An alternative synthesis of the lipophilic tail portion of abediterol using linear-selective hydroformylation

Article abstract of DOI:10.1016/j.tetlet.2019.01.037

Abediterol is a compound currently in development for the treatment of respiratory disease at AstraZeneca. In this work we employ hydroformylation, an under-utilised reaction in the synthesis of pharmaceutical intermediates, as a key step to access the li

Full text of DOI:10.1016/j.tetlet.2019.01.037

Accepted Manuscript
An alternative synthesis of the lipophilic tail portion of abediterol using linear-
selective hydroformylation
Rachel H. Munday, Louis Goodman, Gary M. Noonan
PII:
S0040-4039(19)30066-8
https://doi.org/10.1016/j.tetlet.2019.01.037
TETL 50570
DOI:
Reference:
Tetrahedron Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
2 December 2018
17 January 2019
20 January 2019
Please cite this article as: Munday, R.H., Goodman, L., Noonan, G.M., An alternative synthesis of the lipophilic
tail portion of abediterol using linear-selective hydroformylation, Tetrahedron Letters (2019), doi: https://doi.org/
10.1016/j.tetlet.2019.01.037
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

An alternative synthesis of the lipophilic tail portion of abediterol using linear-selective
hydroformylation
Rachel H Mundaya, Louis Goodmana, Gary M. Noonana*
a Pharmaceutical Technology and Development, AstraZeneca, Silk Road Business Park, Macclesfield SK10 2NA, United Kingdom
*Corresponding Author e-mail: gary.noonan1@astrazeneca.com
Abstract
Abediterol is a compound currently in development for the treatment of respiratory disease at
AstraZeneca. In this work we employ hydroformylation, an under-utilised reaction in the synthesis of
pharmaceutical intermediates, as a key step to access the lipophilic amine portion of this molecule.
The route described herein addresses some of the issues identified in the original route. Namely,
hazardous materials are avoided and increased levels of control - from a process chemistry point of
view - are demonstrated, with isolation of intermediates possible at multiple points in the synthesis.
This work provides ‘proof-of-concept’ for an alternative synthetic route and provides high-quality
material using a series of robust transformations.
Introduction
Abediterol (AZD0548) is a highly-potent, long-acting β2-agonist that is currently in development at
AstraZeneca for the treatment of respiratory disease.1 The currently employed synthetic route to the
lipophilic amine of abediterol is shown below (Scheme 1) and commences from a commercially
available fluorinated ester (1).
F
F
F
F
F F
c
a
b
O
OH
O
Br
)
6
(
O
(1)
(2)
F
F
O
F
F
O
(
)
6
(
)
O
6
(
)
6
NH3
-O
d, e
NH
F
F
NHPhth
OH
O
O
HO
OH
HN
(3)
Steps a-e: 30% overall yield
O
Abediterol
Scheme 1: Current synthetic route.
Reagents and conditions: (a) LiAlH4, THF, reflux; (b) 1,6-dibromohexane (5.2 eq.), TBAB, NaOHaq (33% w/w); (c)
potassium phthalimide, acetone, reflux; (d) hydrazine, EtOH, reflux; (e) fumaric acid, EtOH. (30 % isolated yield
over telescoped sequence, Steps a-e)
This synthesis, although relatively concise, uses over five equivalents of 1,6-dibromohexane in step
(b), to minimise the formation of the unwanted symmetrical di-ether. Residual 1,6-dibromohexane is
then present in the starting material for the phthalimide substitution in step (c), which leads to the
formation of further impurities. Finally, the phthalimide group is removed using hydrazine, a highly
toxic and potentially explosive compound.2 Hydrazine use is also of regulatory concern and is on the
European Chemicals Agency (ECHA) ‘candidate list of substances of very high concern for
authorisation’ and as a result requires an application for authorisation of use, which involves a 45-day
public consultation period.3 Therefore, an alternative route that avoids the use of hydrazine is
desirable.

