Generation of amine dehydrogenases with increased catalytic performance and substrate scope from ε-deaminating L-Lysine dehydrogenase

Article abstract of DOI:10.1038/s41467-019-11509-x

Amine dehydrogenases (AmDHs) catalyse the conversion of ketones into enantiomerically pure amines at the sole expense of ammonia and hydride source. Guided by structural information from computational models, we create AmDHs that can convert pharmaceutically relevant aromatic ketones with conversions up to quantitative and perfect chemical and optical purities. These AmDHs are created from an unconventional enzyme scaffold that apparently does not operate any asymmetric transformation in its natural reaction. Additionally, the best variant (LE-AmDH-v1) displays a unique substrate-dependent switch of enantioselectivity, affording S- or R-configured amine products with up to >99.9% enantiomeric excess. These findings are explained by in silico studies. LE-AmDH-v1 is highly thermostable (T_m of 69 °C), retains almost entirely its catalytic activity upon incubation up to 50 °C for several days, and operates preferentially at 50 °C and pH 9.0. This study also demonstrates that product inhibition can be a critical factor in AmDH-catalysed reductive amination.

Full text of DOI:10.1038/s41467-019-11509-x

A R T I C L E
https://doi.org/10.1038/s41467-019-11509-x
OPEN
G e n e r a t i o n o f a m i n e d e h y d r o g e n a s e s w i t h
i nc r e a s e d c a t al y ti c p e r f o r m a n c e a n d s ub s tr a te
s c o p e f ro m ε- de a m i n a t i n g L- L y s i ne d e h y d r o ge n a s e
1
1
1
1
1
V a s i l is T s e l i o u , T a nj a K n a u s , M a r c e l o F . M a sm a n , M a r i a L . C o r ra d o
& F r a n c e s c o G . M u t t i
A m i n e d e h yd r og e n a s es ( A m D Hs) c a t a l y s e t h e c o n v er s i on o f k e t on e s i n t o e n a n t i o m e ri c a l l y
p u re a mi n e s a t t h e s o l e e x p e n s e o f a m m o n i a a n d h y d ri d e s o u r c e . G u i d e d b y s t r u c t ur a l
i n f or m a ti on f r o m c o m p u ta ti o n a l m o d e l s , w e c r e a t e A m D H s t h a t c a n c o n ve r t p ha rm a c e ut i -
c a l l y r e l e va n t a ro m at i c k e to n e s w i th c o n v e r s i o n s u p t o q ua nt i t a t i v e a n d p er fe c t c h e m i c a l a n d
o p t i c a l p u ri t ie s. T h e s e A m DH s a r e c r e a t e d f r o m a n u nc on v en t i o na l e nz y m e s c a ff o l d t h a t
a p pa re nt ly d o e s n o t o p e r a te a n y a s y m m e t r i c t r a n sf or m a ti on i n i t s n a t ur a l r e a c ti on . A d d i -
t i on a l l y, t h e b e s t v ar i a n t ( L E - A m D H - v 1 ) d i s p l a y s a u n i q ue s u b s t r a t e - d e pe nd e n t s w i t c h o f
e n a nt io se l e c t iv i t y , a f f o r d i n g S- o r R- c o n fig u r e d a m i n e p ro d uc t s w i t h u p t o > 9 9 . 9 % e na n t i o -
m e r i c e x c e ss . T h e s e fin d i n g s a r e e x p l a i n e d b y i n s i l i c o s t u d i e s . L E - A m D H - v 1 i s h i g h l y t h e r -
m o s t a b l e ( Tm o f 6 9 ° C ) , r e t a i n s a l mo s t e n t i r e l y i t s c a t a l y t i c a c t i v i t y u po n i n c u b a t i o n u p t o
5 0 ° C f o r s e ve r al d ay s , a n d o p e r a te s p r e f e r e n ti a l l y a t 5 0 ° C a n d p H 9 . 0 . T h i s s t u d y a l s o
d e m o n s t r at e s t h a t p r o d u c t i n h i b i t i o n c a n b e a c r i t i c a l f a c t o r i n A m D H- c a t a l y s ed r e d u c t i ve
a m i n at io n .
1 Van ‘t Hoff Institute for Molecular Sciences, HIMS-Biocat, University of Amsterdam, Science Park 904, 1098 XH Amsterdam, The Netherlands.
Correspondence and requests for materials should be addressed to T.K. (email: t.knaus@uva.nl) or to F.G.M. (email: f.mutti@uva.nl)
N A T U R E C OM M U N I C A T I ON S |
(2 0 1 9 ) 1 0 : 3 7 1 7 | h t t p s :/ / d o i . o r g / 1 0 . 1 0 3 8 / s 4 1 4 6 7 - 0 1 9 - 1 1 5 0 9 - x | w ww .n a t u r e . co m / n a t u r e co m m u n ic a t io n s
1

A R T I C L E
N A T U R E C O M MU N ICA T IO N S | h t t p s : / / d o i .o r g / 1 0 .1 0 3 8 / s4 1 4 6 7 - 0 1 9 -1 15 0 9 - x
n 2007, when the Roundtable of the American Chemical Society’s using a genome-mining approach25,26. Conversely, known
Green Chemistry Institute identified the most aspirational che- engineered AmDHs have been obtained starting exclusively
mical transformations to challenge the pharmaceutical industry, from L-amino acid dehydrogenases (deaminating at the
I
the reductive amination of prochiral ketones with free ammonia α-amino group) and by mutating the same highly conserved
ranked second on the list1. Indeed, α-chiral amines constitute amino acid residues in the active site (i.e., lysine and
approximately 40% of the optically active drugs that are currently aspartate)27–31. Other AmDHs have been produced either by
commercialised mainly as single enantiomers2. These amines are domain-shuffling or introducing further mutations into these
classically synthesised industrially through the asymmetric hydro- first generation variants32,33
.
genation of activated intermediates such as enamides, enamines, or
Accordingly, the diversity of substrate scope and activity
pre-formed N-substituted imines3; however, such strategies lack among these AmDHs is poor because they were engineered from
efficiency because multiple chemical steps are required. Although very similar scaffolds and following an identical strategy. For
significant progress in the asymmetric reductive amination of instance, all of the known AmDHs—whether native or engi-
carbonyl-containing compounds has been recently achieved in neered enzymes—exhibit mediocre or no activity towards the
organometallic catalysis (e.g., using Ir or Ru or Pd-based catalysts) amination of acetophenones, tetralones, and chromanones, the
and organocatalysis4, a number of enzymatic routes offer more atom- related enantiomerically pure amines of which are recurrent
efficient approaches, particularly for the synthesis of active pharma- motives in many pharmaceuticals such as Sertraline, Norsertra-
ceutical ingredients (APIs) in enantiomerically pure forms5. Fur- line, Rotigotine, Rivastigmine, and Cinacalcet, among others3.
thermore, the often perfect stereoselectivity of enzymes eliminates the
Herein, we show that AmDHs can be engineered from a wild-
need for recrystallisation steps to upgrade the chemical purity of the type enzyme that apparently does not operate any asymmetric
enantiomer product, and there is no concern regarding the removal transformation. This property is revealed for an ε-deaminating
of traces of toxic heavy metals.
