Telescopic one-pot condensation-hydroamination strategy for the synthesis of optically pure L-phenylalanines from benzaldehydes

Article abstract of DOI:10.1016/j.tet.2015.12.063

A chemo-enzymatic telescopic approach was designed for the synthesis of L-arylalanines in high yield and optical purity, starting from commercially available and inexpensive substituted benzaldehydes. The method exploits a chemical Knoevenagel–Doebner condensation (optimised to give complete conversions in a short reaction time, employing microwave irradiation) and a biocatalytic phenylalanine ammonia lyase mediated hydroamination (for the stereoselective addition of ammonia). The two reactions can be run sequentially in one pot, bringing together the advantages of chemical and biological catalysis. The preparative applicability was demonstrated with the synthesis of five L-dihalophenylalanines (71–84% yield, 98–99% ee) of relevance as molecular probes, for medicinal chemistry and for the synthesis of pharmaceutical ingredients.

Full text of DOI:10.1016/j.tet.2015.12.063

Accepted Manuscript
Telescopic one-pot condensation-hydroamination strategy for the synthesis of
optically pure L-phenylalanines from benzaldehydes
Fabio Parmeggiani, Syed T. Ahmed, Nicholas J. Weise, Nicholas J. Turner
PII:
S0040-4020(15)30311-2
10.1016/j.tet.2015.12.063
TET 27389
DOI:
Reference:
To appear in: Tetrahedron
Received Date: 15 October 2015
Revised Date: 20 December 2015
Accepted Date: 23 December 2015
Please cite this article as: Parmeggiani F, Ahmed ST, Weise NJ, Turner NJ, Telescopic one-pot
condensation-hydroamination strategy for the synthesis of optically pure L-phenylalanines from
benzaldehydes, Tetrahedron (2016), doi: 10.1016/j.tet.2015.12.063.
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.

ACCEPTED MANUSCRIPT
Telescopic one-pot condensation-hydroamination strategy for the synthesis
of optically pure L-phenylalanines from benzaldehydes
Fabio Parmeggiani, Syed T. Ahmed, Nicholas J. Weise, Nicholas J. Turner*
Manchester Institute of Biotechnology (MIB), School of Chemistry,
University of Manchester, 131 Princess Street, M1 7DN, Manchester (UK)
* corresponding author, e-mail: nicholas.turner@manchester.ac.uk
Graphical abstract
PAL
2
NH3
3
1
16 examples
67-89% yield
97-99% ee
Abstract
A chemo-enzymatic telescopic approach was designed for the synthesis of L-
arylalanines in high yield and optical purity, starting from commercially
available and inexpensive substituted benzaldehydes. The method exploits a
chemical Knoevenagel-Doebner condensation (optimised to give complete
conversions in a short reaction time, employing microwave irradiation) and
a biocatalytic phenylalanine ammonia lyase mediated hydroamination (for the
stereoselective addition of ammonia). The two reactions can be run
sequentially in one pot, bringing together the advantages of chemical and
biological catalysis. The preparative applicability was demonstrated with
the synthesis of five L-dihalophenylalanines (71-84% yield, 98-99% ee) of
relevance as molecular probes, for medicinal chemistry and for the
synthesis of pharmaceutical ingredients.
Keywords
Biocatalysis; Ammonia lyases; Amino acids; Chemo-enzymatic synthesis,
Enantioselective catalysis; Microwave-assisted synthesis
1

ACCEPTED MANUSCRIPT
1.
Introduction
Non-natural amino acid analogues are widespread chiral building blocks
implicated in the production of various natural and synthetic bioactive
compounds. Of particular interest are ring-substituted phenylalanine
analogues which can be incorporated into natural polypeptides to enable
novel chemical biological studies1 or used as intermediates for
peptidomimetic and small-molecule pharmaceuticals.2
Biocatalysis offers many alternatives for the synthesis of these useful
building blocks.3 For L-phenylalanine analogues, one of the best strategies
is the stereoselective hydroamination of cinnamic acid derivatives mediated
by phenylalanine ammonia lyases (PALs, EC 4.3.1.24-25).4 In nature, these
enzymes catalyse the abstraction of ammonia from the primary metabolite L-
phenylalanine to yield the corresponding acrylic acid. This can then be
incorporated into various natural products such as antibiotic compounds in
bacteria5 and phenylpropanoids (including lignin) in terrestrial plants.6
The discovery of the reversibility of the natural reaction in the presence
of high concentrations of ammonia,7 has enabled an easy and efficient
approach to L-arylalanines, starting from a range of simple prochiral
arylacrylic acids.
The vast majority of cinnamic acid derivatives both on industrial and
laboratory scale are synthesised through the simple Knoevenagel
condensation8 as modified by Doebner,9 based on the reaction of an aromatic
aldehyde and malonic acid in the presence of a base (usually a secondary
amine such as piperidine or pyrrolidine) in refluxing pyridine. The
benzylidenemalonic acid intermediate decarboxylates in situ, affording the
corresponding cinnamic acid, which can be isolated in high yield by
diluting the mixture with ice, acidifying it and filtering the crude
product. The reaction is very efficient, tolerates a wide range of
substituted aryl groups, and affords the (E)-stereoisomer of the product
with perfect diastereoselectivity.
In spite of the simplicity of this synthesis, in order to reduce the
overall processing time, minimise waste and increase the productivity of
the system, it would be desirable to combine the two synthetic steps in a
simple one-pot strategy, especially for those amino acid derivatives where
the corresponding cinnamic acids are not commercially available, but the
aldehyde is. Indeed, from a fine chemical manufacturing perspective it is
generally appealing to run more than one reaction concurrently in the same
vessel in a so-called “one-pot cascade” combination, wherever possible.
Biocatalysis is particularly suited for this purpose, since many enzymes
are compatible with each other and work under similar mild conditions:
various successful multi-enzymatic cascade systems appeared in the
2

