cool to 23 °C over 20 h. The suspension was filtered
through a 25 µm filter medium, the cake was washed with water (2 X 1.5 L) and the solid
collected and dried under vacuum at 45 °C for 48 h to yield crude 1 (3.51 kg, 77%, 97.7 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
purity). Recrystallization: In a 100 L double jacketed reactor were charged crude 1 (3.51 kg,
10.31 mol, 1.0 equiv), iPrOH (27.0 L, 7.5 vol), AcOH (74.1 g) and water (27.0 L, 7.5 vol). The
suspension was warmed to reflux and turned to a solution after 30 min of reflux. Heating was
stopped and the reaction medium was allowed to cool to 23 °C over 20 h. The suspension was
filtered through a 25 µm filter medium, the cake was washed with a mixture of water (3.6 L) and
AcOH (7.3 g) and the solid collected and dried under vacuum at 45 °C for 48 h to yield 1 (2.86
kg, 81%, 98.5 A% purity). 1H NMR (400 MHz, DMSOꢀd6):
δ 9.11 (s, 1H), 8.96 (s, 1H), 7.30ꢀ
7.31 (m, 4), 7.19ꢀ7.20 (m, 1H), 6.79 (d, J = 8.3 Hz, 2H), 6.7 (d, J = 7.9 Hz, 2H), 6.28 (d, J = 2.4
Hz, 1H), 6.16 (dd, J = 8.3, 2.4 Hz, 1H), 4.85ꢀ4.87 (m, 1 H), 4.13 (d, J = 12.9 Hz, 2H), 2.85 (t, J =
13
11.9 Hz, 1H), 2.70 (t, J = 12.7 Hz, 2H), 1.64 (d, J = 12.1 Hz, 2H), 1.40ꢀ1.41 (m, 5H). C NMR
(101 MHz, DMSOꢀd6) δ 156.6, 156.0, 155.2, 146.3, 127.9, 126.7, 126.1, 125.9, 122.5, 106.0,
ACS Paragon Plus Environment
24

Page 25 of 30
Organic Process Research & Development
1
2
102.4, 49.3, 44.4, 34.7, 31.8, 31.7, 22.9; mp: 200ꢀ201 °C; HRMS (m/z, ES+) for C20H25N2O3
(M+H)+ calcd. 341.1865, measd. 341.1859.
3
4
5
6
7
8
9
tert-butyl 4-(2-Hydroxy-4-methoxyphenyl)-3,6-dihydropyridine-1(2H)-carboxylate (28). To
a stirred solution of NꢀBocꢀ4ꢀpiperidone (2.00 g, 0.01 mol, 1.0 equiv) and 3ꢀMethoxyphenol
(2.49 g, 0.02 mol, 2.0 equiv) in 2ꢀMeTHF (20.00 mL, 10.0 vol) was added Ti(OiPr)4 (1.49 ml,
0.01 mol, 0.5 equiv). An orange solid formed upon addition and solubilized slowly. TLC
analysis after 5 h showed complete conversion The reaction mixture was quenched with 1 N HCl
(20 mL) and extracted with 2ꢀMeTHF (2 X 20 mL). The combined organic layers were washed
with brine (10 mL) and evaporated to dryness. The orange residue was stirred overnight in
iPrOH (20 mL) and AcOH (0.29 mL, 0.01 mol, 0.50 equiv) at 80 °C to fully dehydrate the
intermediate. LCMS analysis after 15 h showed complete elimination. The solution was
evaporated to dryness and purified by chromatography eluting with DCM/EtOAc 9:1 to yield 28
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
(2.18 g, 71%) as a white crystalline solid. H NMR (400 MHz, DMSOꢀd6) δ 9.45 (s, 1H), 6.98
(d, J = 8.4 Hz, 1H), 6.39 (d, J = 2.6 Hz, 1H), 6.35 (dd, J = 8.4, 2.5 Hz, 1H), 5.76 (s, 1H), 3.96ꢀ
13
3.90 (m, 2H), 3.68 (s, 3H), 3.47 (t, J = 5.7 Hz, 2H), 2.44ꢀ2.36 (m, 2H), 1.43 (s, 9H). C NMR
(101 MHz, DMSOꢀd6) δ 159.3, 155.5, 153.9, 134.4, 129.3, 121.6, 121.2, 104.3, 101.6, 78.7, 54.9,
40.2, 28.4, 28.2; HRMS (m/z, ESꢀ) for C17H22NO4 (MꢀH)ꢀ calcd. 304.1549, measd. 304.1555.
(S)-4-(4-Hydroxy-3,6-dioxocyclohexa-1,4-dien-1-yl)-N-(1-phenylethyl)piperidine-1-
carboxamide
(33).
(S)ꢀ4ꢀ(5ꢀFluoroꢀ2,4ꢀdihydroxyphenyl)ꢀNꢀ(1ꢀphenylethyl)piperidineꢀ1ꢀ
carboxamide6 (0.10 g, 0.27 mmol, 1.0 equiv) was dissolved in a mixture of water (10.0 mL,
100.0 vol) and EtOH (10.0 mL, 100.0 vol) and placed in an open beaker under air. To this
solution was added an aqueous solution of sodium hydroxide (0.27 mL of a 10 M solution, 2.70
mmol, 10.0 equiv). The dark red reaction mixture was stirred at 23 °C for 1 h. TLC analysis
ACS Paragon Plus Environment
25

