Chemoenzymatic Synthesis of Optically Pure l- and d-Biarylalanines through Biocatalytic Asymmetric Amination and Palladium-Catalyzed Arylation

Article abstract of DOI:10.1021/acscatal.5b01132

A chemoenzymatic approach was developed and optimized for the synthesis of a range of N-protected nonnatural l- and d-biarylalanine derivatives. Starting from 4-bromocinnamic acid and 4-bromophenylpyruvic acid using a phenylalanine ammonia lyase (PAL) and an evolved d-amino acid dehydrogenase (DAADH), respectively, both enantiomers of 4-bromophenylalanine were obtained and subsequently coupled with a panel of arylboronic acids to give the target compounds in high yield and optical purity under mild aqueous conditions.

Full text of DOI:10.1021/acscatal.5b01132

Letter
pubs.acs.org/acscatalysis
Chemoenzymatic Synthesis of Optically Pure L- and D‑Biarylalanines
through Biocatalytic Asymmetric Amination and Palladium-
Catalyzed Arylation
Syed T. Ahmed, Fabio Parmeggiani, Nicholas J. Weise, Sabine L. Flitsch, and Nicholas J. Turner*
School of Chemistry, Manchester Institute of Biotechnology (MIB), University of Manchester, 131 Princess Street, M1 7DN,
Manchester, United Kingdom
*
S Supporting Information
ABSTRACT: A chemoenzymatic approach was developed and optimized for the
synthesis of a range of N-protected nonnatural L- and D-biarylalanine derivatives.
Starting from 4-bromocinnamic acid and 4-bromophenylpyruvic acid using a
phenylalanine ammonia lyase (PAL) and an evolved D-amino acid dehydrogenase
(
DAADH), respectively, both enantiomers of 4-bromophenylalanine were obtained
and subsequently coupled with a panel of arylboronic acids to give the target
compounds in high yield and optical purity under mild aqueous conditions.
KEYWORDS: biocatalysis, biarylalanines, amino acids, amination, Suzuki-Miyaura coupling, cascade reactions
7
n increasing number of drugs in development contain
nonnatural amino acids in their core motif; in particular,
14 motor protein KIFC1 inhibitors, and reverse cholesterol
transport facilitators.
8
A
Although a number of chemical methods have been
described in the literature for the synthesis of nonnatural
amino acids, to date, no chemoenzymatic approach to
biarylalanines has been reported. Herein, we present two
efficient biocatalytic strategies (hydroamination of 4-bromocin-
namic acid 3 and reductive amination of 4-bromophenylpyruvic
acid 4) for the preparation of both enantiomers of 4-
bromophenylalanine 2, followed by protection and Suzuki−
Miyaura coupling with a panel of arylboronic acids, to afford
the target compounds L- and D-1 (Scheme 1).
the biarylalanine moiety (Figure 1). For example, L-
biarylalanines have been incorporated in dipeptidyl peptidase
9
1
2
4
(DPP IV) inhibitors, α4β7 integrin inhibitors, viral 3C-
3
protease inhibitors, and endothelin-converting enzyme inhib-
itors, and D-enantiomers are featured in botulinum toxin
inhibitors, amyloid-β-peptide aggregation inhibitors, kinesin-
4
5
6
Scheme 1. Chemoenzymatic Approach to L- or D-1
One of the most attractive biocatalytic routes to optically
pure L-arylalanines is the asymmetric hydroamination of
arylpropenoic acids catalyzed by phenylalanine ammonia lyases
10
(
PALs), with 100% atom economy and no need for cofactor
regeneration systems. In nature, PAL catalyzes the deamination
11
of L-phenylalanine to cinnamic acid, and the reverse reaction
Received: June 1, 2015
Revised: August 7, 2015
Figure 1. Patented pharmaceuticals containing L- and D-biarylalanines
as chiral building blocks.
©
XXXX American Chemical Society
5410
DOI: 10.1021/acscatal.5b01132
ACS Catal. 2015, 5, 5410−5413

ACS Catalysis
Letter
+
has been exploited in the synthesis of novel L-phenylalanine
analogues from substituted cinnamic acids.
free extract (10 mM 4, 200 mM NH , 100 mM carbonate
4
12
buffer, pH 9.0) using methanol as a cosolvent and the glucose/
glucose dehydrogenase (GDH) system for the regeneration of
the NADPH cofactor, giving complete conversion to D-2 with
>99% ee by HPLC.