L-lysine dehydrogenase from Geobacillus stearothermophilus
Classical industrially applied and laboratory scale biocatalytic (LysEDH), and it is exploited for the creation of AmDHs that
methods convert α-chiral racemic amines into enantiomerically perform the reductive amination of pharmaceutically relevant
pure amines via either kinetic resolution (KR) or dynamic prochiral ketones with perfect stereo- and chemo-selectivity.
kinetic resolution (DKR) using hydrolases6, or deracemisation
using monoamine oxidases7, the latter of which also gives
access to optically active secondary and tertiary amines. Results
Although KR is limited to a theoretical maximum of 50% yield, Molecular modelling and preliminary activity studies. The
DKR and deracemisation require an additional chemical cata- wild-type LysEDH catalyses the (formally) irreversible oxidative
lyst or a stoichiometric reagent6,7. Emerging enzymatic meth- deamination of the ε-amino group of L-lysine (1b, Fig. 1a)34. This
ods involve the use of ammonia lyases8,9
,
Pictet- apparent irreversibility stems from the subsequent spontaneous
Spenglerases10,11, or berberine bridge enzymes12,13, all of cyclization of the ω-oxo amino acid product (1a). However, the
which are active towards valuable types of intermediates for the reverse reaction must be possible if substrates devoid of the α-
synthesis of chiral amines; however, their substrate scope is amine moiety (e.g., 6-oxo hexanoic acid) could be accepted by
rather restricted. Direct biocatalytic amination of unfunctio- LysEDH (2a, Fig. 2b). Initial experiments on the reductive ami-
nalised C-H bonds has been enabled by creating cytochrome nation of 2a (10 mM) in HCOONH4/NH3 (2 M, pH 9.0) revealed
P411, which requires tosyl azide as the nitrogen source14; that LysEDH (90 µM) catalyses this reaction (44% analytical
however, this challenging strategy is limited to benzylic C-H yield). The experiment was conducted in the presence of formate
amination and requires a subsequent deprotection step to dehydrogenase (Cb-FDH) for NADH recycling (Fig. 2a) as
remove the tosyl group. As such, asymmetric reductive ami- described in Methods section (Biocatalytic reductive amination in
nation of prochiral ketones and hydroamination of olefins are analytical scale)23,35. Under the same conditions, substrates
currently more atom-efficient methods for introducing an α- devoid of the ω-carboxyl moiety such as hexanal (23a), heptanal
chiral amine moiety. However, the biocatalytic hydroamination (24a), 2-heptanone (5a), 2-octanone (25a), acetophenone (8a)
of alkene moieties is restricted to the synthesis of α- and β- and α-chromanone (13a) were not converted (see Supplementary
amino acids8,9, whereas chemocatalytic hydroaminations yield Fig. 1 for the structures of 23a, 24a and 25a). Thus, the α-amino
preferentially secondary and tertiary amines with non-perfect group (i.e., present in L-lysine) is not essential for productive
enantiomeric excess15. The stereoselective conversion of substrate binding and amination catalysed by LysEDH, whereas
ketones to amines can be accomplished either by a formal the α-carboxyl group is crucial. These findings motivated us to
reductive amination using ω-transaminases16 or
reductive amination using amine dehydrogenases (AmDHs), of LysEDH. Due to the lack of a crystal structure of LysEDH, a
imine reductases (IReds) or reductive aminases (RedAms)17,18
homology model was created based on published homologous
a truly undertake a structural examination of the active site architecture
.
IReds and RedAms are normally employed for the synthesis of structures. After several rounds of homology modelling, in silico
secondary and tertiary amines17,18, although this property was modifications, and energy minimisation followed by molecular
also recently observed in some cases with AmDHs19. Transa- dynamic refinement, the model of LysEDH was obtained with
minases have proven to be useful and efficient biocatalysts for both coenzyme (NADH) and ligand 1c bound in the active site in
the stereoselective amination of ketones in laboratory, as well as the reactive pose (Fig. 1b, Supplementary Methods).
industrial scale settings16,20,21; however, their downside is the
Our analysis of the model identified seven amino acid residues
requirement for supra-stoichiometric amounts of an amine as potential targets for mutagenesis, which can be classified into
donor and/or more enzymes and cofactors16,20–22. In this two categories based on their role in the binding pocket (Fig. 1b).
context, the AmDH-FDH system enables the efficient reductive Group one encompasses residues that are appointed in the
amination of prochiral ketones (i.e., TONs > 103) while binding of the α-amino and α-carboxylic groups of 1b, namely
requiring only HCOONH4 buffer and catalytic NAD23. A H181, Y238, and T240 and R242, respectively. These residues
natural AmDH activity on aliphatic substrates was reported create a hydrophilic cavity that accommodates the hydrophilic
about two decades ago; however, the gene was never identi- α-amino acid moiety of 1b. On the opposite side, residues F173,
fied24. Native AmDHs were very recently identified for the V172 and V130 create a hindered hydrophobic environment that
reductive amination of aliphatic ketones with S-stereoselectivity forces the substrate into its ideal reactive pose and prevents its
2
N A T UR E C O M M U N I C A TI O N S |
(2 0 1 9 ) 1 0 : 3 7 1 7 | h t t p s : / / d o i . o r g / 1 0 . 1 0 3 8 / s 4 1 4 6 7 - 0 1 9 - 1 1 5 0 9 - x | w ww .n a t u r e . c o m / n a t u re c o m m u n i c a t i o n s

N A T UR E C O M M UN I CA T I O NS | h t t p s : //d o i . o r g / 10 . 1 0 38 / s 4 1 4 6 7 - 01 9 - 1 1 5 0 9 - x
A R T I C L E
analytical yield and high stereoselectivity (ee > 99% S, Table 1).
Moreover, preparative scale production of 3b starting from 3a
(150 mg, 1.040 mmol), which aimed at identifying the product’s
absolute configuration, resulted in 73% isolated yield with ee >
99% S (Supplementary Methods).
a
+
NH3
+
1b
1c
NH3
NAD+
NADH
–
O
–
O
+
H3N
+
H2N
O
O
NH3
H2O
+
NH3
–
O
– H2O
Asymmetric synthesis of amines catalysed by LysEDH variants.
The mutation F173A in the wild-type LysEDH created an addi-
tional large hydrophobic binding pocket in the active site, which
might favour the binding of aromatic substrates such as 8a.
Analysis of our LysEDH model also evidenced that residues
V172, V130 and R242 might be suitable targets for mutagenesis
(Fig. 1b). Therefore, a focused library of 14 variants (LE-AmDH-
v1 to v14, Fig. 2d) was created and screened against a number of
aliphatic and aromatic ketones (Fig. 2b, group B, 4–16a) in
HCOONH4/NH3 buffer (pH 9.0, 2 M) using LE-AmDH variants
(90 μM), Cb-FDH (19 μM), NAD+ (1 mM) for 48 h at 30 °C. The
screening outcome is summarised in Fig. 2c (with further details
in Supplementary Tables 3–7). Notably, in a large majority of
cases, the amine product was obtained in enantiomerically pure
form (ee > 99%), albeit with the opposite configuration (R) than
that observed for the natural binding mode of L-lysine (pro-S).