Organic Process Research & Development
Page 26 of 30
1
2
3
4
5
6
7
8
9
(DCM/MeOH 9:1) after this time showed 1 major dark red spot. The reaction was quenched by
the addition of 50 mL of a 2 M aqueous citric acid solution and turned yellow. The aqueous layer
was extracted with EtOAc (2 X 100 mL) and the combined organic layers were washed with
water, dried over MgSO4 and evaporated to dryness under vacuum (temperature of the heating
bath was kept <30 °C) to yield a yellow oil. This residue was purified by flash chromatography
over silica gel using DCM/EtOAc 1:1 and the red spot was isolated to yield 33 (509 mg, 51%).
as an orange oil. A rapid darkening of the isolated sample was noticed upon concentration and
we were able to record only a 1H NMR and a LC/MS analysis before it degrades into multiple by
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
products. H NMR (400 MHz, DMSOꢀd6) δ 11.44 (br, 1H), 7.36ꢀ7.25 (m, 4H), 7.23ꢀ7.13 (m,
1H), 6.75 (d, J = 7.9 Hz, 1H), 6.47 (d, J = 1.1 Hz, 1H), 5.93 (s, 1H), 4.90ꢀ4.75 (m, 1H), 4.19ꢀ
4.07 (m, 2H), 2.84ꢀ2.61 (m, 3H), 1.62 (d, J = 12.9 Hz, 2H), 1.36 (d, J = 7.1 Hz, 3H), 1.33ꢀ1.22
(m, 2H). LCMS (m/z, ES+) for C20H22N2O4 (M+H)+ calcd. 355.1, measd. 355.3.
ASSOCIATED CONTENT
1H and 13C NMR spectra. This information is available free of charge via the Internet at
http://pubs.acs.org/
AUTHORS INFORMATION
Corresponding Authors
*Eꢀmail: Thibaud.Gerfaud@galderma.com
*Eꢀmail: JeanꢀGuy.Boiteau@galderma.com
ORCID
Thibaud Gerfaud : 0000ꢀ0003ꢀ2101ꢀ2185
ACS Paragon Plus Environment
26

Page 27 of 30
Organic Process Research & Development
1
2
3
4
JeanꢀGuy Boiteau : 0000ꢀ0002ꢀ8184ꢀ471X
5
6
Notes
7
8
9
The authors declare no competing financial interest.
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
ACKNOWLEDGMENTS
The authors would like to thank AnneꢀLaurence Ghilini and Céline Mathieu for analytical
support.
ACS Paragon Plus Environment
27

