Enantioselective biocatalytic formal α-amination of hexanoic acid to l-norleucine

Article abstract of DOI:10.1039/c8ob02212g

A three-step one-pot biocatalytic cascade was designed for the enantioselective formal α-amination of hexanoic acid to l-norleucine. Regioselective hydroxylation by P450_CLA peroxygenase to 2-hydroxyhexanoic acid was followed by oxidation to the ketoacid by two stereocomplementary dehydrogenases. Combination with final stereoselective reductive amination by amino acid dehydrogenase furnished l-norleucine in >97% ee.

Full text of DOI:10.1039/c8ob02212g

Organic &
Biomolecular Chemistry
View Article Online
View Journal | View Issue
COMMUNICATION
Enantioselective biocatalytic formal α-amination
of hexanoic acid to L-norleucine†
Cite this: Org. Biomol. Chem., 2018,
6, 8030
1
a
a
a
Alexander Dennig,
Somayyeh Gandomkar,
Emmanuel Cigan,
Received 7th September 2018,
Accepted 5th October 2018
a
b
a
a
Tamara C. Reiter, Thomas Haas, Mélanie Hall * and Kurt Faber
*
DOI: 10.1039/c8ob02212g
rsc.li/obc
A three-step one-pot biocatalytic cascade was designed for the cascade relies on catalytic amounts of hydrogen peroxide, air
enantioselective formal α-amination of hexanoic acid to (O ) and three enzymes for full conversion to final products,
2
4
L-norleucine. Regioselective hydroxylation by P450CLA peroxygen- liberating water as the only by-product. The handle provided
ase to 2-hydroxyhexanoic acid was followed by oxidation to the by the oxo-functionality renders these compounds ideal for
ketoacid by two stereocomplementary dehydrogenases. subsequent transformation to e.g., amino acids, aldehydes,
5
Combination with final stereoselective reductive amination by amines, or alkanes. Non-canonical amino acids (ncAAs) in
amino acid dehydrogenase furnished L-norleucine in >97% ee.
particular may be ideally obtained through amination of 2-oxo-
acids.
Upgrading of bio-based chemicals using biological methods
presents the advantage of combining the use of natural
resources as starting material and nature-derived catalysts
for performing the reaction(s) of choice. This can translate
into a major asset for commercialization of corresponding
end-products, which do not suffer from a strong chemical foot-
print and in turn benefit from a better acceptance from end-
consumers attentive to global change. In order to render such
bio-synthetic schemes efficient and environmentally sound, a
few strategies may be implemented that should result in the
maximization of reaction yield and minimization of reagents
and waste. In this regard, (artificial) linear cascade reactions
Inspired by pioneering work from Wandrey et al. on the
development of hydrogen borrowing bioprocesses to access
6
amino acids from corresponding α-hydroxy acids, and follow-
4
ing up our recent work on fatty acid oxyfunctionalization, we
envisaged a biocatalytic cascade for the formal enantio-
selective α-amination of hexanoic acid (1a) to ncAA
L-norleucine (4a, Scheme 1). The concept of hydrogen borrow-
ing by Wandrey and Kula was initially applied to amination of
DL-lactate to L-alanine through formation of pyruvate as the
intermediate, using two stereocomplementary lactate dehydro-
genases, in combination with alanine dehydrogenase for final
6
reductive amination. The protocol was then adapted to
1
appear ideal as the combination of several biocatalytic steps
L-leucine production by employing L- and D-hydroxyisocaproate
dehydrogenases (Hic-DHs; L-Hic-DH from Lactobacillus confu-
sus and D-Hic-DH from Lactobacillus casei) along with L-leucine
in sequence without the need for intermediate purification
steps significantly alleviates consumption of resources (includ-
2
ing solvents) and reduces operating time.
7
dehydrogenase (L-Leu-DH from Bacillus cereus). A similar
Saturated fatty acids are receiving increasing consideration
as a source of raw materials, which upon targeted functionali-
8
design was also applied to e.g., mandelic acid amination.