Remarkably, the DAADH-catalyzed reductive amination of 4-
phenylphenylpyruvic acid under the same conditions (see
Supporting Information) afforded 82% conversion, proving
once again the broad substrate spectrum of this engineered
biocatalyst. However, the strategy based on the synthesis of D-2
as a gateway intermediate to a wide range of coupled products
is more efficient and versatile than optimizing the biotransfor-
mation conditions for each different biaryl substrate.
The biotransformation of 4-bromocinnamic acid 3 was tested
under the standard conditions for PAL aminations (5 mM 3, 5
M aqueous ammonia, pH 9.6) using three different wild-type
PALs overproduced in Escherichia coli whole cells: PcPAL from
Petroselinum crispum (parsley), RgPAL from the red yeast
Rhodotorula glutinis and AvPAL from the cyanobacterium
Anabaena variabilis. The final conversions obtained are reported
in Table 1, with the best results being provided by AvPAL,
a
Table 1. PAL Amination of 3
a
PAL variant
conv (%)
Having access to both enantiomers of 2 through different
biocatalytic routes, we turned our attention to the development
of the arylation step. Three palladium catalysts (5−7) were
PcPAL
44
67
71
72
75
80
RgPAL
AvPAL
15−17
shown to be active under aqueous conditions (Scheme 2),
AvPAL-F107L
AvPAL-F107I
AvPAL-F107A
and to investigate their suitability, a model reaction was set up
between 4-bromobenzoic acid 4 and phenylboronic acid 9a.
aExperimental conditions: 25 mg mL−1 lyophilized E. coli cells
Scheme 2. Catalysts for Aqueous Suzuki Coupling
producing PAL, 5 mM 3, 5 M NH OH, pH 9.6, 37 °C, 18 h.
4
a
Determined by HPLC.
which has recently emerged as a promising candidate for
preparative-scale biotransformations because of its broader
1
3
substrate scope.
Catalyst 5, in a water/ethanol mixture as solvent, gave a good
yield of compound 10 (Table 2, entry 1), whereas in neat
water, the conversion dropped considerably to 24% (entry 2).
Phenol and biphenyl were identified as the predominant side
products; however, with exclusion of oxygen, the conversion
increased to 36% (entry 3). In an oxidative environment,
palladium can react with oxygen to form a palladium−peroxo
intermediate, which catalyzes the homocoupling and the
oxidation of the boronic acid.20 By increasing catalyst
concentration and temperature, up to 86% yield of 10 in neat
water was obtained (entry 6) without the need for a nitrogen
atmosphere.
Analysis of the active site of AvPAL suggested that the
conversion might be improved by reducing the steric clash
between the bromine atom and the residue F107 in the
aromatic binding pocket (Figure 2). We therefore designed
Davis et al. have reported the use of catalyst 6 in the
bioconjugation of proteins and peptides via a Suzuki−Miyaura
18
cross-coupling reaction; however, the reaction with 6 was
slow, and a maximum yield of 41% was obtained (entry 7). We
applied the best conditions from catalyst 5 to catalyst 6 (entry
Figure 2. Active site model of 3 bound to AvPAL (MIO = 4-
methylideneimidazol-5-one).
8
): although the yield increased, it was still lower than with 5.
variants in which F107 was mutated to smaller hydrophobic
residues, that is, F107I, F107L, and F107A. For all three
variants, the conversions were found to be higher compared
with WT (Table 1), and the highest value was obtained with
the variant F107A. HPLC analysis on a chiral stationary phase
showed that the phenylalanine product L-2 was obtained with
Catalyst 7 was completely insoluble under the conditions tested
(entry 9), so it was not pursued any further. Na CO as a less
expensive alternative to Cs CO3 (entry 10) gave lower
conversion. To reduce the reaction times (24 h), microwave
irradiation was employed, affording similar yields after only 20
min (entry 11).
2
3
2
>
99% ee.
The optimized conditions from the model reaction (entry
11) were applied to the coupling of N-Boc-4-bromo-L-
phenylalanine, L-11, with 9a, affording the final biaryl product,
L-1a, in 93% conversion and >99% ee by HPLC (Scheme 3).