ACCEPTED MANUSCRIPT
literature in the last decade,10 only rarely involving PALs.11 Equally
advantageous has been the integration of chemical and biocatalytic steps,12
successfully exploited for a range of different purposes, such as to shift
the equilibrium of thermodynamically limited enzymatic reactions, to
synthesise the enzyme substrate in situ from a simpler precursor, or to
diversify and further manipulate the product of a biotransformation.
In some instances, though, the extreme difference in conditions (e.g.,
temperature and solvent) make the cascade strategy not feasible. In those
situations, similar advantages can still be achieved with a so-called
“telescopic” approach, where the reagents for the second step are added
only when the first one is completed, and the second reaction is run under
conditions that are independent on the previous.
Since the conditions for the Knoevenagel-Doebner condensation and for the
PAL-mediated amination are definitely incompatible, we ruled out the
possibility to combine them as a cascade system, but we envisaged that a
telescopic strategy could be feasible. Herein, we report on our attempts to
combine the two reactions in a one-pot strategy for the conversion of a
range of available and inexpensive arylaldehydes 1 to the corresponding L-
phenylalanines L-3 in good yield and excellent optical purity (Scheme 1).
Scheme 1. Telescopic synthesis of L-phenylalanines from benzaldehydes.
2.
Results and discussion
2.1 Preliminary experiments on the feasibility of the telescopic approach
A typical Knoevenagel-Doebner condensation procedure involves heating under
reflux for a few hours the aromatic aldehyde (1 equiv.), malonic acid (3
equiv.) and a base such as piperidine or pyrrolidine (2-10 mol %) in
pyridine as a solvent. Such a procedure, applied to benzaldehyde 1a (as a
model substrate, used throughout the optimisation described below) affords
quantitative conversion to trans-cinnamic acid 2a (Scheme 2).
3

ACCEPTED MANUSCRIPT
Scheme 2. Classical Knoevenagel-Doebner condensation of 1a.
Scheme 3. Classical PAL-mediated hydroamination of 2a.
The substrate/product concentrations for the Knoevenagel-Doebner reaction
are in the molar range, while for the biocatalytic amination (Scheme 3)
concentrations of 20-50 mM are rarely exceeded. Therefore, we expected that
the PAL-mediated amination step could be conveniently telescoped by
diluting the reaction mixture with buffered ammonia solution and adding the
biocatalyst. For this purpose, we tested the crude Knoevenagel-Doebner
reaction mixture (1:50 dilution in the reaction buffer, giving a 20 mM
final concentration of 2a) as a substrate for the amination mediated by PAL
under standard conditions (5 M aqueous ammonia solution, pH adjusted to 9.6
with H2SO4, 37°C, 24 h). Two PALs from different kingdoms overproduced in
E. coli BL21(DE3) were tested as a whole-cell biocatalyst: RgPAL from the
red yeast Rhodotorula glutinis13 and AvPAL from the cyanobacterium Anabaena
variabilis.14 No conversion to L-3a was observed, even after prolonged
incubation time or with increased biocatalyst loadings. This could be
attributed to the inhibitory effect of residual pyridine: indeed, control
experiments showed that 2.5% v/v of pyridine completely suppresses the
activity of both the enzymes tested, and concentrations as low as 1.0% v/v
can considerably reduce it (Table 1).
Table 1. Effect of increasing pyridine concentrations on the amination of
cinnamic acid 2a mediated by RgPAL and AvPAL.
Pyridine conc.
[% v/v]
0
1.0
Conv. RgPAL Conv. AvPAL
[%]a
61
7
[%]a
47
4
2.5
5.0
10.0
<1
<1
<1
<1
<1
<1
Experimental conditions: 5 mM 2a, 40 mg mL-1 PAL biocatalyst (wet cells), 5
M NH4OH, pH 9.6 (adj. with H2SO4), 37ºC, 24 h. a Calculated by HPLC analysis
on a non-chiral stationary phase.
4

ACCEPTED MANUSCRIPT
A variation of the Knoevenagel-Doebner condensation in the literature calls
for replacing pyridine with DMSO,15 a safer and more environmentally
friendly solvent that is also well-known for being much better tolerated by
enzymes.16 PAL aminations generally do not require cosolvents due to the
good solubility of cinnamates in the alkaline reaction medium, therefore
little attention in the previous literature has been devoted to the
cosolvent tolerance of this class of enzymes. Thus, we first investigated
the influence of varying amounts of DMSO on the conversion of cinnamic acid
2a to phenylalanine 3a by RgPAL and AvPAL under standard conditions (Figure
1).
80
2 h
RgPAL
24 h
60
40
20
0
0
2.5
5
10
20
30
40
DMSO conc. [% v/v]
80
60
40
20
0
2 h
AvPAL
24 h
0
2.5
5
10
20
30
40
DMSO conc. [% v/v]
Figure 1. Effect of increasing DMSO concentrations on the amination of
cinnamic acid 2a mediated by RgPAL and AvPAL. Experimental conditions: 5 mM
2a, 40 mg mL-1 PAL biocatalyst (wet cells), 5 M NH4OH, pH 9.6 (adj. with
H2SO4), 37ºC.
Remarkably, both enzymes appeared to tolerate considerable amounts of this
cosolvent: high concentrations lead to lower conversions but the reaction
still proceeded with a significant drop only over a concentration of 20%
v/v. For RgPAL, addition of DMSO up to 10% v/v did not change the overall
reaction rates appreciably. For AvPAL, the initial rates did not seem to
drop up to 20-30% v/v, giving after 2 h almost the same conversions as
achieved with no cosolvent. On the other hand, the stability of the
5

ACCEPTED MANUSCRIPT
biocatalyst seemed to be affected, since the final conversions after 24 h
became progressively lower with 20-40% v/v of cosolvent. Nevertheless, the
modest decrease in activity of both PALs in the presence of substantial
amounts of DMSO (as opposed to what observed with pyridine) proved that the
telescopic approach outlined before could, in principle, be feasible.
To test this strategy, the same Knoevenagel-Doebner condensation was
repeated in the presence of DMSO instead of pyridine, according to
literature conditions,15 affording again quantitative conversion of 1a to
2a. The resulting mixture was then diluted with ammonia solution and
incubated with PAL as described above for the reaction in pyridine. In this
case, PAL activity was almost fully retained, affording 58 and 51%
conversion to L-3a (over 24 h) with RgPAL and AvPAL, respectively.
Having thus proven the practical applicability of our system, we then
turned to the optimisation of each step individually, in order to maximise
the overall efficiency of the process.
2.2 Optimisation of the Knoevenagel-Doebner condensation step
With the aim of reducing the reaction times as much as possible, we
attempted to optimise the conditions for cinnamic acid synthesis by
employing microwave heating. The uniform and rapid heating of reaction
mixtures allows faster reaction times to be achieved and in some cases even
higher yields.
A series of optimisation reactions were set up to probe the efficiency of
the Knoevenegel-Doebner condensation under microwave conditions, using 1a
as the model substrate (Table 2). Initially we tested three temperatures
(60, 90, 120°C) with a reaction time of 5 min. Surprisingly, the highest
conversion was seen with the lowest temperature (entry 1) and a similar
trend was observed when the reaction time was increased to 20 minutes
(entries 4-6). In both cases, a higher temperature resulted in lower
conversion, which we postulate is due to the thermal decomposition of DMSO
under microwave irradiation, evident from the darkening of the reaction
mixtures at higher temperatures. The instability of DMSO in the presence of
acids or bases at high temperature is known in the literature, with side
products identified as dimethyl sulphide, bis(methylthio)methane and methyl
disulphide.17 The highest temperature where no decomposition was observed
was 60°C, with a reaction time of 45 minutes giving 93% conversion (entry
8). Another possibility is that, at higher temperatures, the rate of
1
decarboxylation of malonic acid to give acetic acid (observed by H NMR)
and carbon dioxide, is increased more than that of the desired reaction,
thus resulting in overall lower consumption of the starting aldehyde.
6