Complementary approaches for enzymatic asymmetric syn-
3
zation, grant access to a broad variety of synthons. We
9,10
thesis of ncAAs are numerous.
recently developed an atom-efficient biocatalytic oxidative
cascade for the transformation of fatty acids to corresponding
α-ketoacids via internal H O recycling. The two-step one-pot
Enantioselective formal amination of saturated fatty acids
to enantiopure α-amino acids in an enzymatic cascade may
proceed through (i) regioselective α-hydroxylation to 2-hydroxy
2
2
1
1
acids catalyzed by P450 peroxygenase, followed by (ii) dehy-
drogenation catalyzed by Hic-DHs to 2-oxo-acids and (iii) final
stereoselective reductive amination by L-Leu-DH (Scheme 1).
a
12
Department of Chemistry, University of Graz, Heinrichstrasse 28, 8010 Graz,
Austria. E-mail: melanie.hall@uni-graz.at, kurt.faber@uni-graz.at
Since the last two steps share complementary redox states of
b
Creavis, Evonik Industries, Bau 1420, Paul Baumann Strasse 1, 45772 Marl,
7
the nicotinamide cofactor, we opted for the H
2
O
2
-driven
Germany
hydroxylation catalyzed by P450 enzyme (via the so-called
†
Electronic supplementary information (ESI) available: Protocols to access
1
3
peroxide shunt ) in the first step. The reaction can also be
enzymes, procedure for enzymatic steps, analytical methods, GC-MS and HPLC
traces. See DOI: 10.1039/c8ob02212g
driven by O
2
in conjunction with NADH and a suitable electron
8030 | Org. Biomol. Chem., 2018, 16, 8030–8033
This journal is © The Royal Society of Chemistry 2018

View Article Online
Organic & Biomolecular Chemistry
Communication
Scheme 1 Linear biocatalytic cascade for the enantioselective formal α-amination of fatty acids to amino acids. P450CLA: P450 peroxygenase from
Clostridium acetobutylicum; Hic-DH: 2-hydroxyisocaproate dehydrogenase from Lactobacillus confusus (L-Hic-DH) and Lactobacillus casei (D-Hic-
DH); L-Leu-DH: leucine dehydrogenase from Bacillus cereus.
1
4
a
transport chain. However, the peroxide shunt offers the Table 2 α-Hydroxylation of fatty acids 1a–b by P450CLA
advantage of simplified reaction design by not impacting the
redox neutrality of the two combined last steps.‡
b
Entry
1a–b (10 mM)
Conv. (%)
[2a–b] (mM)
TON
Since P450 from Clostridium acetobutylicum (P450CLA
efficiently functions as peroxygenase, it was selected as the
)
1
2
1a
1b
98
>99
9.8
>9.9
3267
>3300
catalyst for the first step and could be obtained through
a
Reactions performed in Tris-HCl (pH 7.5, 100 mM) using 3 µM
recombinant expression in E. coli in a soluble pure form from P450CLA and 2.5–5% (v/v) EtOH (co-solvent) at RT and 170 rpm
4
shaking in closed glass vials. 1.6 mM H O was supplemented every
hour (total 19.2 mM). No β-hydroxylation product detected.
2 2
high yielding preparation. In contrast to the previously devel-
b
4
oped internal H
2
O
2
-regenerating system, which allowed the
use of catalytic amounts of hydrogen peroxide in a closed loop,
the oxidant should be provided in this isolated hydroxylation
step in (at least) stoichiometric amounts. Concomitant heme- (Table 2), forming corresponding 2-hydroxy acids 2a–b with
4
dependent enzyme inactivation at high peroxide concentra- high regioselectivity (>95%). The moderate coupling efficiency
1
1a
−1
tion
tation of H
can be practically avoided by fed-batch supplemen- of ∼50% (molproduct molperoxide ) justifies the excess of sup-
2 2
O
2 2
. To identify a suitable amount of peroxide for plied peroxide and is likely connected to depletion of H O
1
6
the stepwise addition, a preliminary test was conducted; one through spontaneous disproportionation. Under such reac-
portion (0.8 mM or 1.6 mM) was added to the buffer solution tion design, fed-batch supplementation proved already
containing the substrate (10 mM of 1a–b) and catalyst, and superior to batch addition of the oxidant, which could not
4
conversion of saturated fatty acids to corresponding hydroxy afford full conversion at 10 mM substrate. Notably, a TON of
acids 2a–b was monitored (Table 1). Data indicated that per- ∼3300 could be reached in the H
centage yield is reduced at higher peroxide concentration 1b by P450CLA (entry 2, Table 2).