We also tested the PAL-catalyzed hydroamination of 4-
phenylcinnamic acid under the same conditions (see
Supporting Information). No conversion was observed, proving
that our chemoenzymatic approach is essential for the PAL-
mediated synthesis of L-2.
To access D-2, we selected an NADPH-dependent D-amino
acid dehydrogenase (DAADH, engineered from a meso-
diaminopimelate dehydrogenase from Corynebacterium gluta-
micum), which has been previously reported to be a very
effective catalyst for the reductive amination of a wide range of
Addition of Boc O after the PAL biotransformation formed
2
the desired product L-11, but also side products tert-butyl
carbamate 12 (by reaction with free ammonia) and 4-
phenylcinnamic acid 13 (from the coupling between 3 and
9a). The coupling reaction performed in the presence of
varying amounts of 12 and 3 led to lower yields (Figure S1a,
Supporting Information). Furthermore, protection and cou-
pling attempts on the crude PAL biotransformation mixture
yielded no product due to the high ammonia concentration in
14
aliphatic and aromatic α-ketoacids. The reductive amination
of 4-bromophenylpyruvic acid 4 was tested with DAADH cell-
5
411
DOI: 10.1021/acscatal.5b01132
ACS Catal. 2015, 5, 5410−5413

ACS Catalysis
Letter
Table 2. Optimisation of the Reaction between 8 and 9a under Aqueous Conditions
a
entry
cat. (mol %)
solvent
conditions
4 h, air
temp (°C)
base
conv (%)
1
2
3
4
5
6
7
8
9
5 (2)
H O/EtOH 2:1
50
50
50
50
50
80
50
80
50
80
120
Cs CO3
62
24
36
71
43
86
41
56
< 5
68
84
2
2
5 (2)
H O
4 h, air
Cs CO3
2
2
5 (2)
H O
4 h, N2
Cs CO3
2
2
5 (10)
5 (10)
5 (10)
6 (4)
H O
4 h, N2
Cs CO3
2
2
H O
4 h, air
Cs CO3
2
2
H O
24 h, air
4 h, air
Cs CO3
2
2
H O
KP buffer pH 7.5
2
i
6 (10)
7 (2−10)
5 (10)
5 (10)
H O
24 h, air
4 h, air
Cs CO3
2
2
H O
Cs CO3
2
2
1
0
1
H O
24 h, air
20 min, air, MW
Na CO3
2
2
1
H O
Cs CO3
2
2
a
Determined by HPLC.
Scheme 3. One-Pot Protection and Arylation of L-2
Therefore, for the PAL-mediated synthesis of L-1a, the
removal of ammonium salts and unreacted 3 from the reaction
mixture was performed by adsorption on an ion-exchange resin,
affording quantitative recovery of pure L-2 ready for use in the
following step. Boc protection and cross-coupling could be
performed in one pot to afford compound L-1a (>99% ee) in
7
0% isolated yield, resulting in an overall yield of 53% from 3
(
Table 3).
In the case of the DAADH biotransformation, the
substantially lower ammonia concentration and the complete
consumption of the starting material prevent competing side
reactions and catalyst deactivation. Therefore, it was possible to
successfully run the whole sequence as one-pot system, giving
the buffer, as demonstrated by control experiments with
increasing NH4 concentration (Figure S1b).
+
a
b
Table 3. Chemoenzymatic Synthesis of Compounds L-1a−k and D-1a−k
L-1a−k
D-1a−k
conv. from L-2
isol. yield from L-2
overall isol. yield from 3
overall conv. from 4
overall isol. yield from 4
g
c
d
e
f
product
R
(%)
(%)
(%)
(%)
(%)
1
1
1
1
1
1
1
1
1
1
1
a
b
c
d
e
f
H
75
99
65
85
95
80
65
99
99
99
82
70
81
55
70
85
58
61
89
90
80
51
53
60
41
40
63
33
42
65
64
58
39
86
99
99
99
98
74
95
99
97
99
95
57
56
66
64
66
47
62
61
68
70
40
2-F
3-F
4-F
4-Cl
3-Cl
2-MeO
3-MeO
4-MeO
g
h
i
j
3,4-methylenedioxy
4-Ph
k
a
b
c
Via hydroamination followed by one-pot protection and coupling. Via one-pot reductive amination, protection and coupling. Analytical
d
e
f
conversion of L-2 to L-1 (determined by HPLC). Isolated yield of L-1 from L-2. Overall isolated yield of L-1 from 3. Analytical conversion of 4 to
D-1 (determined by HPLC). Overall isolated yield of D-1 from 4.
g
5
412
DOI: 10.1021/acscatal.5b01132
ACS Catal. 2015, 5, 5410−5413

ACS Catalysis
Letter
D-1a (>99% ee) with overall conversion of 86% and isolated
yield of 57% from 4 (Table 3).