Organic Process Research & Development
Page 28 of 30
1
2
3
4
REFERENCES
5
6
7
8
[1] Briganti, S.; Camera, E.; Picardo, M. Pigment Cell Res. 2003, 16, 101.
[2] Brenner, M.; Hearing, V. J. Photochem. Photobiol. 2008, 84, 539.
[3] a) Schallreuter K. U.; Kothari S.; Chavan B.; Spencer J. D. Exp. Dermatol. 2008, 17, 395; b)
Cooksey, C. J.; Garratt, P. J.; Land, E. J.; Pavel, S.; Ramsden, C. A.; Riley, P. A.; Smit, N. P. J.
Biol. Chem. 1997, 272, 26226; c) Stratford M. R. L.; Ramsden C. A.; Riley P. A. Bioorg. Med.
Chem. 2013, 21, 1166.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
[4] Chang, T. S. Int. J. Mol. Sci. 2009, 10, 2440.
[5] Hypopigmentation effect have already been demonstrated for resorcinols see: a) Kim, D.S;
Kim, S.Y.; Park, S.H.; Choi, Y.G. Kwon, S.B.; Kim, M.K.; Na, J.I.; Youn, S.W.; Park, K.C.
Biol. Pharm. Bull. 2005, 28, 2216; b) Khemis, A.; Kaiafa, A.; QueilleꢀRoussel, C.; Duteil, L.;
Ortonne, J. P. Br. J. Dermatol. 2007 156, 997.
[6] Boiteau, JꢀG.; Bouquet, K.; Talano, S.; MilloisꢀBarbuis, C. Patent WO 2010/063774 A1,
2010.
[7] For an example of Suzuki coupling involving 2,4ꢀdimethoxyphenylboronic acid with enol
triflate, see: Kolodziej, S. A.; Hockerman, S. L.; Boehm, T. L.; Carroll, J. N.; DeCrescenzo, G.
A.; McDonald, J. J.; Mischke, D. A.; Munie, G. E.; Fletcher, T. R.; Rico, J. G.; Stehle, N. W.;
Swearingen, C.; Becker D. P. Bioorg. Med. Chem. Lett. 2010, 20, 3557.
[8] Liu, X.; Go, M. Bioorg. Med. Chem. 2007, 15, 7021.
[9] a) Neiderl, J. B.; Vogel, H. J. J. Am. Chem. Soc. 1940, 62, 2512; b) Schneider, U.; Schneider,
H. J. Chem. Ber. 1994, 127, 2455.
ACS Paragon Plus Environment
28

Page 29 of 30
Organic Process Research & Development
1
2
3
4
5
6
[10] a) Causse, M. H. Bull. Soc. Chim. Paris, Ser. 1892, 7, 563; b) Livant, P.; Webb, T. R.; Xu,
W. J. Org. Chem. 1997, 62, 737; c) Poupelin et al. Eur. J. Med. Chem. 1980, 15, 253
7
8
9
[11] Filimonov, S. I.; Savinsky, N. G.; Evstigneeva, E. M. Mendeleev Commun. 2003, 13, 194.
[12] Boiteau, JꢀG.; Musicki, B. Patent WO 2014/125233 A2, 2014.
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
[13] Various solvents (DCM, acetone, EtOH, iPrOH, MeOH, THF) and antisolvents (water,
TBME, heptane, chlorobenzene) combinations at various concentrations (from 10 to 25 V/P)
were tested. Only 2 successive recrystallizations in chlorobenzene/acetone (7:3 ratio, 11 V/P)
enabled to reduce the amount of 22 from 5% to <0.5% but only in low (35%) yield.
[14] See the Supporting Information for details.
[15] Aycock, D. F. Org. Process Res. Dev. 2007, 11, 156
[16] No precipitate was observed under the optimized conditions in the absence of resorcinol.
[17] Attempts to characterize the solid formed were unsuccessful; the solid was insoluble in all
deuterated solvents commonly used for NMR analysis.
[18] Ti(OiPr)4 can be obtained from SigmaꢀAldrich for 100€/L (30€/mol) whereas Ti(OtBu)4 is
sold 10820€/L (4180€/mol). Mixed organotitanium complexes (e.g. TiClx(OR)y) along with
other types of ligands (e.g. αꢀketo esters…) have not been investigated: focus was put on cheap
and commercially available titanium (IV) sources with operationally simple conditions in order
to be able to scaleꢀup the process rapidly.
[19] a) Brieger, G.; Nestrick, T. J. Chem. Rev. 1974, 74, 567; b) Ram, S.; Ehrenkaufer, R. E.
Synthesis 1988, 91.
[20] Internal specification for clinical API batches at Nestlé Skin Health are set to <10 ppm for
Pd and Ti.
[21] Takeda, K.; Haruo, O. Synthetic Commun. 1982, 12, 213.
ACS Paragon Plus Environment
29