O -driven hydroxylation of
2 2
(
compare 0.8 mM to 1.6 mM, Table 1). The reaction appears to
Since regioselective α-hydroxylation by P450CLA is poorly
4
proceed quickly, with 87% of maximum product formation enantioselective (max. 36% ee on 1b ), two stereocomple-
observed with 1a after 5 min reaction time (entry 1, Table 1). mentary enzymes are necessary to convert first intermediates
Stepwise addition of 1.6 mM every 60 min was chosen to maxi- 2a–b to 2-oxo-acids 3a–b. 2-Hydroxyisocaproate dehydrogen-
mize product concentration. Long-term pulsing (12 h) in a con- ases from Lactobacillus confusus (L-Hic-DH) and Lactobacillus
centration range usually troublesome for many heme-depen- casei (D-Hic-DH) were selected and characterized in the oxi-
1
5
17
dent proteins
was generally well-tolerated by P450CLA dative reaction. Both compounds (analytical standards of
(
Table 2). Full conversion of fatty acids (1a–b) could be rac-2a–b) were successfully oxidized by the two enzymes using
obtained (10 mM) using max. 2 mol. equivalents of H
2
O
2
substrate concentrations of 1 and 10 mM (Fig. S2, ESI†).
a
2 2
Table 1 Characterization of P450CLA in fatty acid (1a–b) hydroxylation upon one-time addition of sub-stoichiometric quantities of H O
b
c
Specific activity (U mg−1)
d
Entry
1a–b (10 mM)
Time (min)
[H
2 2
O ] (mM)
[2a–b] (mM)
Percentage yield (%)
TON
1
2
2
4
5
6
1a
1a
1a
1b
1b
1b
5
15
30
5
15
30
0.8
0.8
1.6
0.8
0.8
1.6
0.69 ± 0.06
0.71 ± 0.03
0.88 ± 0.07
0.49 ± 0.02
0.51 ± 0.01
0.94 ± 0.05
87
88
55
61
63
59
695
706
880
488
506
938
2.9
1.0
0.6
2.0
0.7
0.7
a
Reactions performed in triplicate in Tris-HCl (pH 7.5, 100 mM) using 1 µM P450CLA and 5% (v/v) EtOH (co-solvent) at RT and 170 rpm shaking
b
−1
c
in closed glass vials. molproduct molperoxide (theoretical yield controlled by peroxide as limiting reagent). Turnover number (at 0.8 and 1.6 mM
H O respectively, with 1 µM P450, TONmax = 800 and 1600, respectively). MWP450 = 48 kDa, values only relevant at the initial rate.
2 2
d
This journal is © The Royal Society of Chemistry 2018
Org. Biomol. Chem., 2018, 16, 8030–8033 | 8031

View Article Online
Communication
Organic & Biomolecular Chemistry
+
Table 3 Characterization of L- and D-Hic-DHs in dehydrogenation of genation (both Hic-DHs and NAD ) and reductive amination
a
rac-2-hydroxyacids 2a–b
(
L-Leu-DH and ammonium salt) steps (module 2) were added
at once. The reaction was investigated on 1a–b and detection
of amino acid products was performed by ninhydrin staining/
thin layer chromatography. A positive response could be only
observed in the reaction with 1a, and formation of norleucine
Concentration
(mM)
Specific
activity (U mg−
1
Entry
rac-2a–b
Enzyme
)
1
2
3
4
5
6
7
8
rac-2a
rac-2a
rac-2b
rac-2b
rac-2a
rac-2a
rac-2b
rac-2b
1
10
1
10
1
10
1
10
L-Hic-DH
L-Hic-DH
L-Hic-DH
L-Hic-DH
D-Hic-DH
D-Hic-DH
D-Hic-DH
D-Hic-DH
1.5
5.1
1.3
4.4
2.7
3.7
3.5
6.4
(
4a) was confirmed with various control reactions and GC-MS
analysis (Fig. S5, ESI†).