Route design and selection for the synthesis of an active pharmaceutical intermediate are complex
decisions with many factors to be considered.4 For example, the current synthesis contains only a
single isolation i.e. the precipitation and filtration of the fumarate salt. Demonstration of ‘control’ in
a pharmaceutical process is a key regulatory requirement and a telescoped process that uses a large
excess of reagents, generating relatively large quantities of impurities, is sub-optimal from this point
of view. The ultra-high potency of abediterol means that the projected annual global supply after
launch is of the order of kilograms. Therefore, our synthetic efforts focused on the production of
material using an increased number of potential isolation points, robust chemistry, and minimal
reagent excess where possible. Thus, we envisioned the alternative route vide infra that employs a
linear-selective hydroformylation as a key step.
Hydroformylation is now a highly developed reaction technology, with millions of tonnes of aldehydes
produced every year via this olefin homologation process.5 The linear-selective hydroformylation of
unfunctionalized olefins is particularly refined, driven largely by the identification, and subsequent
development, of wide bite-angle diphosphine and bulky diphosphite ligands.6 Despite the obvious
potential utility of this reaction, there are relatively few reports of its application to the synthesis of
pharmaceutical intermediates.5a, 7 Some examples of the use of this reaction in the synthesis of active
pharmaceutical intermediates are given in Figure 1, with the key ‘hydroformylation disconnection’
and the carbon atom derived from carbon monoxide highlighted. Our planned application of a
hydroformylation approach to the current target and subsequent downstream chemistry is described
below.
CF3
OH
H
C
N
N
C
Tolterodine
Cinacalcet
H
O
N
N
C
H
O
S
O
OH
O
ilepatril
Figure 1: API’s previously accessed through hydroformylation showing the key disconnection. The C-atom in
bold is added via hydroformylation
Results and Discussion
Our alternative synthetic strategy is shown in Scheme 2. Alkylation of intermediate (2) - the product
of the first step of the currently employed synthetic route, with 5-bromopent-1-ene, would form olefin
(4). Then, linear selective hydroformylation would convert (4) to (5). The resulting aldehyde would be
reduced to the alcohol and the protected amine subsequently formed via a Mitsunobu reaction.8
Deprotection would generate the free amine, which can then be converted to salt (3).

F
F
F
F
F
F F
OH
O
O
O
O
(2)
(4)
(5)
F
O
-O
+H3N
O
HO
(3)
Scheme 2: Proposed strategy for the synthesis of (3)
Reduction of ethyl 2,2-difluoro-2-phenyl-acetate using lithium aluminium hydride produced the
desired product (2) in 75% yield and the subsequent phase-transfer-catalysed alkylation gave (4) in
80% yield. With the olefin in-hand we next turned our attention to the hydroformylation reaction.
Initially, we carried out the reaction using a rhodium-diphosphite pre-catalyst system (Biphephos as
ligand), previously shown to produce the linear aldehyde in high yield.7f To our surprise, the reaction
generated close to a 1:1 mixture of linear and branched aldehydes. Repeating the initial reaction at a
variety of temperatures did not change the selectivity significantly. These results prompted us to look
at alternative ligands to increase selectivity. As expected, variation of the catalyst system had a large
effect on reactivity/selectivity. However, we also found that significant changes in reaction
concentration - for a given catalyst system - produced very different profiles, in terms of both
regioselectivity and by-product formation. Table 1 shows a small selection of results obtained using
the Rh-Xantphos6d and Rh-BISBI6c catalyst systems to illustrate this concentration effect.
F
F
O
O
F
F
[Rh(acac)(CO)2] (1 mol%)
Ligand (1.2 mol%)
O
(5)
Solvent
H2/CO (1:1)
80 °C, 16 h
+
(4)
Isomeric aldehydes
and olefins
alkenes
+alkanec
(%)b
5
44
6
branched
aldehydes
(%)b
Linear
aldehyde
(%)b
92
45
92
76
Solvent
Ligand
(rel. vol.)a
Xantphosd
Xantphosd
BISBId
10
100
10
3
11
2
BISBId
100
11
13
a Refers to the number of solvent volumes of toluene relative to the weight of substrate i.e. 10 mL solvent for 1 g of substrate = 10 rel. vol.; b Percentage of
total peak area in GC; c Note: Side-products not isolated and characterised, were designated ‘alkenes+alkane’, as these products have very similar GC
retention times to the olefin, whereas the retention time for the ‘oxo-products’ are ~ 1 min later on the chromatogram d For ligand structures see ESI
Table 1: Effect of concentration on selectivity
Under otherwise identical conditions, the relative proportion of reaction by-products - which we
ascribe to the hydrogenation product and alkene isomerisation products - increases significantly under
dilute conditions. The regioselectivity also decreases under dilute reaction conditions. These changes
are not particularly surprising, as the reaction mechanism for a hydroformylation reaction is complex,
with the reaction outcome very much dependent on the conditions employed.9
Having carried out reactions using a variety of monodentate and bidentate ligands, in a number of
solvents, we found that the diphosphine ligands Xantphos6d and BISBI6c gave relatively clean formation
of the linear aldehyde, when the reactions were carried out using 10 relative volumes of toluene. The
hydroformylation process was carried out successfully on 4 g-scale using the Rh-Xantphos system and