The mutation F173G (v2) resulted in a significant loss of
activity compared to the beneficial F173A (v1). Compared with
LE-AmDH-v1, LE-AmDH-v2 showed an approximately three-
fold decrease of conversion values for the amination of 8a and 9a
and a two-fold decrease for the amination of 10a. Moreover, LE-
AmDH-v2 did not convert 4a, 5a, 12a, 15a and 16a into the
corresponding amines, whereas LE-AmDH-v1 yielded, respec-
tively 78%, 10%, 61%, 26% and 58% conversion for the same
substrates. All variants bearing the mutation F173G (LE-AmDH-
v2, v3 and v4) generally showed very low levels of activity towards
the tested substrates, thus indicating that an increase of flexibility
in the active site is detrimental. Substitution of the neighbouring
V172 either to alanine or glycine while maintaining the beneficial
F173A mutation (LE-AmDH-v5 and v6) provoked a dramatic
reduction of catalytic activity. Interestingly, the residue R242
appears to be involved in the binding of the α-carboxylic group of
L-lysine and it is located at the entrance of the active site in a
region that is exposed to an aqueous environment (Supplemen-
tary Fig. 4). To investigate if R242 regulates accessibility of
substrates to the active site, we also created the variants R242M
(v14), R242M/F173A (v7), R242M/F173V (v13) and R242W/
+
N
H
–
O
C
O
O
O
1a
b
NADH
V130
1c
A131
F173
V172
Y238
H181
T240
R242
Fig. 1 M o d e l o f t h e a c t i v e s i t e o f t h e ε- d e a m i n a t ing L- ly s i n e d e h y d r o gen a s e
f r om G. stearothermophilus. a S c h em a t ic r e p r e s e n t a ti o n o f t h e r e a ct i o ns
u n de r g o n e b y t h e n a t u r a l s u b s t r a te 1b. b M o d e l o f t h e a c t i ve s i t e c o n t a i n i n g
i t s n a t u r a l s u b st r a t e 1c. R e s i d u e s H 18 1 , Y 2 3 8 a n d T 2 4 0 ( i n c y a n ) a p p e a r t o
+
s ta b i l i s e t h e a l p h a N H g r o u p , w h e re a s R 2 4 2 a p p e a r s t o i n t e r a c t w i t h t h e
3
C O O− o f t h e s u b s tr a t e . R e s id ue s V 17 2 , F 17 3 a n d V 13 0 c r e a t e a h i n d e r e d
h y dr o p h ob i c e n v i r o n m ent . A l l h i gh li g h t e d r e s i d u e s a r e a m e n a b l e t a r ge t s f o r
p r ot e i n e n g i n e er i ng s t u d i e s
rotation. In fact, in our model, the distance between the departing
hydride of NADH and the pro-chiral carbon of the imine F173A (v8). LE-AmDH-v7 converted substrates 8a (6%) and 9a
intermediate 1c is 2.7 Å, which is below the sum of the van der
Waals’ radii of carbon and hydrogen atoms36. Since F173 is the 13a (38%, 41%, 8%, 15% conversions, ee > 99% R, Supplementary
bulkiest hydrophobic residue, we envisioned that its mutation
into a less bulky residue such as alanine or glycine would retain the binding pocket at this site, with minimal conformational
(19%, ee > 99% R), whereas LE-AmDH-v13 converted 8–10a and
Table 4–6), thus indicating that increasing the hydrophobicity of
the overall hydrophobic properties of the cavity while also
permitting the binding of bulkier substrates in different
orientations. Hence, ketones might be accepted besides simpler
aldehydes.
rearrangement, reduces enzyme activity.
Finally, we investigated if the mutation of V130, which is
located on the opposite face of F173A, to a less bulky
hydrophobic amino acid residue such as alanine or glycine,
would enable a different binding mode for the substrate. LE-
As mentioned above, the wild-type LysEDH converted 2a to 2b
with 44% analytical yield. As 2b is a terminal (achiral) amine AmDH-v1 (F173A) was combined with mutations V130A or
product, information on the reaction’s stereoselectivity was not
attainable; however, the reductive amination of 3a as substrate exhibited a significant level of activity towards 8a (74%
(i.e., the structurally related methyl-ketone of 2a) would result in
an α-chiral amine. Since according to our model, the hydride of (8–13b) were obtained in enantiomerically pure form (ee > 99%
NADH must be transferred towards the Si-face of the prochiral
carbonyl moiety, this experiment would reveal a possible amination of 9a (78% conversion) and other ketones (8–9a,
stereoselectivity of LysEDH in the conversion of 3a. Indeed, 3b
was produced in an enantiomerically pure S-configuration (3a 50 AmDH-v12 also aminated 4a and 15a (64 and 42% conversion),
mM; LysEDH 90 µM; ee > 99%) and with 90% analytical yield.
When LysEDH F173A (herein referred as LE-AmDH-v1) was These results indicate that a productive pro-S binding pose was
V130G to generate LE-AmDH-v11 and v12. LE-AmDH-v11
conversion) and 9a (65% conversion), and all amine products
R). LE-AmDH-v12 was also stereospecific (ee > 99% R) for the
12–13a, 16a, Supplementary Tables 4–7). Nonetheless, LE-
albeit with imperfect stereoselectivity (ee 83% R and 77% R).
tested under the same conditions, 3a was aminated with > 99%
generated in these two reactions in addition to the favoured pro-
N A T U R E C OM M U N I C A T I ON S |
(2 0 1 9 ) 1 0 : 3 7 1 7 | h t t p s :/ / d o i . o r g / 1 0 . 1 0 3 8 / s 4 1 4 6 7 - 0 1 9 - 1 1 5 0 9 - x | w ww .n a t u r e . co m / n a t u r e co m m u n ic a t io n s
3

A R T I C L E
N A T U R E C O M MU N ICA T IO N S | h t t p s : / / d o i .o r g / 1 0 .1 0 3 8 / s4 1 4 6 7 - 0 1 9 -1 15 0 9 - x
a
b
Group A:
O
O
F
O
17a
O
21a
3a
19a
20a
2a
18a
O
O
O
NH3
O
H2O
LE-AmDHs
HO
HO
NH2
O
R
F
O
O
F
Group B:
R
R′
R′
6a
5a
7a
4a
8a
O
9a
O
O
O
NADH/H+ NAD+
O
F
HCOO–
10a
13a
–
O
O
O
O
11a
O
12a
HCO3
O
14a
15a
16a
O
O
Cb-FDH
c
d
LE-AmDH-v1
LE-AmDH-v2
LE-AmDH-v3
LE-AmDH-v4
LE-AmDH-v5
LE-AmDH-v6
LE-AmDH-v7
LE-AmDH-v8
LE-AmDH-v9
LysEDH variant
LE-AmDH-v1
LE-AmDH-v2
LE-AmDH-v3
LE-AmDH-v4
LE-AmDH-v5
LE-AmDH-v6
LE-AmDH-v7
LE-AmDH-v8
LE-AmDH-v9
LE-AmDH-v10
LE-AmDH-v11
LE-AmDH-v12
LE-AmDH-v13
LE-AmDH-v14
Mutations
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
F173A
F173G
V172A, F173G
V172G, F173G
V172G, F173A
V172A, F173A
R242M, F173A
R242W, F173A
V130A, F173V
V130G, F173V
V130A, F173A
V130G, F173A
R242M, F173V
R242M
LE-AmDH-v10
LE-AmDH-v11
LE-AmDH-v12
LE-AmDH-v13
LE-AmDH-v14
4a 5a 6a 7a 8a 9a 10a 11a 12a 13a 14a 15a 16a
Fig. 2 I n i t i a l s c r ee n i n g w i t h L E - A m DH v a r i a n t s . a G e n e r a l s c h e m e o f t h e b i o ca t a l y t i c r e d u ct i v e a m i n a t io n p e r f o r m e d b y L E- A m D H v a r i a n t s . A c a t a ly t i c
+
a
m
o
u
n
t
o
f
N
A
D
(
1
m
M
)
w
a
s
a
p
p
l
i
e
d
.