We finally focused on the conversion of hexanoic acid (1a)
for quantification of final product 4a, aiming at demonstrating
the feasibility of a linear one-pot cascade toward L-norleucine
(Scheme 2). Various substrate concentrations were tested
a
Reactions performed in Tris-HCl (pH 8.5, 100 mM) using 0.01
−1
+
mg mL each of L- or D-Hic-DH (freeze-dried whole cells), 10 mM NAD , (1–5 mM). Products were derivatized and analyzed by GC and
5
% (v/v) EtOH (co-solvent) and 1 or 10 mM 2a–b; activity is determined
by monitoring the increase in absorbance at 340 nm (EtOH is not oxi-
dized by Hic-DHs).
chiral HPLC (Fig. S6–S8, ESI†). The amination reaction was
confirmed to take place with high enantioselectivity, as for-
mation of L-norleucine by L-Leu-DH in perfect enantiopurity
(
ee > 97%) could be confirmed (Fig. S8, ESI†). Up to 34% con-
7
Overall, both Hic-DHs displayed good dehydrogenase activity version to the amino acid product was reached at lower sub-
−
1
at 10 mM (up to 5.1 U mg on 2a with L-Hic-DH and 6.4 strate concentration, and 1 mM 4a could be formed starting
U mg− on 2b with D-Hic-DH, entries 2 and 8, respectively, from 5 mM 1a (Scheme 2).
1
Table 3), whereas lower substrate concentration decreased
overall activity.
In the present design, despite theoretical hydride shuffling
between 2a and 4a through NAD (Scheme 1), the nicotin-
+
The basic pH required by Hic-DHs for oxidation of amide cofactor was added in stoichiometric amounts, as the
α-hydroxy acids necessitates a pH adjustment after completion reversibility of the reaction catalyzed by Hic-DHs otherwise
1
8
of the hydroxylation step (1 → 2). This was performed by favors formation of the hydroxy acid product. Nevertheless,
adding 5 M NaOH in 1 µL increments until the desired pH was compared to other linear biocatalytic cascades to non-natural
1
0c
reached. For the reductive amination step, L-Leu-DH was amino acids,
chosen, as it displays good activity on aliphatic substrates.
the sequence hydroxylation-dehydrogenation-
9
reductive amination allows access to enantiopure amino acids
Activity of recombinant L-Leu-DH on L-norleucine (4a) was con- through a simplified reaction design, with initiation of the
firmed at pH 8.5 in the deamination direction (14.5 U mL− of reaction by hydrogen peroxide, in the absence of a cofactor re-
activity determined for cell-free lysate, Fig. S3, ESI†). In cycling system otherwise needed in other cascades, such as the
addition, compatibility of the last two steps of the cascade (oxi- transformation of substituted benzenes to tyrosine deriva-
1
1
9
dation and amination) was tested (Fig. S4, ESI†), indicating tives. In comparison, oxy- and amino-functionalization of
that conversion of rac-2a to 4a proceeded smoothly at 100 mM styrene derivatives in a modular cascade remains a comp-
ammonium salt. Finally, a sequential one-pot protocol can be lementary example of linear biocatalytic transformation, which
envisaged. The cascade to final amino acid products was elegantly relies on external glucose and ammonium salt for
2
0
designed in two modules occurring in the same vessel: after access to non-natural amino acids. The recently reported
estimated completion of the hydroxylation step (module 1, combination of P450, ADH and amine dehydrogenase for
Table 2) and addition of catalase (1200 U mL− ), the pH was amination of alkanes highlights the current interest in such
1
2
1
set to 8.5 (see above) and all components required for dehydro- biotransformations.