ACCEPTED MANUSCRIPT
Table 2. Optimisation of the Knoevenagel-Doebner reaction for the synthesis
of 2a employing microwave heating.
Entry Time
[min]
Temp.
[ºC]
60
90
120
60
70
90
60
60
60
60
60
60
60
60
60
Base
Malonic acid Conv.
[equiv.]
1.1
1.1
1.1
1.1
1.1
1.1
1.1
1.1
1.1
1.1
2.0
2.0
2.0
3.0
2.0
2.0
2.0
[%]a
75
1
2
3
5
5
5
2 mol% piperidine
2 mol% piperidine
2 mol% piperidine
2 mol% piperidine
2 mol% piperidine
2 mol% piperidine
2 mol% piperidine
2 mol% piperidine
2 mol% piperidine
6 mol% piperidine
2 mol% piperidine
2 mol% piperidine
2 mol% piperidine
2 mol% piperidine
2 mol% Et2NH
51b
57b
87
4
5
6
7
8
9
10
11
12
13
14
15
16
17
20
20
20
30
45
60
30
30
20
10
10
20
20
20
76
52b
89
93
93
81
>99
>99
91
92
57
60
60
2% v/v conc. NH4OH
5% v/v conc. NH4OH
77c
97c
a
1
Calculated by H NMR (by integration of the singlet at 9.82 ppm for 1a
and the doublet at 6.27 ppm for 2a); b Considerable darkening and formation
of subproducts was observed; c Formation of β-phenylalanine was observed.
In light of our initial temperature/time optimisation study, two further
experiments were carried out: increasing the base from 2 mol% to 6 mol%
(entry 10) and using 2 equivalents of malonic acid (entry 11) to improve
reaction time and drive it to completion. The former gave lower conversion
of 81%, while the latter yielded full conversion. By shortening the
reaction time to 20 min, the yield did not drop (entry 12), but reducing it
to 10 min the reaction was no longer complete (entry 13), even with 3
equivalents of malonic acid (entry 14). This indicates the reaction must be
performed for at least 20 min under these conditions.
We also looked into the use of an alternative base in our screening.
Diethylamine (also a secondary amine similar to piperidine) gave a
substantially lower conversion of 57% under the same conditions of entry 12
(entry 15), possibly because of its higher volatility. Employing ammonia as
the base was also very attractive due to it being already a component
required for the second step of the telescopic process: a conversion of 97%
was measured using concentrated ammonium hydroxide instead of piperidine
(entry 17). However, in this case, we also observed the formation of a
white precipitate (not seen with diethylamine or piperidine), that was
isolated and found to be mostly composed of β-phenylalanine (confirmed by
1H NMR and HPLC), drastically lowering the overall yield of cinnamic acid.
Indeed, a high-yield synthesis of aromatic β-amino acids from the same
starting materials uses similar conditions with ammonium formate or acetate
7

ACCEPTED MANUSCRIPT
as the ammonia source.18 Employing ammonium hydroxide, it is not surprising
to observe this competing side reaction.
Thus, in order to achieve complete conversions and avoid the formation of
side-products such as the β-amino acid, we reverted back to piperidine as
the base, leading to the following optimised conditions: 60°C, 20 min, 2
mol% piperidine, 2.0 equiv. malonic acid.
2.3 Optimisation of the PAL-mediated hydroamination step
The standard buffer for PAL-mediated aminations consists of aqueous ammonia
solution (5 M) brought to pH 9.6 with sulphuric acid. In the attempt to
maximise the conversion of the biocatalytic reaction on the raw mixture
from the previous step in DMSO (>99% conv. to 2a), we screened several
other ammonium sources at different concentrations and pH values (Table 3).
Even though good results were obtained with the standard buffer,
considerably higher conversions were achieved with a more concentrated
solution (13% w/v, approximately 7 M) adjusted to pH 10.0 with CO2
(previously employed in a large-scale application of PAL reactions7c). In
contrast, saturated or highly concentrated solutions of ammonium salts
(used for similar reactions with related aminomutase enzymes19) perfomed
poorly, likely due to the combined effect of the higher ionic strength and
the lower pH values.
Table 3. Screening of different buffers for the PAL-mediated hydroamination
of 2a in the raw Knoevenagel-Doebner condensation mixture.
Reaction buffer
Conv. RgPAL Conv. AvPAL
[%]a
59
81
3
[%]a
48
55
7
5 M NH4OH (adj. with H2SO4), pH 9.6
7 M NH4OH (adj. with CO2), pH 10.0
sat. (NH4)2SO4, pH 8.2
sat. NH4Cl, pH 8.4
2.5 M (NH4)2CO3, pH 9.0
16
56
13
37
Experimental conditions: 1:50 dilution of Knoevenagel-Doebner mixture (20
a
mM final conc. 2a), 40 mg mL-1 PAL biocatalyst (wet cells), 37ºC, 24 h.
Calculated by HPLC analysis on a non-chiral stationary phase.
In addition, we investigated the maximum substrate loading practically
feasible for the hydroamination to occur in good yields: different volume
ratios of the condensation mixture and ammonia buffer were combined and
incubated with different concentrations of AvPAL (Figure 2). RgPAL afforded
high conversions up to 1:10 dilution (100 mM 2a), while AvPAL showed a
gradual decrease in efficiency with increasing ratios. In order to avoid a
significant drop in conversions and to keep the substrate loading as high
as possible, a 1:20 dilution ratio was chosen as a trade-off.
8