T
h
e
r
e
d
u
c
i
n
g
e
q
u
i
v
a
l
e
n
t
s
,
a
s
w
e
l
l
a
s
t
h
e
n
i
t
r
o
g
e
n
s
o
u
r
c
e
o
r
i
g
i
n
a
t
e
d
f
r
o
m
t
h
e
b
u
f
f
e
r
o
f
t
h
e
r
e
a
c
t
i
o
n
:
H
C
O
O
N
H
/
N
H
4
3
( p H 9 . 0 , 2 M ) . T h e s u b s tr a t e c o n c e n t r a t io n w a s 1 0 m M . A m D H a n d C b - F D H w e r e u s e d i n fin a l c o n c e n t r a t io n s o f 9 0 μM a n d 1 9 µM , r e s p e ct i v el y. R e ac ti o n
v o l u m e : 0 . 5 m L ; r e ac t i o n t i m e : 4 8 h ; t e m p e r a t u re : 3 0 ° C ; a g i t a t i o n o n a n o r b i t a l s h a ke r a t 1 7 0 r p m . b G r o u p B : s u b s t r a t e s u s e d f o r s c r e e n i n g o f t h e L y s E D H
v a r i an t s r e p o r t ed h e r e. G r o u p A : a d d i t io n a l s u b s t r a te s t e s t e d w it h L E- A m D H- v 1. c H e a t m a p o f t h e s c r e e n i n g o u t c o m e , i n w h i ch G r o u p B s u b s t r at e s w e r e
t e st e d t o r ed u c e s c r ee n i ng e f f o r t ( r e s u lt s e x p r e s se d i n c o n v er s i o n) . d L i s t o f m u t a t i on s i n t r o d u c e d i n t h e L y s E D H s c a f f o l d
R. Architectural modulation of the active site by testing the
The thermal robustness of LE-AmDH-v1 motivated us to
F173V mutation generated inactive variants (LE-AmDH-v9 and determine its thermostability data at different pH values using
v10). Figure 2c shows that the best variant for the reductive HCOONH4/NH3 buffer (2 M, pH 7.0–9.5) in the presence, as well
amination of the target aromatic ketones turned out to be LE- as absence of ligand 8a and coenzyme (NADH). In all cases
AmDH-v1. Accordingly, further studies in this work were (Supplementary Methods and Supplementary Table 17), the Tm
conducted with that AmDH.
was found to be 65 °C, which indicates no influence of pH in the
thermal stability of LE-AmDH-v1. Notably, Tm was increased to
69 °C when NADH was present in the mixture. All of these data
align with the thermal stability profile of the wild-type LysEDH34,
thus indicating that the mutation F173A did not affect the
stability of LE-AmDH-v1.
Finally, the long-term stability of LE-AmDH-v1 was tested in
HCOONH4/NH3 (2 M) buffer at pH 9.0 at different temperatures
and using various batches of enzyme for a maximum of 7 days
(Supplementary Methods). Remarkably, no loss of activity was
observed within this time period for samples incubated at 4 °C
and at room temperature. Samples incubated at 40 °C exhibited a
consistent residual activity between 80 and 90% after 7 days,
whereas samples incubated at 50 °C showed an average residual
activity of ca. 60% and a maximum residual activity of 80% after
7 days. Notably, the samples incubated at 60 °C (close to the Tm)
still showed a residual activity of ca. 50% after 70 hours. These
data further support the real applicability of LE-AmDH-v1 in
organic synthesis.
Biocatalytic optimisation studies on LE-AmDH-v1. The tem-
perature dependence and influence of the pH value were inves-
tigated for the reductive amination catalysed by LE-AmDH-v1
using 8a as test substrate. As it originates from the thermophile
bacterium Geobacillus stearothermophilus, the wild-type LysEDH
displays high stability and activity at elevated temperatures34. The
same features were observed for the LE-AmDH-v1 variant. Fig-
ure 3a (and Supplementary Table 8) shows that conversions of 8a
were above 90% after 16 h for reactions performed in the range of
30–60 °C. Remarkably, at 50 °C and 60 °C, conversions were
found to already be 80 and 90%, respectively, after the first 4 h.
Conversely, the conversion at 20 °C did not exceed 90% even after
32 h. A subsequent evaluation of the influence of buffers
(HCOONH4/NH3 2 M) at different pH values revealed the
highest conversions at pH 9.0–9.5 (Fig. 3b, Supplementary
Table 9).
4
N
A
T
U
R
E
C
O
M
M
U
N
I
C
A
T
I
O
N
S
|
(
2
0
1
9
)
1
0
:
3
7
1
7
|
h
t
t
p
s
:
/
/
d
o
i
.
o
r
g
/
1
0
.
1
0
3
8
/
s
4
1
4
6
7
-
0
1
9
-
1
1
5
0
9
-
x
|
w
w
w
.
n
a
t
u
r
e
.
c
o
m
/
n
a
t
u
r
e
c
o
m
m
u
n
i
c
a
t
i
o
n
s

N A T UR E C O M M UN I CA T I O NS | h t t p s : //d o i . o r g / 10 . 1 0 38 / s 4 1 4 6 7 - 01 9 - 1 1 5 0 9 - x
A
R
T
I
C
L
E
that produces either S-configured ω-amino acid 3a or R-config-
ured α-chiral amines in elevated enantiomeric excess (Table 1).
Table 1 Reductive amination of a panel of ketones and
aldehydes employing LE-AmDH-v1
Reaction intensification. Using the optimum pH value (pH 9.0),
a panel of ketones (8a, 9a, 13a, 17–19a and 21a) was tested for
reductive amination from 10 mM up to 100 mM final substrate
concentrations at two different temperatures (30 °C and 50 °C),
since higher temperatures may affect the stability of the enzyme
under process reaction conditions (e.g., shaking). The conversions
and productivities are displayed in Fig. 3c to i (also see Supple-
mentary Tables 10, 11). The highest productivity was observed at
50 °C and 100 mM substrate concentration for aminations of 8a
(84 mM), 9a (89 mM) and 18a (58 mM). The highest productivity
for amination of 17a (at 50 °C) was observed with 50 mM of
substrate concentration resulting in 46 mM of the corresponding
amine product. LE-AmDH-v1 could also produce 69.6 mM of
19b (at 50 °C) and 38.3 mM of 13b (at 30 °C) starting from
75 mM of 19a and 13a, respectively. Finally, the highest activity
was observed with benzaldehyde (21a), which was fully converted
by LE-AmDH-v1 at 200 mM scale at 50 °C.
Substrate
10 mM (Conv. %) 50 mM
(Conv. %)
> 9 9
ee (%)a
2a
3a
4a
8a
> 99
> 99
8 7
N . A .
> 9 9 ( S)
9 9 ( R)
b
> 9 9
7 1
b
b
> 99
9 8
8 7
3 6
7 9
8 2
8 6
8 6
9 8
9 5
6 5
2 6
5 0
5 5
8 4
8 1
> 9 9 .9 (R)
> 9 9 .9 (R)
9 9 . 8 (R)
N . D .
9a
10a
11a
12a
13a
14a
15a
16a
17a
18a
19a
20a
21a
b
> 9 9 ( R)
9 9 . 4 (R)
N . A .
8 9 ( R)
9 7 ( R)
> 9 9 .8 (R)
7 6
9 9
9 9
> 99
8 8
5 3
9 2
9 6
9 8
7 3
> 9 9
b
b
b
> 9 9 .7 ( R)
>
9
9
.
8
(
R
)
b
> 9 9 ( R)
c
c
> 99
N . A .
R e a c t i o n c o n di t i on s : s ub s t r a t e ( 1 0 m M o r 5 0 m M) , L E - A m D H - v 1 ( 9 0 µM ) , C b - F D H ( 19 µM ),
Reductive amination reactions in preparative scale. In order to
confirm the potential usefulness of LE-AmDH-v1 for the man-
ufacturing of APIs, we performed preparative scale reductive
amination reactions (100 mL reaction volume) starting from 8a
(600 mg, 5 mmol) and 9a (671 mg, 5 mmol), which were con-
verted into (R)-8b and (R)-9b with >99 and 97% conversion,
respectively. The unreacted ketone substrate was removed by
extraction with MTBE under acidic conditions. After basification,
the amine products (R)-8b and (R)-9b were extracted with MTBE
in 82 and 65% isolated yields, respectively. No further purification
was required. By considering the detection limit of the GC
measurement and concentration of the analytical samples (Sup-
plementary Methods), we could conclude that (R)-8b and (R)-9b
were obtained in >99.9% ee and total impurities were below the
detection limit (i.e., surely <0.15%); these values comply with the
requirement for API commercialisation.