Scheme 2 Conversion of 1a to L-4a in a three-step one-pot cascade: the reaction proceeds in Tris-HCl buffer (pH 7.5, 100 mM) containing 3 µM
P450CLA, 5% (v/v) EtOH (co-solvent), and 0.8 mM H
2
O
2
(from 320 mM stock) supplemented every 30 min (total 19.2 mM, module 1). After com-
−1
−1
pletion of the first step, catalase (1200 U mL ) is added and the pH is adjusted to 8.5, followed by addition of 0.07 U mL Leu-DH (cell-free lysate),
−1
+
0
4 2 4
.01 mg mL L- and D-Hic-DHs (freeze-dried whole cells), 50 mM (NH ) SO and 10 mM NAD (module 2).
8032 | Org. Biomol. Chem., 2018, 16, 8030–8033
This journal is © The Royal Society of Chemistry 2018

View Article Online
Organic & Biomolecular Chemistry
Communication
H. J. Schafer, Angew. Chem., Int. Ed., 2011, 50, 3854–3871;
Conclusions
(
c) K. Hill, Pure Appl. Chem., 2000, 72, 1255–1264.
In this work, we could demonstrate as proof-of-principle the
enantioselective biocatalytic formal α-amination of hexanoic
acid to L-norleucine, which proceeds without intermediate
purification in a linear one-pot three-step sequential cascade,
with pH adjustment after the first step. The key step was the
regioselective oxy-functionalization of non-activated C–H
bonds by P450 monooxygenase, followed by dehydrogenation
to the ketoacid and stereoselective reductive amination by
amino acid dehydrogenase. Reductive amination of
α-ketoacids to chiral amino acids by amino acid dehydrogen-
ases is well-documented, with industrial examples showcasing
4 S. Gandomkar, A. Dennig, A. Dordic, L. Hammerer,
M. Pickl, T. Haas, M. Hall and K. Faber, Angew. Chem., Int.
Ed., 2018, 57, 427–430.
5 S. Kim, S. Cheong, A. Chou and R. Gonzalez, Curr. Opin.
Biotechnol., 2016, 42, 206–215.
6 C. Wandrey, E. Fiolitakis, U. Wichmann and M. R. Kula,
Ann. N.Y. Acad. Sci., 1984, 434, 91–94.
7 B. Bossow and C. Wandrey, Ann. N.Y. Acad. Sci., 1987, 506,
325–336.
8 V. Resch, W. M. F. Fabian and W. Kroutil, Adv. Synth.
Catal., 2010, 352, 993–997.
9 Y. P. Xue, C. H. Cao and Y. G. Zheng, Chem. Soc. Rev., 2018,
47, 1516–1561.
2
2
the practical utility of the catalysts employed. The combi-
nation with a P450 peroxygenase, reported here for the first
time, appears promising for functionalization of bio-based 10 (a) A. S. Bommarius, M. Schwarm and K. Drauz, Chimia,
chemicals to access relevant chiral synthons. Conversions to
the final product are at present moderate; extension of this
work to preparative-scale synthesis of non-natural amino acids
with improved redox balance and final yield is currently under
investigation.
2001, 55, 50–59; (b) M. Breuer, K. Ditrich, T. Habicher,
B. Hauer, M. Kesseler, R. Sturmer and T. Zelinski, Angew.
Chem., Int. Ed., 2004, 43, 788–824; (c) J. Schrittwieser,
S. Velikogne and W. Kroutil, Catalysts, 2018, 8, 205.
11 (a) M. Girhard, S. Schuster, M. Dietrich, P. Durre and
V. B. Urlacher, Biochem. Biophys. Res. Commun., 2007, 362,
1
14–119; (b) T. Fujishiro, O. Shoji, S. Nagano, H. Sugimoto,
Y. Shiro and Y. Watanabe, J. Biol. Chem., 2011, 286, 29941–
9950.
2 A. S. Bommarius, K. Drauz, W. Hummel, M. R. Kula and
C. Wandrey, Biocatalysis, 1994, 10, 37–47.
Conflicts of interest
2
There are no conflicts to declare.