ACCEPTED MANUSCRIPT
100
80
RgPAL
AvPAL
60
40
20
0
Vol. ratio
Subs. conc.
DMSO conc. 2% v/v
1:50
20 mM
1:20
50 mM
5% v/v
1:10
1:5
200 mM
20% v/v
100 mM
Volume ratios
10% v/v
Figure 2. Effect of increasing the condensation mixture/buffer ratio on the
amination of cinnamic acid 2a mediated by RgPAL and AvPAL. Experimental
conditions: 40 mg mL-1 PAL biocatalyst (wet cells), 7 M NH4OH, pH 10.0 (adj.
with CO2), 37ºC, 24 h.
To conclude our optimisation study, we performed a time course experiment
to select the most reasonable biotransformation time for further screening
(Figure 3). RgPAL reached the thermodynamic limit for the reaction in
approximately 2 h, while AvPAL did not get to the same level even after 24
h. Unsubstituted 2a is known to be a better substrate for RgPAL than for
AvPAL, but the latter enzyme showed a much broader substrate scope and
higher activity on more interesting non-natural substrates.20 Therefore, we
decided to perform the screening-scale biotransformations with both enzymes
for 24 h.
100
80
60
40
RgPAL
AvPAL
20
0
0
5
10
15
20
25
Time [h]
Figure 3. Time course of the amination of cinnamic acid 2a (from
Knoevenagel-Doebner condensation) mediated by RgPAL and AvPAL. Experimental
conditions: 1:20 dilution of condensation mixture, 40 mg mL-1 PAL
biocatalyst (wet cells), 7 M NH4OH, pH 10.0 (adj. with CO2), 37ºC.
The enantiomeric excess value of L-3a was >99% in both cases (by HPLC on a
chiral stationary phase), proving that DMSO and the other additional
components do not affect the enantioselectivity of PAL amination of 2a.
9

ACCEPTED MANUSCRIPT
2.4 Extension to known and unreported substrates
The general applicability of our strategy was demonstrated by testing a
panel of halogenated aldehydes 1b-p as starting materials, to afford the
corresponding halocinnamic acids 2b-p, that were converted in a telescopic
fashion into the corresponding L-halophenylalanines L-3b-p (Table 4).
Aldehydes 1b-i were employed to validate our methodology, since the
corresponding cinnamic acids are known to be substrates for PALs,7,11 while
aldehydes 1j-p were investigated to expand the substrate scope (the
amination of several difluorocinnamic acids with a plant PAL has been
reported in the literature,21 but substrates 2j-p have not been tested
before, to the best of our knowledge). Quantitative conversions were
obtained in all cases for the Knoevenagel-Doebner step, and good to very
high conversions with almost all substrates for the PAL hydroamination step
(indeed the latter were almost identical to those achieved with the pure
halocinnamic acid as substrate under cosolvent-free conditions).
Table 4. Evaluation of the substrate scope of the telescopic chemo-
enzymatic synthesis of L-phenylalanines L-3a-p from benzaldehydes 1a-p.
Substrate
R
RgPAL
AvPAL
ee L-3
ee L-3
Conv.
[%]a
84
Conv.
[%]a
41
93
93
65
96
70
88
69
41
93
95
95
55
97
94
[%]b
>99
>99
>99
>99
98
[%]b
>99
>99
>99
>99
98
1a
1b
1c
1d
1e
1f
1g
1h
1i
1j
1k
1l
1m
1n
1o
1p
H
2-F
3-F
4-F
2-Cl
3-Cl
4-Cl
2-Br
93
92
39
97
94
21
96
4
22
<1
43
73
10
90
25
98
>99
98
>99
>99
99
98
99
99
97
>99
98
c
4-Br
-
3,4-F2
2,4-Cl2
3,4-Cl2
3,5-Cl2
2-Cl-4-F
3-Cl-4-F
4-Cl-3-F
99
c
-
98
>99
c
-
>99
>99
97
c
-
94
Experimental conditions: 1:20 dilution of condensation mixture (optimised
conditions), 40 mg mL-1 PAL biocatalyst (wet cells), 7 M NH4OH, pH 10.0
(adj. with CO2), 37ºC, 12 h. a Calculated by HPLC analysis on a non-chiral
b
stationary phase;
Calculated by HPLC analysis on a chiral stationary
phase; c - = not determined.
10

ACCEPTED MANUSCRIPT
Monohalogenated substrates gave similar results to those previously
reported with other PALs (some with higher conversion), and all the new
substrates were converted to some extent by at least one of the enzymes
with good yields. In general, higher conversions were more often obtained
with AvPAL, proving once again its broad substrate tolerance and its better
suitability as a biocatalyst for synthetic applications.
It should be mentioned that cinnamic acids 2n-p are not commercially
available (from major suppliers), and those are the cases where our
strategy turns out to be more valuable, since it easily allows access to
optically pure L-phenylalanines without the need to isolate the
intermediate.
To demonstrate the practical applicability of our strategy, we performed a
preparative scale synthesis of five non-commercially available
dihalophenylalanines (L-3j-k, n and o), which were isolated in pure form by
adsorption on ion-exchange resin in 72-84% overall yield starting from the
aldehyde (Scheme 4). These compounds, besides being useful molecular probes
for CH/π interactions22 are relevant building blocks for medicinal chemistry
and drug discovery, and have been used in the synthesis of peptidomimetic
or small-molecule pharmaceuticals. For instance, L-3j has been incorporated
in selective inhibitors of individual nuclear hormone receptors,23 L-3k in
cholecystokinin B antagonists for the treatment of CNS disorders24 and L-3o
in indolizinone antivirals.25
Scheme 4. Preparative scale telescopic chemo-enzymatic conversion of five
dihalobenzaldehydes into L-dihalophenylalanines for medicinal chemistry
applications.
Furthermore, by monitoring the course of the reactions over time, we
observed that a reaction time of 2-4 h was sufficient to achieve conversion
values similar to those obtained on analytical scale after 24 h. This
demonstrates that halogenated substrates are turned over much more quickly
than 2a by AvPAL, with clear benefit for the productivity of the process,
since the preparative scale reactions from the aldehyde to the isolated
pure L-amino acid product could be run in less than 6 h.
11