N AD+ ( 1 m M ) , H C O O N H / NH b u ff e r ( 2 M , p H 9 . 0 ) , T 5 0 ° C , a g it a t i o n o r b i t a l s h a k e r 1 70 r p m ,
4
3
r e a c t i o n t i m e 4 8 h
ND c o u ld n ot b e d e t e rm i n e d , NA n o t a p p l i ca b l e
a
I n s o m e a n al y t i ca l s c al e r e ac t io n s , i t w a s p o ss i b l e t o d e t e r m i n e t h e ee v a lu e s w i t h a n a c c u r a c y
e q u al o r s u p e r io r t o 9 9 . 7 % d ue t o t h e e l e v a t e d c o n v e r s i o n a n d t h e h i g h r e s p o n se G C f a c t o r o f
t h e a m in e p ro d uc t . T h e 9 9 . 7 % ee t h r e s h o ld v a lu e i s i m p o r t a n t , a s i t i s s e t b y l a w f o r t h e
m a n u f a c t u re a n d c o m m e rc ia l is a t i o n o f A P I s ( i . e . , m a x 0 . 1 5 % t o t a l i m p u r i t i e s, w i t h t h e m i n o r
e n a n t i o m e r b e in g c o n si d e re d a s i m p u r i t y) . H i g h s t e r e o s e le c t i v i t y w a s a l s o c o n fir m e d i n t h e
p r e p a r a t i v e s c a le r e ac t io n s
b
I
n
t
h
e
s
e
c
a
s
e
s
,
t
h
e
l
e
v
e
l
o
f
p
r
e
c
i
s
i
o
n
d
e
p
e
n
d
s
o
n
i
n
s
t
r
u
m
e
n
t
a
l
d
e
t
e
c
t
i
o
n
l
i
m
i
t
s
r
a
t
h
e
r
t
h
a
n
t
h
e
i
n
t
r
i
n
s
i
c
s
t
e
r
e
o
s
e
l
e
c
t
i
v
i
t
y
o
f
t
h
e
e
n
z
y
m
e
.
I
n
d
e
e
d
,
t
h
e
S
-
e
n
a
n
t
i
o
m
e
r
w
a
s
n
e
v
e
r
o
b
s
e
r
v
e
d
.
c
R e a c t i on i n H C O O N H / N H b u f fe r ( 2 M , p H 7 . 8 )
4
3
Substrate scope of LE-AmDH-v1. LE-AmDH-v1 exhibited a
significantly extended and complementary substrate scope for
aromatic substrates compared with the currently available
AmDHs (Table 1). Specifically, Rs-AmDH, Bb-AmDH and Cal-
AmDH can efficiently aminate 4-phenylbutan-2-one (and deri-
vatives) and/or phenylacetone (and derivatives)23,28,29,31, whereas
other AmDHs such as Ch1-AmDH, LeuDH, AmDH4,
MsmeAmDH, MicroAmDH, CfusAmDH and EsLeuDH enzymes
Steady state kinetic data and inhibition studies. LE-AmDH-v1
was characterised with respect to its kinetic properties for the
amination of 8a and 21a. First, steady-state kinetics were deter-
mined for the amination of 8a at different temperatures; the most
elevated catalytic profile was observed at 60 °C (Table 2,
Supplementary Methods and Supplementary Table 19). There-
fore, all other enzymatic assays were performed at 60 °C. LE-
AmDH-v1 exhibited both the highest kcat value and the lowest
KM for 21a (Table 2, Supplementary Methods and Supplementary
Table 19).
With the aim of rationalising the higher conversion of 8a into
8b that LE-AmDH-v1 exhibits over Ch1-AmDH, we also
determined the kinetic parameters of Ch1-AmDH for the
reductive amination of 8a (Supplementary Methods and
Supplementary Table 19). Surprisingly, despite the much higher
conversion for the amination of 8a (50 mM) catalysed by LE-
AmDH-v1 (>99%, 90 μΜ enzyme) compared with Ch1-AmDH
(34%, 130 μΜ enzyme) after 48 h23, Ch1-AmDH possesses ca.
two-fold higher kcat and similar KM values of those of LE-AmDH-
v1 for the amination of 8a at 60 °C (Table 2). Therefore, we
hypothesised that the reductive amination of 8a catalysed by Ch1-
AmDH could be affected by severe product inhibition, as has also
been observed with other aminating enzymes such as ω-
transaminases16,39,40. We remark that steady-state kcat and KM
values are commonly determined at an initial zero concentration
of the product of the reaction. As such, solely considering these
parameters for the evaluation of the catalytic effectiveness of a
exhibit
particularly
high
activity
towards
aliphatic
substrates23,25,27,30,33,37. In contrast, elevated biocatalyst loading
of the previously described Ch1-AmDH (130 µM) poorly ami-
nated a 50 mM solution of 8a (34% conversion), derivatives
thereof (17–20a, 9–43% conversion), and 9a (8% conversion)
within 48 h23. Bulky-bulky ketones such as 10a and 11a were not
accepted at all by Ch1-AmDH23. In another case, an elevated
concentration of an AmDH engineered from Exigobacterium
sibiricum (EsLeuDH, 1.45 mM; concentration calculated from a
reportedly amount of 10 U mL−1, activity of 0.17 U mgenzyme
−1
and MWenzyme 40.5 kDa) was necessary to reach ca. 80% con-
version of 8a in 100 h38. In contrast, LE-AmDH-v1 (90 µM)
converted the above-mentioned acetophenone and derivatives
thereof (8a and 17–20a, 50 mM) with conversions up to 98%
within 48 h (Table 1). Propiophenone (9a) was accepted (50 mM,
95% conversion), as well as the more bulky substrates 10a
(50 mM, 65%) and 11a (50 mM, 26%). Moreover, neither Ch1-
AmDH23,32 nor other previously reported AmDHs (e.g., Cal-
AmDH)23,25,27–31,33,37 are capable of aminating either α-tetralone
(12a) or α-chromanone (13a) to any synthetically useful extent.
In contrast, LE-AmDH-v1 exhibited conversions above 50% for
the amination of these substrates (50 mM, Table 1). In addition,
LE-AmDH-v1 proved to be a versatile biocatalyst, such that ali-
phatic ketones (4a, 14–16a, 50 mM) were also accepted (53–84%
conversion) and benzaldehyde (21a, 50 mM) was quantitatively
aminated. Lastly, LE-AmDH-v1 is a high stereoselective catalyst
N
A
T
U
R
E
C
O
M
M
U
N
I
C
A
T
I
O
N
S
|
(
2
0
1
9
)
1
0
:
3
7
1
7
|
h
t
t
p
s
:
/
/
d
o
i
.
o
r
g
/
1
0
.
1
0
3
8
/
s
4
1
4
6
7
-
0
1
9
-
1
1
5
0
9
-
x
|
w
w
w
.
n
a
t
u
r
e
.
c
o
m
/
n
a
t
u
r
e
c
o
m
m
u
n
i
c
a
t
i
o
n
s
5

A
R
T
I
C
L
E
N
A
T
U
R
E
C
O
M
M
U
N
I
C
A
T
I
O
N
S
|
h
t
t
p
s
:
/
/
d
o
i
.
o
r
g
/
1
0
.