produced the linear aldehyde cleanly in very high selectivity (l:b > 30:1). A minor impurity, which we
ascribe to olefin isomerisation was also observed (see ESI). The resulting aldehyde sample was then
reduced to the corresponding alcohol using sodium borohydride. This resulted in a 77% yield after
purification, over the hydroformylation-reduction sequence (Steps (c)-(d), Scheme 3).
As an aside, the alkyl aldehyde product of the hydroformylation reaction is prone to atmospheric
oxidation on standing. The rate of this product degradation can be reduced dramatically by the
addition of an appropriate alcohol additive.10 As well as the high boiling alcohols previously reported
e.g. benzyl alcohol, we have found that the more easily removed n-butanol can also be employed. If
the reduction is carried out soon after the aldehyde is produced, the formation of the acid is not an
issue.
Conversion of the alcohol to the corresponding amine was performed by a Mitsunobu reaction (Step
(e), Scheme 3) using tetrachlorophthalimide as a ‘pre-nucleophile’, producing material of very high
purity (> 99.5% based on proton NMR assay) in 84% yield.11 Tetrachlorophthalimide has been shown
to be an effective ammonia equivalent for the Mitsunobu reaction,11 with the benefit of facile
deprotection in the presence of aqueous methylamine.12 Deprotection using methylamine (Step (f),
Scheme 3) provided the desired amine in 74% yield. Although not demonstrated here, this amine can
be straight-forwardly converted to the corresponding fumarate salt, as per our currently employed
route. This new route produces intermediate (8) in 29 % overall yield; making it competitive in terms
of yield relative to the process shown in Scheme 1 (assuming the fumarate salt formation is high
yielding).
O
F
F
F
F
F
F
F
F
F F
(i)
(ii)
(iii)
(iv)
O
OH
O
O
O
OH
)
6
(
)
(
)
(
3
5
93%
80%
69%
Two-steps
O
(2)
(4)
(5)
(6)
Cl
Cl
O
F
F
F F
-O
O
)
O
O
Cl
(vi)
(v)
NH3
F
F
O
O
N
74%
84%
(
6
Cl
H2N
HO
O
(7)
(8)
(3)
Scheme 3: Hydroformylation route to ‘tail amine’
Reagents and conditions: (a) LiAlH4, THF, reflux; (b) 5-bromopent-1-ene, TBAB, NaOH (aq., 33% w/w), 70 °C; (c)
[Rh(acac)(CO)2] (1 mol %), Xantphos (1.2 mol%), CO:H2 1:1 (3 barg), toluene, 80 °C; (d) NaBH4, MeOH, 0 °C-rt; (e)
triphenylphosphine, 3,4,5,6-tetrachlorophthalimide, DIAD, THF, 0 °C-rt; (f) methylamine (aq., 40% w/w), DMF, 40 °C.
Conclusion
We have developed an alternative synthesis of the lipophilic amine portion of abediterol, which
represents a successful proof-of-concept for a hydroformylation approach and provides the amine
product in moderate yield and high purity. The primary benefit of this route is the increased control,
provided by several optional points of isolation, where the intermediates can be purified and may be
amenable to long-term storage. Also, each intermediate is formed cleanly, with few reaction by-
products; this could lead to shorter processing time, another key consideration when carrying the
synthesis of an active pharmaceutical intermediate. Finally, the undesirable use of hydrazine is
obviated by the route described herein.
Acknowledgements