ACCEPTED MANUSCRIPT
3.
Conclusion
Telescoping reactions steps has several advantages in chemical synthesis,
from the reduction of processing time to the minimisation of waste
effluents. In this paper, we proposed an efficient route to enantiopure
non-natural L-amino acid derivatives from simple precursors (benzaldehydes,
malonic acid and ammonia) using a tandem chemical condensation and
asymmetric biocatalytic hydroamination. The apparent incompatibility of the
differing reaction conditions required has been addressed via cosolvent
tolerance studies of the candidate phenylalanine ammonia lyase
biocatalysts. This has subsequently enabled optimisation of both steps for
a telescoped synthesis via dilution of the crude chemical reaction mixture
in the biotransformation buffer.
The ease and generality of this method allows the expansion of the L-amino
acids range attainable through PAL-mediated amination, by accessing the
diversity of commercially available substituted benzaldehyde starting
materials. As a representative example, we provided a high-yield optimised
route to five optically pure L-dihalophenylalanines that are incorporated
in several candidate active pharmaceutical ingredients.
4.
Experimental
4.1 General
Analytical grade reagents and solvents were obtained from Sigma-Aldrich,
AlfaAesar or Fisher Scientific and used without further purification,
unless stated otherwise.
13
1H and C NMR spectra were recorded on a Bruker Avance 400 spectrometer
(400.1 MHz) without additional internal standard. Chemical shifts are
reported as δ in parts per million (ppm) and are calibrated against
residual solvent signal.
Reverse phase HPLC was performed on an Agilent 1200 Series LC system
equipped with a G1379A degasser, a G1312A binary pump, a G1329 autosampler
unit, a G1316A temperature controlled column compartment and a G1315B diode
array detector.
HRMS analyses were performed using an Agilent 6510 Q-TOF mass spectrometer
connected to an Agilent 1200 Series LC system.
Microwave reactions were carried out in a CEM Discoverer microwave reactor
using sealed reaction vessels. Maximum power output was set to 200 W and
maximum pressure was set to 247 psi. All microwave reactions were conducted
under stirring.
4.2 Analytical methods
The conversions of the Knoevenagel-Doebner condensations were estimated by
1H NMR, comparing the peak areas for the signal at δ = 6.0-6.5 ppm (d,
=CHCOOH) for 2a-p and the signal at δ = 9.0-9.5 ppm (s, CHO) for 1a-p.
The conversions of the PAL biotransformations were measured by HPLC on a
non-chiral reverse-phase Zorbax C-18 Extend column (50 mm × 4.6 mm × 3.5
µm, Agilent), flow rate 1.0 mL min–1, temperature 40°C, detection
wavelength 210 nm. Method 1 (mobile phase: aq. NH4OH 0.1 M pH 10 / MeOH,
9:1): tR (3a) = 1.3 min, tR (2a) = 2.9 min, tR (3b) = 1.7 min, tR (2b) = 4.5
min, tR (3c) = 1.6 min, tR (2c) = 4.1 min, tR (3d) = 1.6 min, tR (2d) = 3.7
min, tR (3j) = 2.1 min, tR (2j) = 5.0 min. Method 2 (mobile phase: aq. NH4OH
12

ACCEPTED MANUSCRIPT
0.1 M pH 10 / MeOH, 8:2): tR (3e) = 2.0 min, tR (2e) = 4.9 min. Method 3
(mobile phase: aq. NH4OH 0.1 M pH 10 / MeOH, 7:3): tR (3f) = 1.6 min, tR
(2f) = 3.8 min, tR (3g) = 2.9 min, tR (2g) = 6.9 min, tR (3h) = 2.5 min, tR
(2h) = 6.1 min, tR (3i) = 3.6 min, tR (2i) = 8.8 min, tR (3n) = 2.9 min, tR
(2n) = 7.0 min, tR (3o) = 3.3 min, tR (2o) = 8.8 min, tR (3p) = 3.9 min, tR
(2p) = 9.3 min. Method 4 (mobile phase: aq. NH4OH 0.1 M pH 10 / MeOH, 6:4):
tR (3k) = 3.1 min, tR (2k) = 7.4 min, tR (3l) = 3.5 min, tR (2l) = 8.1 min,
tR (3m) = 4.4 min, tR (2m) = 10.5 min. Response factors were measured and
taken into account in the calculation of the conversions (a: 2.3, b: 2.7,
c: 2.7, d: 2.8, e: 2.2, f: 2.1, g: 2.0, h: 1.4, i: 1.4, j: 2.9, k: 2.2, l:
2.2, m: 2.2, n: 2.4, o: 2.4, p: 2.4).
The enantiomeric excess values of L-3a-p were measured on a reverse-phase
Crownpak CR(+) column (150 mm × 4 mm × 3.5 µm, Daicel). Flow rate 1.0 mL
min–1, temperature 40°C, detection wavelength 210 nm. Method 1 (mobile
phase: aq. HClO4 1.14% w/v / MeOH 96:4): tR (D-3a) = 5.1 min, tR (L-3a) =
7.9 min, tR (D-3b) = 6.9 min, tR (L-3b) = 9.2 min, tR (D-3c) = 7.6 min, tR
(L-3c) = 10.9 min, tR (D-3d) = 8.0 min, tR (L-3d) = 9.9 min, tR (D-3j) =
10.9 min, tR (L-3j) = 13.2 min. Method 2 (mobile phase: aq. HClO4 1.14% w/v
/ MeOH 86:14): tR (D-3e) = 10.2 min, tR (L-3e) = 13.0 min; tR (D-3f) = 17.1
min, tR (L-3f) = 20.9 min; tR (D-3g) = 13.9 min, tR (L-3g) = 17.6 min; tR
(D-3h) = 13.6 min, tR (L-3h) = 16.4 min; tR (D-3i) = 18.2 min, tR (L-3i) =
23.7 min; tR (D-3k) = 33.4 min, tR (L-3k) = 42.1 min; tR (D-3l) = 46.9 min,
tR (L-3l) = 63.8 min; tR (D-3m) = 48.2 min, tR (L-3m) = 61.6 min; tR (D-3n)
= 12.6 min, tR (L-3n) = 14.7 min; tR (D-3o) = 17.0 min, tR (L-3o) = 22.8
min; tR (D-3p) = 17.2 min, tR (L-3p) = 21.5 min.
4.3 Knoevenagel-Doebner condensation for the synthesis of cinnamic acids
The suitable aldehyde (10 mmol), malonic acid (30 mmol, 3.12 g) and
piperidine (0.5 mL) were dissolved in pyridine (20 mL) and the mixture was
heated under reflux for 4 h. The solution was cooled to room temperature
and poured in ice-cold aqueous HCl (100 mL, 3 M). The white solid
precipitate was filtered, washed with water (3 × 50 mL), aqueous NaHCO3 (20
mL, 5% w/v), then again with water (2 × 50 mL) and dried in an oven (60°C).
If required, the crude solid was recrystallysed from EtOH/H2O.
4.3.1 (E)-3-(3,4-difluorophenyl)acrylic acid (2j). Yield: 1.70 g (92%). 1H
NMR (DMSO-d6, 400 MHz), δ: 7.85-7.91 (m, 1H), 7.55-7.57 (m, 1H), 7.56
(d, J=16.0 Hz, 1H), 7.43-7.50 (m, 1H), 6.57 (d, J=16.0 Hz, 1H). 13C
1
2
NMR (DMSO-d6, 101 MHz), δ: 167.30, 150.35 (dd, JCF=248 Hz, JCF=13
1
2
3
Hz), 149.64 (dd, JCF=244 Hz, JCF=13 Hz), 141.64, 132.03 (dd, JCF=6
4
3
4
Hz, JCF=4 Hz), 125.82 (dd, JCF=7 Hz, JCF=3 Hz), 120.68, 117.88 (d,
2JCF=17 Hz), 116.65 (d, 2JCF=18 Hz).
4.3.2 (E)-3-(2,4-dichlorophenyl)acrylic acid (2k). Yield: 2.04 g (94%). 1H
NMR (DMSO-d6, 400 MHz), δ: 7.92-7.97 (m, 1H), 7.79 (d, J=15.6 Hz,
1H), 7.69-7.72 (m, 1H), 7.44-7.49 (m, 1H), 6.63 (d, J=15.6 Hz, 1H).
13C NMR (DMSO-d6, 101 MHz), δ: 166.96, 137.44, 135.24, 134.32, 130.93,
129.48, 129.36, 127.94, 122.99.
4.3.3 (E)-3-(3,4-dichlorophenyl)acrylic acid (2l). Yield: 1.99 g (92%). 1H
NMR (DMSO-d6, 400 MHz), δ: 7.92-7.97 (m, 1H), 7.79 (d, J=16.3 Hz,
1H), 7.69-7.73 (m, 1H), 7.42-7.49 (m, 1H), 6.63 (d, J=16.3 Hz, 1H).
13C NMR (DMSO-d6, 101 MHz), δ: 167.21, 141.20, 135.12, 132.35, 131.70,
130.91, 129.95, 128.10, 121.57.
4.3.4 (E)-3-(2-chloro-4-fluorophenyl)acrylic acid (2n). Yield: 1.87 g
(93%). 1H NMR (DMSO-d6, 400 MHz), δ: 7.98-8.03 (m, 1H), 7.80 (d,
J=16.0 Hz, 1H), 7.51-7.58 (m, 1H), 7.24-7.32 (m, 1H), 6.59 (d,
13
J=16.0 Hz, 1H). C NMR (DMSO-d6, 101 MHz), δ: 167.08, 162.64 (d,
3
3
1JCF=253 Hz), 137.56, 134.50 (d, JCF=11 Hz), 129.99 (d, JCF=9 Hz),
4
2
128.62 (d, JCF=3 Hz), 122.18, 117.14 (d, JCF=25 Hz), 115.29 (d,
2JCF=21 Hz).
13