1
0
3
8
/
s
4
1
4
6
7
-
0
1
9
-
1
1
5
0
9
-
x
a
b
e
h
c
100
100
80
60
40
20
0
30 °C
50 °C
100
80
60
40
20
0
100
80
60
40
20
0
80
60
40
20
0
20°C
30°C
40°C
50°C
60°C
pH 7
pH 8
pH 8.5
pH 9
pH 9.5
0
10
20
30
40
50
0
5
10
15
20
25
10
30
50
8a (mM)
75
100
Time (h)
Time (h)
d
f
30 °C
50 °C
30 °C
50 °C
30 °C
50 °C
100
80
60
40
20
0
100
80
60
40
20
0
100
80
60
40
20
0
100
80
60
40
20
0
100
80
60
40
20
0
100
80
60
40
20
0
10
30
50
75
100
10
30
50
75
100
10
30
50
75
100
9a (mM)
13a (mM)
17a (mM)
g
i
30 °C
50 °C
30 °C
50 °C
30 °C
50 °C
100
80
60
40
20
0
100
80
60
40
20
0
100
80
60
40
20
0
100
80
60
40
20
0
100
80
60
40
20
0
100
80
60
40
20
0
0
20
40
60
80 100
0
20 40
60
80 100
0
20 40
60
80 100
18a (mM)
19a (mM)
21a (mM)
Fig. 3 O p t i m i s at i o n s t u d i e s e m p l o y in g L E - A m DH - v1. R e a c t i o n c o n d i t i o n s : 0 .5 m L fin a l v o l u m e ; b u f f e r : H C O ON H / N H 2 M p H 7 .0 –9 . 5 ; 1 7 0 r p m o n o r bi t a l
4
3
s ha k e r ; [ s u b s tr a t e ] = 1 0–1 00 m M ; [ N A D+] = 1 m M ; [ L E - A m DH - v1 ] = 9 0 μM ; [ C b - F D H] = 1 9 μM ; r e a ct i o n t i m e : 1 –4 8 h ( a–b) a n d 4 8 h ( c–i) a I nflu e n c e o f
t he t e m p e r a t u r e ( 20 t o 6 0 ° C ) o n t h e r e d u c ti v e a m in a t i o n o f 8a i n H C O ON H / N H 2 M p H 9 . 0 b u f f e r . b I n flu e n c e o f t h e p H v a l u e s u s i n g H C O O N H / N H
4
3
4
3
b u f f e r a t 5 0 ° C o n t h e r e d u c t iv e a m in a t i o n o f 8a. c–i E f f ec t o f t h e s u b s t r a t e c o n c e n t r a t io n o n c o n v e r s i o ns ( l ef t y- a x i s ; c o l u m n s) a n d p r o d u c t i v i t i e s ( r ig h t y-
a x i s ; d a s h ed l i n e s w i th s y m b o l s) a t t wo d i f f e r en t t e m p e r a t u re s : 3 0 a n d 5 0 ° C ) . S o u r c e d a t a a r e p r o v i d ed a s a s o u r c e d a t a fil e . D a t a a r e b a s e d o n s i ng l e
m e as u r e m en t s
biocatalyst can lead to wrong conclusions, particularly when Computational studies. The final refined model of LE-AmDH-v1
product inhibition occurs, as reviewed elsewhere41,42
.
(Fig. 1b, Supplementary Methods and Supplementary Table 12)
Initially, we studied the potential inhibitory effect of the amine was used for in-depth computational studies aimed at explaining
product (R)-8b on LE-AmDH-v1 and Ch1-AmDH at a 15 mM the stereoselective outcome of the reactions. We have shown that
concentration of 8a (Supplementary Methods and Supplementary linear or aromatic ketones (e.g., 4a, 8a) can be quantitatively
Table 20) by determining the IC50 values, which were indeed aminated with perfect stereoselectivity (ee > 99% R). Conversely,
20 mM for LE-AmDH-v1 and 1 mM for Ch1-AmDH (Fig. 4b and the amination of 3a proceeded quantitatively with opposite ste-
Table 2). A subsequent set of in-depth kinetic experiments on reoselectivity (ee > 99% S). A substrate-dependent switch of the
competitive product inhibition followed by Lineweaver-Burk enantioselectivity enabled by introducing a single mutation (as
analysis (Fig. 4c–e, Supplementary Methods and Supplementary the F173A in LE-AmDH-v1) has been observed in very rare cases
Tables 21-22)43 revealed that Ch1-AmDH is ca. 38-fold more in enzyme catalysis such as with an ene-reductase44, a naphtha-
inhibited than LE-AmDH-v1 (Table 2, see KI values). The lene dioxygenase45, a lipase46, and an ω-transaminase47. How-
eff
evaluation of the overall catalytic performance (kapp/KM
)
ever, a substrate-dependent switch of enantioselectivity has been
according to Fox et al.41, demonstrated that LE-AmDH-v1 is observed only in the latter two cases, albeit the resulting enan-
ca. 15-fold more efficient than Ch1-AmDH (Table 2, Supple- tiomeric excess was poor (ca. 58%)46,47. Therefore, LE-AmDH-v1
mentary Methods and Supplementary Table 23). In this is the first example of a biocatalyst possessing substrate-
eff
evaluation, KM acts as the time-averaged and effective value dependent stereo-switchable selectivity while always affording
for the KM over the course of the reaction, thus including the perfect enantiopurity of the products. Other examples from the
competitive product inhibition phenomenon. Notably, an extre- literature describe either more mutations and consequent major
eff
mely high value of KM for the conversion of 8a with Ch1- structural alterations48, or starting from scaffolds with poor
AmDH indicates that the theoretical vmax (i.e., kapp)—although selectivity49,50, or changing the reaction medium51. To explain
mathematically higher—already becomes physically unattainable this singular feature, the iminium intermediates 4c, 8c and 3c
at the beginning of the reaction. In conclusion, our kinetics were docked into the active site of the enzyme. Via molecular
explain the superior catalytic performance of LE-AmDH-v1 in docking and molecular dynamics simulations, we were able to
the reductive amination of 8a. Notably, LE-AmDH-v1 is also identify the putative productive pro-R and pro-S binding modes
capable of converting prochiral ketones that are not accepted by for all three iminium intermediates. The two parameters to
other AmDHs at all.
consider are19: (i) the distance between the departing hydride of
6
N
A
T
U
R
E
C
O
M
M
U
N
I
C
A
T
I
O
N
S
|
(
2
0
1
9
)
1
0
:
3
7
1
7
|
h
t
t
p
s
:
/
/
d
o
i
.
o
r
g
/
1
0
.
1
0
3
8
/
s
4
1
4
6
7
-
0
1
9
-
1
1
5
0
9
-
x
|
w
w
w
.
n
a
t
u
r
e
.
c
o
m
/
n
a
t
u
r
e
c
o
m
m
u
n
i
c
a
t
i
o
n
s

N A T UR E C O M M UN I CA T I O NS | h t t p s : //d o i . o r g / 10 . 1 0 38 / s 4 1 4 6 7 - 01 9 - 1 1 5 0 9 - x
A
R
T
I
C
L
E
Table 2 Steady state kinetic data and inhibitory effects obtained for LE-AmDH-v1 and Ch1-AmDHa
15
eff
Enzyme
Substrate
8a
T (°C)
kapp (min−1
)
KM (mM)
IC50 (mM)b
KI (mM)c
KM (mM)c
kapp/KMeff (M min−1
)
L
E
-
A
m
D
H
-
v
1
6 0
5 0
4 0
6 0
5 0
4 0
6 0
7 . 1 0 . 2
7 . 0 0 . 1
5 . 7 0 . 1
1 7. 7 0 . 4
1 3 . 1 0 . 5
7 . 7 0 . 3
4 3 .6 1 . 7
5 .5 0 .5
7 . 2 0 .4
9 .6 0 .6
5 .6 0 .4
8 . 5 0 .8
1 1 . 6 0 .9
4 . 9 0 .5
2
0
4
.