We would like to thank AstraZeneca plc for funding this work. I would also like to thank Dr Andrew D.
Campbell for useful discussions and input during the preparation of this manuscript.
References
1a
Medicine 2014, 108, 1424; b
Ruiz, S. The Journal of Clinical Pharmacology 2014, 54, 1347.
Beier, J.; Fuhr, R.; Massana, E.; Jiménez, E.; Seoane, B.; de Miquel, G.; Ruiz, S. Respiratory
Timmer, W.; Massana, E.; Jimenez, E.; Seoane, B.; de Miquel, G.;
2
3
Niemeier, J. K.; Kjell, D. P. Organic Process Research & Development 2013, 17, 1580.
ECHA (European Chemicals Agency) In Candidate List of substances of very high concern for
Authorisation (Hydrazine),, 20/06/2011.
4
5a
Cornwall, P.; Diorazio, L. J.; Monks, N. Bioorganic & Medicinal Chemistry 2018, 26, 4336.
Franke, R.; Selent, D.; Borner, A. Chem Rev 2012, 112, 5675; b Van Leeuwen, P. W. N. M.;
Claver, C.; Editors Rhodium Catalyzed Hydroformylation. [In: Catal. Met. Complexes, 2000; 22];
Kluwer, 2000.
6a
Billig, E.; Abatjoglou, A. G.; Bryant, D. R. In Transition metal complex-catalyzed processes;
Union Carbide Corp., USA . 1987. pp 83 pp; b Billig, E.; Abatjoglou, A. G.; Bryant, D. R. In
Homogeneous rhodium carbonyl compound-phosphite ligand catalysts and process for olefin
hydroformylation; Union Carbide Corp., USA . 1988. pp 27 pp. Cont; c Casey, C. P.; Whiteker, G. T.;
Melville, M. G.; Petrovich, L. M.; Gavney, J. A., Jr.; Powell, D. R. J. Am. Chem. Soc. 1992, 114, 5535; d
Kranenburg, M.; van der Burgt, Y. E. M.; Kamer, P. C. J.; van Leeuwen, P. W. N. M.; Goubitz,
K.; Fraanje, J. Organometallics 1995, 14, 3081.
7a
Abrams, M. L.; Buser, J. Y.; Calvin, J. R.; Johnson, M. D.; Jones, B. R.; Lambertus, G.; Landis, C.
R.; Martinelli, J. R.; May, S. A.; McFarland, A. D.; Stout, J. R. Organic Process Research & Development
2016, 20, 901; bBotteghi, C.; Cazzolato, L.; Marchetti, M.; Paganelli, S. J. Org. Chem. 1995, 60, 6612; c
Botteghi, C.; Chelucci, G.; Del Ponte, G.; Marchetti, M.; Paganelli, S. J. Org. Chem. 1994, 59,
7125; d Botteghi, C.; Corrias, T.; Marchetti, M.; Paganelli, S.; Piccolo, O. Org. Process Res. Dev. 2002,
6, 379; e
Botteghi, C.; Marchetti, M.; Paganelli, S.; Persi-Paoli, F. Tetrahedron 2001, 57, 1631; f
Cobley, C. J.; Hanson, C. H.; Lloyd, M. C.; Simmonds, S.; Peng, W.-J. Org. Process Res. Dev.
2011, 15, 284; gRheude, U.; Vicari, M.; Aquila, W.; Wegner, G.; Niekerken, J. In Process for the
production of the C5-acetate used for the manufacture of vitamin A by the hydroformylation of
distillation-purified 3,4-diacetoxy-1-butene; BASF Aktiengesellschaft, Germany . 2002. pp 6 pp; h
Thiel, O.; Bernard, C.; Larsen, R.; Martinelli, M. J.; Raza, M. T. In Methods of synthesizing
cinacalcet and salts thereof; Amgen Inc., USA . 2008. pp 58pp; i Verspui, G.; Elbertse, G.; Sheldon, F.
A.; Hacking, M. A. P. J.; Sheldon, R. A. Chem. Commun. (Cambridge) 2000, 1363; jWhiteker, G. T.;
Cobley, C. J. In Applications of Rhodium-Catalyzed Hydroformylation in the Pharmaceutical,
Agrochemical, and Fragrance Industries, 2012. pp 35.
8
9
Mitsunobu, O.; Yamada, M. Bulletin of the Chemical Society of Japan 1967, 40, 2380.
Brezny, A. C.; Landis, C. R. Accounts of Chemical Research 2018.
10
Sankar, M.; Nowicka, E.; Carter, E.; Murphy, D. M.; Knight, D. W.; Bethell, D.; Hutchings, G. J.
Nat. Commun. 2014, 5, 4332/1.
11
12
Watts, R. E.; Siegel, M.; Khosla, C. Journal of Medicinal Chemistry 2006, 49, 7493.
Hoffmann, M.; Dahmann, G.; Fiegen, D.; Handschuh, S.; Klicic, J.; Linz, G.; Schaenzle, G.;
Schnapp, A.; East, S. P.; Mazanetz, M. P.; Scott, J.; Walker, E. In Preparation of substituted
naphthyridines and their use as therapeutic Syk kinase inhibitors; Boehringer Ingelheim International
GmbH, Germany . 2011. pp 194pp.


Application of hydroformylation to the synthesis of a pharmaceutical intermediate
Improved synthesis with fewer impurities formed


Avoids the use of toxic and explosive hydrazine
F
F
F
F
Steps
O
O
NH2
O
F
F
Hydroformylation route enables a
hydrazine-free synthesis, with no reagents
present in large excess!
O
O
Key intermediate