ACCEPTED MANUSCRIPT
4.3.5 (E)-3-(3-chloro-4-fluorophenyl)acrylic acid (2o). Yield: 1.94 g
(96%). 1H NMR (DMSO-d6, 400 MHz), δ: 7.97-8.02 (m, 1H), 7.70-7.75 (m,
1H), 7.56 (d, J=16.1 Hz, 1H), 7.42-7.48 (m, 1H), 6.59 (d, J=16.1 Hz,
13
1
1H). C NMR (DMSO-d6, 101 MHz), δ: 167.29, 157.94 (d, JCF=249 Hz),
3
3
141.36, 132.37 (d, JCF=4 Hz), 130.22, 129.01 (d, JCF=8 Hz), 120.69
(d, 4JCF=2 Hz), 120.16 (d, 2JCF=18 Hz), 117.30 (d, 2JCF=21 Hz).
4.4 Screening of different conditions for the Knoevenagel-Doebner
condensation
Benzaldehyde 1a (0.05 mol) and malonic acid (0.055-0.150 mol) were
dissolved in DMSO (50 mL) to make a stock solution (final conc. 1 M 1a).
Aliquots (1 mL) were taken, the suitable base was added and the mixture was
heated in a microwave reactor (30-60 min, 60-120°C). To check the
conversion, a sample of the solution (100 ꢀL) was diluted in CDCl3 (600 ꢀL)
and used directly for 1H NMR analysis.
4.5 Preparation of whole-cell biocatalysts
The pET16b-RgPAL and pET16b-AvPAL plasmids containing the pal genes from
Rhodotorula glutinis and Anabaena variabilis, codon-optimised for E. coli,
were obtained as described previously.11b The plasmids were transformed into
E. coli BL21(DE3) as an expression host for protein production. A single
colony of E. coli BL21(DE3) carrying the desired pET16b-PAL plasmid was
used to inoculate LB medium (8 mL) supplemented with ampicillin (100 ꢀg mL–
1), which was grown overnight at 37°C and 220 rpm. The starter culture was
used to inoculate LB-based autoinduction medium26 (800 mL, Formedium)
containing ampicillin (100 ꢀg mL–1) and glycerol (0.5% v/v) which was
incubated at 18°C and 250 rpm for 4 days. The cells were harvested by
centrifugation (5,000 rpm, 20 min, 4°C), washed with KPi buffer (100 mM, pH
8.0) and harvested again by centrifugation (5,000 rpm, 20 min, 4°C). The
cell pellet was then aliquoted and stored at –20°C.
4.6 Screening of different conditions for the PAL amination
Pure 2a or a suitable volume of the Knoevenagel-Doebner condensation
mixture in DMSO was added to the required aqueous NH4OH solution or
ammonium salt solution. Wet E.coli BL21(DE3) cells overproducing the
suitable PAL (40 mg mL–1) were added and the suspension was incubated at
37°C with shaking (180 rpm) for the required time. Biotransformation
samples (10-500 ꢀL depending on substrate concentration) were mixed with
MeOH (500 ꢀL), thoroughly shaken and centrifuged (13000 rpm, 1 min), then
the supernatant was transferred to a 0.45 ꢀm filter vial and used directly
for HPLC analysis.
4.7 Telescopic condensation-hydroamination procedure on preparative scale
The suitable aldehyde 1j-l, 1n or 1o (0.5 mmol), malonic acid (1.0 mmol)
and piperidine (0.01 mmol) were dissolved in DMSO (500 µL) and the mixture
was heated in a microwave reactor (30 min, 60°C). After cooling, ammonia
solution (9.0 mL, 13% w/v, approx. 7 M, adjusted to pH 10.0 by slow
addition of dry ice chunks) was added, followed by E. coli BL21(DE3) cells
overproducing the required PAL (400 mg wet cell paste). The suspension was
stirred at 37°C for several hours, monitoring the conversion by HPLC. For
product isolation, the reaction mixture was acidified to pH < 2.0 using
aqueous H2SO4 (20% w/v) and centrifuged (8000 rpm, 5 min) to remove cells
and precipitated unconverted substrate. Dowex® 50WX8 hydrogen form resin
(5.0 g), packed in a disposable plastic column, was washed with deionised
water (20 mL) and aqueous H2SO4 (20 mL, 5% w/v). The supernatant from the
biotransformation was loaded onto the resin (1 mL min–1). The resin bed was
washed with deionised water until the flow-through tested neutral, then the
product was eluted with aqueous NH4OH (20 mL, 5% w/v). Fractions containing
the product were pooled and dried overnight in a centrifugal evaporator, to
afford the corresponding L-phenylalanine L-3j-l, L-3n or L-3o as a white
solid.
14