4
8
0
4
6
7
1
5
a
C
h
1
-
A
m
D
H
8a
1
0 .1 2 0
1 6 8 3 8
1
L
E
-
A
m
D
H
-
v
1
21a
a
3
2
T h e d e t er m in a t io n o f t h e c a t a ly t i c p a r a m e t e r s o f C h 1 - A mD H b y B o mm a ri u s a n d c o w o rk e r s ( re f . ) a r e v e ry s i m il a r t o t h o s e o b t a i n e d i n t h i s s t u d y : ka p p 1 4 . 4 m i n −1 ( 6 0 ° C) , 1 0. 8 m i n−1 ( 50 ° C ) a n d 7 . 2
m i n−1 ( 4 0 ° C ) ; KM 5 . 2 m M ( 6 0 ° C ) . S o u rc e d a t a a r e p r o v id e d a s a s o u r c e d a t a fil e . r e p r e s e n t t h e d e v i a t i o n ( n = 2 i n d e p e n d e n t e x p e r i m e n t s u s in g t w o d i f f e r e n t e n z y m e b a t c h e s )
b
D e te r m i ne d a t a c o n c e n t ra t i on o f 1 5 m M 8a
c
4
1
e f f
(R) - 8b w as u s e d a s i nh i bi t o r. F o l l o w i n g F o x e t a l . ( r e f. ) , KM w a s c a l cu l a t e d b a s e d o n t h e d e t e r m i n e d KI , [ S ]0 = 1 00 m M a n d c = 9 9 % ( f o r f u r t h e r d e t a i l s , s e e S u p p l e m e n t a r y M e t h o d s)
a
O
LE-AmDH-v1 or
Ch1-AmDH
NH2
NH2
+
HCOONH4/NH3 4 M pH 9
NADH 250 μM
(R)-8b
8a
(R)-8b
b
c
LE-AnDH-v1
12
10
8
10 mM (R)-8b
5 mM (R)-8b
0 mM (R)-8b
1.6
1.2
0.8
0.4
LE-AmDH-v1
Ch1-AmDH
15
50
IC
15
IC
6
50
4
2
0
0
5
10
15
20
25
–0.2
0.0
0.2
0.4
0.6
(R)-8b (mM)
1/c(8a) (mM–1
)
d
e
Ch1-AmDH
3.0
0.4 mM (R)-8b
0.2 mM (R)-8b
0 mM (R)-8b
LE-AmDH-v1
Ch1-AmDH
0.6
0.4
0.2
2.5
2.0
1.5
1.0
0.5
0.0
–KI
–KI
–4 –2
0
2
4
6
8
10
–0.2
0.0
0.2
0.4
0.6
1/c(8a) (mM–1
)
(R)-8b (mM)
Fig. 4 P r o d u c t i n h i b i t i on s t u d i e s e m p l o yi n g L E - A m DH -v 1 a n d C h 1 - A m D H . a O v e r a l l s c h e m e o f t h e k i n e t i c e x p e r i m en t s , i n w h i ch 8a i s u s e d a s s u bs t r at e a nd
1 5
(
R
)
-
8
b
a
s
c
o
m
p
e
t
i
t
i
v
e
i
n
h
i
b
i
t
o
r
.
b
I
n
h
i
b
i
t
o
r
y
e
f
f
e
c
t
(
I
C
)
o
f
t
h
e
a
m
i
n
e
p
r
o
d
u
c
t
(
R
)
-
8
b
a
t
1
5
m
M
o
f
s
u
b
s
t
r
a
t
e
c
o
n
c
e
n
t
r
a
t
i
o
n
8
a
.
c
a
n
d
d
P
r
i
m
a
r
y
d
o
u
b
l
e
5 0
r e c i pr o ca l p l o t s f o r c o m p e t it i ve i n h i b i t io n a c c o rd in g t o L i n e w e a v e r- B u r k a n a l y s i s f o r L E- A m D H-v 1 a n d C h 1 - A m D H , r e s p e ct i v e l y. e S e c o n d a r y p lo t s o f t h e
s lo p e s o b t ai n e d f r o m t h e p r i m a r y L i n e w e a v e r- B u r k p l o t s v s . i n h i b i t o r c o n c e n t ra t i o n s . T h e a b s o l u t e v a l u e o f t h e i n t e r ce p t w i t h t h e x- a x i s p r ov id e s t h e
c o m p e ti t iv e i n h i b i t i on c o n s t a n t ( KI) . S o u r c e d a t a a r e p r o v i d e d a s a s o u r c e d a t a fil e . D a t a p o i n t s o f F i g . 4 b a r e t h e a v e r a g e o f n = 2 i n d ep e n d e nt
m e as u r e m en t s , b u t t h e d ev i a t i o n i s s o l o w t h a t t h e e r r o r b a r i s s m a l l e r t h a n t h e d o t s ; t h e o n l y v i s i b l e e r r o r b a r i s f o r t h e p i n k l i n e a t ka p pc a . 5 . F i g ur e 4 c , d
a r e b a s e d o n s i n g l e m e a s u re me n t s . F i gu r e 4 e i s a s e c o n d a r y p l o t d e r i v e d f r o m F i g . 4c , d
NADH and pro-chiral carbon of the iminium intermediate carbon of 8c. Starting from the pro-S conformation of 8c, only
(Supplementary Methods); and (ii) the behaviour of a dihedral 0.6% of MD snapshot is favourable for reduction. The same
angle as defined by the three atoms bonded to the pro-chiral scenario occurs for the iminium intermediate 4c (generated
carbon of the iminium intermediate and the hydride atom of the from 4a). As expected, the situation is reversed for the reduction
NADH. This angle determines the optical configuration of the of 3c (generated from 3a). In this case, the pro-S conformation
amine product (for details, see Methods section and Supple- assumes a value below the threshold distance at 70.6% of
mentary Methods). The data are summarised in Fig. 5a.
the simulation time compared with at only 4.30% of the time for
In the case of the iminium intermediate (8c) generated from the pro-R.
8a, the pro-R conformation is highly favoured with 30.5% of the
Figure 5b represents a superposition of the structure of the
MD snapshot (within the simulation time) below the threshold wild-type LysEDH and the LE-AmDH-v1 (i.e., F173A variant)
distance of 3 Å between the hydride of NADH and prochiral with bound iminiums of L-lysine (1c) and acetophenone (8c).
N
A
T
U
R
E
C
O
M
M
U
N
I
C
A
T
I
O
N
S
|
(2 0 1 9 ) 1 0 : 3 7 1 7 | h t t p s :/ / d o i . o r g / 1 0 . 1 0 3 8 / s 4 1 4 6 7 - 0 1 9 - 1 1 5 0 9 - x | w ww .n a t u r e . co m / n a t u r e co m m u n ic a t io n s
7

A R T I C L E
N
A
T
U
R
E
C
O
M
M
U
N
I
C
A
T
I
O
N
S
|
h
t
t
p
s
:
/
/
d
o
i
.
o
r
g
/
1
0
.
1
0
3
8
/
s
4
1
4
6
7
-
0
1
9
-
1
1
5
0
9
-
x
a
b
6.5
6.0
5.5
5.0
4.5
4.0
3.5
3.0
2.5
2.0
1.5
Total average
Average under threshold
H181
NADH
F173
8c
A131
Y238
1c
R242
T240
Fig. 5 P r o - c h i r a l p r e f er e n ce s o f L E - A m DH- v1 . a T o t a l a v e r a ge h y d r i d e /pro- c h i r a l c a r b o n d i s t a n c e o v e r a m i n i m u m o f 6 M D s i m u l a t i o n s a n d a v e r a g e u nd e r
t he d i s t an c e t h r e sh o l d ( 3 Å ) . T h e p e r c e n t i l e s i n d i c a t e h o w o f t en t h e a v e r a g e d i s t a n c e w a s u n d e r t h e g i v en t h r es h o l d. F o r d e t a i l s , s e e S u p p l e m e nta r y
M e th o d s. b S u p er p o s i t io n o f t h e n a t u r a l s u b s t r a te a n d 8a i n t h e a c t i v e s i t e o f t h e wild type e n z y m e . T h e n a t u r a l i n t e r m e d i a te 1c i s d e p i c t e d i n m a g e n ta ,
w h e r e a s t h e i m i n i u m o f a c e t o p h e n o ne ( 8c) i s d e p i c t e d i n y e l l o w . T h e m u t a t i o n p o i n t ( F 1 7 3 A ) i s m a r ke d i n r e d a n d t h e c o f a c t or N A D H i s d e p i c t e d i n g r e e n .