1
1
3 (a) I. G. Denisov, T. M. Makris, S. G. Sligar and
I. Schlichting, Chem. Rev., 2005, 105, 2253–2277;
Acknowledgements
(b) O. Shoji and Y. Watanabe, J. Biol. Inorg. Chem., 2014, 19,
The authors would like to thank Thomas Lins for excellent
technical assistance. The authors acknowledge the financial
support by the University of Graz.
529–539; (c) V. B. Urlacher and M. Girhard, Trends
Biotechnol., 2012, 30, 26–36.
4 (a) R. Bernhardt, J. Biotechnol., 2006, 124, 128–145;
1
(
b) I. G. Denisov and S. G. Sligar, in Cytochrome P450, ed.
P. R. Ortiz de Montellano, Springer, Heidelberg, 2015, pp.
9–109.
Notes and references
6
4
15 A. Dennig, M. Kuhn, S. Tassoti, A. Thiessenhusen, S. Gilch,
T. Bulter, T. Haas, M. Hall and K. Faber, Angew. Chem., Int.
Ed., 2015, 54, 8819–8822.
6 E. Koubek, M. L. Haggett, C. J. Battaglia, K. M. Ibne-Rasa,
H. V. Pyun and J. O. Edwards, J. Am. Chem. Soc., 1963, 85,
‡
In our previous work, oxidation of 2-hydroxyacids was performed by two
2 2
stereocomplementary (α-hydroxyacid oxidases, thereby releasing H O required
in the first step (‘catalytic reagent’ concept). This design was not selected for the
present three-step cascade, as final NADH-dependent reductive amination by
amino acid dehydrogenase would then require either stoichiometric amounts of
reduced nicotinamide or sacrificial substrate for cofactor recycling, in turn
increasing waste formation and affecting atom-economy.
1
2263–2268.
1
7 Both enzymes are active on 2-oxohexanoic acid in the
reductive direction, see: E. Busto, N. Richter, B. Grischek
and W. Kroutil, Chem. – Eur. J., 2014, 20, 11225–11228.
1
2
J. H. Schrittwieser, S. Velikogne, M. Hall and W. Kroutil,
Chem. Rev., 2018, 118, 270–348.
L. Martínez-Montero and I. Lavandera, in Modern 18 (a) H. Schutte, W. Hummel and M. R. Kula, Appl. Microbiol.
Biocatalysis: Advances Towards Synthetic Biological Systems,
ed. G. Williams and M. Hall, Royal Society of Chemistry,
Biotechnol., 1984, 19, 167–176; (b) W. Hummel, H. Schutte
and M. R. Kula, Appl. Microbiol. Biotechnol., 1985, 21, 7–15.
London, 2018, pp. 351–386; J. H. Schrittwieser, 19 A. Dennig, E. Busto, W. Kroutil and K. Faber, ACS Catal.,
S. Velikogne and W. Kroutil, in Modern Biocatalysis: 2015, 5, 7503–7506.
Advances Towards Synthetic Biological Systems, ed. G. 20 S. K. Wu, Y. Zhou, T. W. Wang, H. P. Too, D. I. C. Wang and
Williams and M. Hall, Royal Society of Chemistry, London,
018, pp. 387–438.
(a) J. C. Philp, R. J. Ritchie and J. E. M. Allan, Trends
Z. Li, Nat. Commun., 2016, 7, 11917.
21 H. L. Yu, T. Li, F. F. Chen, X. J. Luo, A. T. Li, C. Yang,
G. W. Zheng and J. H. Xu, Metab. Eng., 2018, 47, 184–189.
2
3
Biotechnol., 2013, 31, 219–222; (b) U. Biermann, 22 A. S. Bommarius, M. Schwarm and K. Drauz, J. Mol. Catal.
U. Bornscheuer, M. A. R. Meier, J. O. Metzger and B: Enzym., 1998, 5, 1–11.
This journal is © The Royal Society of Chemistry 2018
Org. Biomol. Chem., 2018, 16, 8030–8033 | 8033