(3) Prior, A. M.; Kim, Y.; Weerasekara, S.; Moroze, M.; Alliston, K.
R.; Uy, R. A. Z.; Groutas, W. C.; Chang, K.-O.; Hua, D. H. Bioorg. Med.
Chem. Lett. 2013, 23, 6317−6320.
To demonstrate the generality of our approach, using the
optimized conditions for the chemoenzymatic synthesis of L-
and D-1a, we employed a panel of substituted phenylboronic
acids 9b−k to afford L- and D-biarylalanine derivatives L- and D-
(
4) De Lombaert, S.; Fink, C. A.; Firooznia, F.; Hoyer, D. W.; Jeng,
A. Y. U.S. Patent 6,613,782 B2, September 2, 2003.
5) Roques, B.; Anne, C.; Turcaud, S.; Fournie-Zalunski, M.-C. U.S.
Patent 2004/0176333 A1, September 9, 2004.
6) Taniguchi, A.; Sasaki, D.; Shiohara, A.; Iwatsubo, T.; Tomita, T.;
(
1
b−k in high yield and >99% ee (Table 3).
(
As an example of the practical relevance of these building
Sohma, Y.; Kanai, M. Angew. Chem., Int. Ed. 2014, 53, 1382−1385.
(7) Yang, B.; Lamb, M. L.; Zhang, T.; Hennessy, E. J.; Grewal, G.;
Sha, L.; Zambrowski, M.; Block, M. H.; Dowling, J. E.; Su, N.; Wu, J.;
Deegan, T.; Mikule, K.; Wang, W.; Kaspera, R.; Chuaqui, C.; Chen, H.
J. Med. Chem. 2014, 57, 9958−9970.
blocks, we exploited our chemoenzymatic approach to L-1d in
1
the synthesis of the DPP IV inhibitor 15 (Scheme 4) in 30%
overall yield from 3.
(
8) Sircar, J. C.; Mencel, J. J. U.S. Patent 2007/0105783 A1, May 10,
007.
9) (a) Asymmetric Synthesis and Application of α-Amino Acids;
Soloshonok, V. A., Izawa, K., Eds.; American Chemical Society:
Washington, DC, 2009; pp 2−417. (b) Groger, H.; Dietz, F. R.
Encyclopedia of Chemical Biology; John Wiley & Sons, Inc.: Weinheim,
008; pp 1−12. (c) Bommarius, A. S.; Schwarm, M.; Karlheinz, D.
Chimia 2001, 55, 50−59.
10) (a) Turner, N. J. Curr. Opin. Chem. Biol. 2011, 15, 234−240.
b) Heberling, M. M.; Wu, B.; Bartsch, S.; Janssen, D. B. Curr. Opin.
Scheme 4. Chemoenzymatic Synthesis of DPP IV Inhibitor
2
(
1
5
̈
2
(
(
Chem. Biol. 2013, 17, 250−260. (c) Louie, G. V.; Bowman, M. E.;
Moffitt, M. C.; Baiga, T. J.; Moore, B. S.; Noel, J. P. Chem. Biol. 2006,
1
3, 1327−1338. (d) Cooke, H. A.; Christianson, C. V.; Bruner, S. D.
Curr. Opin. Chem. Biol. 2009, 13, 460−468. (e) Wanninayake, U.;
Walker, K. D. J. Am. Chem. Soc. 2013, 135, 11193−11204.
(
In summary, we designed a green, efficient route to a range of
biarylalanines through the marriage of two enantiocomple-
mentary enzymatic transformations with a combinatorial
chemocatalytic coupling. The modular independence of the
bio- and chemocatalytic conversions shown here may be more
broadly applicable in the field of medicinal chemistry, allowing
similar expansion to the product range of other biocatalysts.
11) Weisshaar, B.; Jenkins, G. I. Curr. Opin. Plant Biol. 1998, 1,
2
(
51−257.