ACCEPTED MANUSCRIPT
4.7.1 (S)-2-amino-3-(3,4-difluorophenyl)propanoic acid (L-3j). Yield: 84 mg
1
(84%), 99% ee by HPLC. H NMR (D2O+NaOH, 400 MHz), δ: 7.03-7.18 (m, 2H),
6.90-6.99 (m, 1H), 3.36-3.44 (m, 1H), 2.81-2.90 (dd, J=13.6,6.0 Hz, 1H),
13
2.70-2.79 (dd, J=13.6,7.2 Hz, 1H). C NMR (D2O+NaOH, 101 MHz), δ: 182.06,
1
2
1
150.94 (dd, JCF=243 Hz, JCF=12 Hz), 150.09 (dd, JCF=242 Hz, 2JCF=13 Hz),
3
4
3
4
135.33 (dd, JCF=5 Hz, JCF=3 Hz), 125.51 (dd, JCF=6 Hz, JCF=3 Hz), 117.83
2
2
(d, JCF=17 Hz), 116.97 (d, JCF=17 Hz), 57.33, 39.94. HRMS (ESI), m/z:
calcd. mass 202.0680 [M+H]+, found 202.0689 [M+H]+.
4.7.2 (S)-2-amino-3-(2,4-dichlorophenyl)propanoic acid (L-3k). Yield: 88 mg
1
(75%), 98% ee by HPLC. H NMR (D2O+NaOH, 400 MHz), δ: 7.45 (d, J=2.0 Hz,
1H), 7.16-7.26 (m, 2H), 3.43-3.52 (m, 1H), 2.95-3.06 (dd, J1=13.4,6.4 Hz,
13
1H), 2.81-2.90 (dd, J=13.4,7.6 Hz, 1H). C NMR (D2O+NaOH, 101 MHz), δ:
182.12, 134.83, 134.53, 132.43, 132.31, 128.90, 127.06, 56.40, 38.13. HRMS
(ESI), m/z: calcd. mass 234.0089 [M+H]+, found 234.0093 [M+H]+.
4.7.3 (S)-2-amino-3-(3,4-dichlorophenyl)propanoic acid (L-3l). Yield: 97 mg
1
(83%), 98% ee by HPLC. H NMR (D2O+NaOH, 400 MHz), δ: 7.21-7.26 (m, 1H),
7.17 (s, 1H), 6.90-6.98 (m, 1H), 3.21-3.27 (m, 1H), 2.68-2.75 (m, 1H),
13
2.51-2.62 (m, 1H). C NMR (D2O+NaOH, 101 MHz), δ: 181.88, 138.85, 131.31,
131.00, 130.20, 129.61, 129.11, 57.23, 39.94. HRMS (ESI), m/z: calcd. mass
234.0089 [M+H]+, found 234.0083 [M+H]+.
4.7.4 (S)-2-amino-3-(2-chloro-4-fluorophenyl)propanoic acid (L-3n). Yield:
79 mg (73%), 99% ee by HPLC. 1H NMR (D2O+NaOH, 400 MHz) 7.18-7.29 (m, 2H),
6.96-7.05 (m, 1H), 3.42-3.55 (m, 1H), 2.96-3.09 (dd, J=13.6,6.4 Hz, 1H),
2.84-2.92 (dd, J=13.6,8.0 Hz, 1H). 13C NMR (D2O+NaOH, 101 MHz), δ: 182.26,
160.00 (d, 1JCF=245 Hz), 134.26 (d, 3JCF=11 Hz), 132.46 (d, 3JCF=8 Hz), 132.04
(d, 4JCF= 3 Hz), 116.42 (d, 2JCF=25 Hz), 113.78 (d, 2JCF=21 Hz), 56.51, 37.91.
HRMS (ESI), m/z: calcd. mass 218.0384 [M+H]+, found 218.0397 [M+H]+.
4.7.5 (S)-2-amino-3-(3-chloro-4-fluorophenyl)propanoic acid (L-3o). Yield:
1
77 mg (71%), 99% ee by HPLC. H NMR (D2O+NaOH, 400 MHz), δ: 7.23-7.29 (m,
1H), 7.06-7.12 (m, 2H), 3.35-3.42 (m, 1H), 2.81-2.88 (dd, J=13.6,5.6 Hz,
13
1H), 2.68-2.79 (dd, J=13.6,7.2 Hz, 1H). C NMR (D2O+NaOH, 101 MHz), δ:
1
4
182.00, 157.79 (d, JCF= 245 Hz), 135.45 (d, JCF=3 Hz), 131.01, 129.30
(3JCF=7 Hz), 119.70 (d, JCF=18 Hz), 116.32 (d, JCF=21 Hz), 57.34, 39.77.
2
2
HRMS (ESI), m/z: calcd. mass 218.0384 [M+H]+, found 218.0385 [M+H]+.
5.
Acknowledgements
This work was funded by the Biotechnology and Biological Sciences Research
Council (BBSRC) and Glaxo-SmithKline (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). N. J. T. acknowledges the Royal Society for
a Wolfson Research Merit Award.
6.
References
1
(a) Acevedo-Rocha, C. G., Hoesl, M. G.; Nehring, S.; Royter, M.;
Wolschner, C.; Wiltschi, B.; Antranikian, G.; Budisa, N. Catal. Sci.
Technol. 2013, 3, 1198-1201; (b) Spicer, C. D.; Davis, B. G., Chem.
Commun. 2011, 47, 1698-1700; (c) Peters, F. B.; Brock, A.; Wang, J.;
Schultz, P. G. Chem. Biol. 2009, 16, 148–152.
15