12) (a) Paizs, C.; Katona, A.; Retey, J. Chem. - Eur. J. 2006, 12,
2
739−2744. (b) De Lange, B.; Hyett, D. J.; Maas, P. J. D.; Mink, D.;
Van Assema, F. B. J.; Sereinig, M.; De Vries, A. H. M.; De Vries, J. G.
ChemCatChem 2011, 3, 289−292.
(
J. P.; Moore, B. S. Biochemistry 2007, 46, 1004−1012. (b) Wang, L.;
Gamez, A.; Archer, H.; Abola, E. E.; Sarkissian, C. N.; Fitzpatrick, P.;
Wendt, D.; Zhang, Y.; Vellard, M.; Bliesath, J.; Bell, S. M.; Lemontt, J.
F.; Scriver, C. R.; Stevens, R. C. J. Mol. Biol. 2008, 380, 623−635.
13) (a) Moffitt, M. C.; Louie, G. V.; Bowman, M. E.; Pence, J.; Noel,
ASSOCIATED CONTENT
Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acscatal.5b01132.
■
*
S
(
2
c) Lovelock, S. L.; Lloyd, R. C.; Turner, N. J. Angew. Chem., Int. Ed.
014, 53, 4652−4656. (d) Lovelock, S. L.; Turner, N. J. Bioorg. Med.
Chem. 2014, 22, 5555−5557.
14) Vedha-Peters, K.; Gunawardana, M.; Rozzell, J. D.; Novick, S. J.
Figure S1, experimental section, and copies of spectra
(
PDF)
(
J. Am. Chem. Soc. 2006, 128, 10923−10929. (b) Rozzell, D.; Novick, S.
AUTHOR INFORMATION
Corresponding Author
E-mail: nicholas.turner@manchester.ac.uk.
■
J. U.S. Patent WO 2006/113085 A2, October 26, 2006.
(
15) (a) Ganesamoorthy, S.; Shanmugasundaram, K.; Karvembu, R. J.
Mol. Catal. A: Chem. 2013, 371, 118−124. (b) Burda, E.; Hummel, W.;
Groger, H. Angew. Chem., Int. Ed. 2008, 47, 9551−9554.
(16) (a) Chalker, J. M.; Wood, C. S. C.; Davis, B. G. J. Am. Chem. Soc.
*
Notes
̈
The authors declare no competing financial interest.
2
009, 131, 16346−16347. (b) Spicer, C. D.; Triemer, T.; Davis, B. G.
ACKNOWLEDGMENTS
This work was funded by the Biotechnology and Biological
Sciences Research Council (BBSRC) and Glaxo-SmithKline
J. Am. Chem. Soc. 2012, 134, 800−803.
(
■
17) Yamamoto, K.; Watanabe, M.; Ideta, K.; Mataka, S.; Thiemann,
T. Z. Naturforsch. B 2005, 60, 1299−1307.
18) Adamo, C.; Amatore, C.; Ciofini, I.; Jutand, A.; Lakmini, H. J.
Am. Chem. Soc. 2006, 128, 6829−6836.
(
(
GSK) under the Strategic Longer and Larger (sLoLa) grant
initiative (ref BB/K00199X/1). N.J.W. was supported by the
European Union’s 7th Framework program FP7/2007-2013
under Grant Agreement No. 289646 (KYROBIO). We thank
the Royal Society for a Wolfson Research Merit Award (N.J.T.)
and Dr. Emma Jones from GSK.
REFERENCES
■
(
1) Qiao, L.; Baumann, C. A.; Crysler, C. S.; Ninan, N. S.; Abad, M.
C.; Spurlino, J. C.; DesJarlais, R. L.; Kervinen, J.; Neeper, M. P.;
Bayoumy, S. S.; Williams, R.; Deckman, I. C.; Dasgupta, M.; Reed, R.
L.; Huebert, N. D.; Tomczuk, B. E.; Moriarty, K. J. Bioorg. Med. Chem.
Lett. 2006, 16, 123−128.
(
2) Papst, S.; Noisier, A. F. M.; Brimble, M. A.; Yang, Y.; Krissansen,
G. W. Bioorg. Med. Chem. 2012, 20, 5139−5149.
5
413
DOI: 10.1021/acscatal.5b01132
ACS Catal. 2015, 5, 5410−5413