ACCEPTED MANUSCRIPT
2
3
Soloshonok, V. A.; Izawa K. (Eds.), Asymmetric synthesis and application
of α-amino acids, American Chemical Society, Washington DC, 2009.
(a) Gröger, H.; Dietz, F. R. in: Encyclopedia of Chemical Biology, John
Wiley & Sons, Weinheim, 2008, pp 1−12. (b) Bommarius, A. S.; Schwarm,
M.; Karlheinz, D. Chimia 2001, 55, 50−59.
4
5
6
7
(a) Turner, N. J. Curr. Opin. Chem. Biol. 2011, 15, 234−240; (b)
Heberling, M. W.; Wu, B.; Bartsch, S.; Janssen, D. B. Curr. Opin. Chem.
Biol. 2013, 17, 250-260.
(a) Williams, J. S.; Thomas, M.; Clarke, D. J. Microbiology 2005, 151,
2543–2550; (b) Xiang, L.; Moore, B. S. J. Biol. Chem. 2002, 277, 32505–
32509; (c) Xiang, L.; Moore, B.S. J. Bacteriol. 2005, 187, 4286–4289.
(a) Ritter, H.; Schulz, G. E. Plant Cell 2004, 16, 3426–3436; (b)
Emiliani, G.; Fondi, M.; Fani, R.; Gribaldo, S. Biol. Direct 2009, 4,
7.
(a) Yamada, S.; Nabe, K.; Izuo, N.; Nkamichi, K.; Chibata, I. Appl. Env.
Microbiol. 1981, 42, 773-778; (b) Paizs, C.; Katona, A.; Retey, J.
Chem. Eur. J. 2006, 12, 2739-2744 (c) 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; (d) Ahmed, S. T.;
Parmeggiani, F.; Weise, N. J.; Flitsch, S. L.; Turner, N. J. ACS Catal.
2015, 5, 5410–5413.
8
Knoevenagel, E. Chem. Ber. 1898, 31, 2596-2619.
9
Doebner, O. Chem. Ber. 1900, 33, 2140-2142.
10
(a) Garcia-Junceda, E.; Lavandera, I.; Rother, D.; Schrittwieser, J. H.;
J. Mol. Catal. B: Enzym. 2015, 114, 1-6; (b) Ricca, E.; Brucher, B.;
Schrittwieser, J. H. Adv. Synth. Catal. 2011, 353, 2239–2262; (c)
Santacoloma, P. A.; Sin, G.; Gernaey, K. V.; Woodley, J. M. Org.
Process Res. Dev. 2011, 15, 203-212.
11
12
(a) Paisz, C.; Katona, A.; Retey, J.; Eur. J. Org. Chem. 2006, 1113-
1116; (b) Parmeggiani, F.; Lovelock, S. L.; Weise, N. J.; Ahmed, S. T.;
Angew. Chem. Int. Ed. 2015, 54, 4608–4611.
(a) Gröger, H.; Hummel, W.; Curr. Opin. Chem. Biol. 2014, 19, 171-179;
(b)Gröger, H.; Hummel, W. in: Riva, S.; Fessner W.-D. (Eds.), Cascade
Biocatalysis: Integrating Stereoselective and Environmentally Friendly
Reactions, Wiley-VCH, Weinheim, 2014, Chapter 19, pp. 427-456.
Takac, S.; Akay, B.; Ozdamar, T. H. Enzyme Microb. Technol. 1995, 17,
445-452.
13
14
(a) Moffitt, M. C.; Louie, G. V.; Bowman, M. E.; Pence, J.; Noel, J. P.;
Moore, B. S. Biochemistry, 2007, 46, 1004-1012 (b) Wang, L.; Gamez, A.;
Archer, H.; Abola, E. E.; Saekissian, C. N.; Fitzpatrick, P.; Wendt,
16

ACCEPTED MANUSCRIPT
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
Kingsbury, C. A.; Max, G.; J. Org. Chem. 1978, 43, 3131-3139.
(a) Castro, G. R.; Knubovets, T. Crit. Rev. Biotechnol. 2003, 23, 195-
231; (b) Kvittingen, L. Tetrahedron 1994, 50, 8253-8274.
Head, D. L.; McCarty, C. G. Tetrahedron Lett. 1973, 16, 1405-1408.
(a) Rodionow, W. M.; Postovskaja, E. A. J. Am. Chem. Soc. 1929, 51, 841-
847: (b) Rodionow, W. M. J. Am. Chem. Soc. 1929, 51, 847-852; (c) Tan,
C. Y. K.; Weaver, D. F. Tetrahedron 2002, 58, 7449-7461
Weise, N. J.; Parmeggiani, F.; Ahmed, S. T.; Turner, N. J. J. Am. Chem.
Soc. 2015, 137, 12977-12983.
15
16
17
18
19
20
21
Lovelock, S. L.; Turner, N. J. Bioorg. Med. Chem. 2014, 22, 5555-5557.
Gloge, A.; Zon, J.; Kovari, A.; Poppe, L.; Retey, J. Chem. Eur. J. 2000,
6, 3386-3390.
22
Fujita, T.; Nose, T.; Matsushima, A.; Okada, K.; Asai, D.; Yamauchi, Y.;
Shirasu N.; Honda, T.; Shigehiro, D.; Shimohigashi, Y. Tet. Lett. 2000,
41, 923-927.
23
24
Geistlinger, T. R.; Guy, R. K. J. Am. Chem. Soc. 2003, 125, 6852-6853.
Augelli-Szaffan, C. E.; Horwell, D. C.; Kneen, C.; Ortwine, D. F.;
Pritchard, M. C.; Purchase, T. S.; Roth, B. D.; Trivedi, B. K.; Hill,
D.; Suman-Chauhan, N.; Webdale, L. Bioorg. Med. Chem. 1996, 4, 1733-
1745.
25
26
Ruesbam, F.; Dragovich, P., U.S. Patent 0227774 A1, 2008.
Studier, F. W. Prot. Expr. Purif. 2005, 41 ,207-234.
17