I t c a n b e o b s er ve d t h a t o n c e t h e m u t a t io n F 17 3 A i s i n t r o d u ce d , t h e a r o m a t i c r i n g o f 8c g e t s a c c o m m o d a te d i n t o t h e n e w l y c r e a t e d h y d r o p h ob ic c a v i t y,
w h i c h w a s p r ev i o u s l y o c c u p ie d b y t h e p h e n y l g ro u p o f F 17 3 ; t h a t p r o d u c es a n i n v e r s i o n o f t h e c h i r a l p r ef e r e n ce s o f t h i s m u t a n t t o w a r d s a r o m a t ic s u bs t r a t e s
a n d t o w a r d s s u b st r a t e s w i t h ( r e l a t i v e ly) s h o r t h y d r o p h ob i c c h a i n s . S o u r c e d a t a a r e p r o v i d ed a s a s o u r c e d a t a fil e . E r r o r b a r s r e p r e s e nt t h e s t an da r d
d ev i a t i o n o f n = 6 i n d e p e n d e n t e x p e r im en t s
Ligand 8c is oriented to have its phenyl ring in the new cavity dimethylaminopyridine in acetic anhydride (50 µL of stock solution 50 mg mL−1
,
409 mM). The samples were shaken in an incubator at RT for 30 min, after which
water (500 µL) was added and the samples were shaken for an additional 30 min.
After centrifugation, the organic layer was dried with MgSO4. Enantiomeric excess
was determined by GC with a variant Chiracel DEX-CB column. Details on the GC
analysis and methods are reported in Supplementary Figs. 5–18 and Supplementary
Tables 1 and 2. Analytical reference compounds were synthesised as reported
in Supplementary Methods and Supplementary Tables 14–16 or purchased as
reported in Supplementary Notes.
created by the mutation F173A. This binding is not possible in
the wild-type enzyme due to a steric clash with the side chain of
F173. In contrast, 1c always assumes the “natural” conformation
as reported in the initial homology model (Fig. 1b).
Discussion
In the cases of 2a and 3a, the reactions were quenched by the addition of 100 μL
HCl (3 M) and the samples were centrifuged (15 min, 21.1 g). The supernatants
were analysed by RP-HPLC (Supplementary Methods)
In summary, guided by structural information obtained from
computational studies, we have generated AmDHs starting from
an enzyme that does not operate any apparent asymmetric
transformation in its natural reaction. The best variant (LE-
AmDH-v1) is highly thermostable (Tm: 69 °C), also exhibiting Biocatalytic reductive amination in preparative scale. Preparative scale reac-
tions of 8a (50 mM, 600 mg, 5 mmol) and 9a (50 mM, 671 mg, 5 mmol) were
performed in a total volume of 100 mL of HCOONH4/NH3 buffer (2 M, pH 9.0) at
highly retained catalytic activity (>99% at RT, 90% at 40 °C and
80% at 50 °C) upon incubation for 7 days. LE-AmDH-v1 operates
preferentially at 50 °C and pH 9.0, thus affording pharmaceuti-
cally relevant amines such as 8b (and derivatives), 12b, 13b and
9b in enantiomerically pure R configuration and elevated pro-
ductivities. Such biocatalytic performance is in part due to a
remarkably reduced product inhibition, as investigated for the
amination of 8a. LE-AmDH-v1 also quantitatively aminated 3a
into an S configured product (ee > 99%), thus providing a rare
example of substrate-dependent stereo-switchable selectivity.
Finally, in silico studies provided insight into the role of the
mutations upon ligand binding and explained the enantioselective
preferences of LE-AmDH-v1.
50 °C in 250 mL Erlenmeyer flasks containing LE-AmDH-v1 (90 μΜ, 4 mg mL−1),
FDH (16 μM, 0.68 mg mL−1), NAD+ (1 mM). After 48 h, a > 99% conversion was
obtained for 8a and a 63% was obtained for 9a. For the latter reaction, additional
aliquots of FDH (8 μΜ) and NAD+ (1 mM) were added and the reaction was
incubated for another 48 h, thus obtaining 97% conversion. The reactions were
acidified with concentrated HCl (6 mL) and unreacted starting material was
extracted with MTBE (2 × 80 mL). After the addition of KOH (10 M, 12 mL), the
amine products were extracted with MTBE (2 × 120 mL). The combined organic
layers were dried over MgSO4 and concentrated under reduced pressure. Avoiding
any further purification step, (R)-8b and (R)-9b were obtained with 82% (col-
ourless liquid, 495 mg, 4 mmol) and 65% (clear, yellow liquid, 437 mg, 3 mmol)
isolated yields. The purity of the products was analysed by 1H-NMR (400 MHz in
CDCl3) (Supplementary Figs. 19 and 20) and GC equipped with chiral column
(Supplementary Tables 2 and 24, and Supplementary Methods).
Methods
In silico modifications. All in silico modifications were performed using Yasara as
software52, and selecting AMBER 03 as force field53. Prior to the introduction of
any in silico mutation into the model structure, the protonation state of all the
atoms was corrected automatically with the exception of the atoms involved in the
hydride shift from the cofactor to the substrate. The protonation state of these
atoms was corrected manually upon visual inspection. After the introduction of a
mutation, the energy of the system was minimised following a three-step protocol
that enables the adjustment of the model structure without creating any possible
unwanted deformation. In step one, only the atoms constituting the mutated amino
acid residue were subjected to energy minimisation. In step two, the process for
energy minimisation was repeated by including all the atoms of the amino acid
residues that are located within 6 Å distance from the mutated residue. In step
three, the energy of the overall structure was minimised.
Biocatalytic reductive amination in analytical scale. SDS-page documenting the
purity of the AmDHs is reported in Supplementary Methods. LysEDH variant
(90 μM, 4 mg mL−1), NAD+ free acid (1 mM), formate dehydrogenase from
Candida boidinii (Cb-FDH, 19 μΜ, 0.81 mg mL−1) and substrate (4–21a,
10–200 mM) were added to an ammonium formate buffer (HCOONH4/NH3
0.5 mL, 2 M, pH 9). Biotransformations were performed at various temperatures
and for different lengths of time in Eppendorf tubes in a horizontal position on
orbital shakers (170 rpm). The reactions were quenched by the addition of aqueous
10 M KOH (100 μL) and extracted with dichloromethane (1 × 600 μL). After cen-
trifugation, the organic phase was dried with MgSO4 and the conversion was
measured by GC-FID. Prior to injection into a chiral column for the determination
of the ee, derivatisation of the samples was performed by the addition of 4-
8
N A T UR E C O M M U N I C A TI O N S |
(
2
0
1
9
)
1
0
:
3
7
1
7
|
h
t
t
p
s
:
/
/
d
o
i
.
o
r
g
/
1
0
.
1
0
3
8
/
s
4
1
4
6
7
-
0
1
9
-
1
1
5
0
9
-
x
|
w
w
w
.
n
a
t
u
r
e
.
c
o
m
/
n
a
t
u
r
e
c
o
m
m
u
n
i
c
a
t
i
o
n
s

N A T UR E C O M M UN I CA T I O NS | h t t p s : //d o i . o r g / 10 . 1 0 38 / s 4 1 4 6 7 - 01 9 - 1 1 5 0 9 - x
A R T I C L E
Homology modelling generation. The homology model was created as a hybrid
Determination of the IC5015. Product inhibition studies were performed in
model and using multiple templates as specified in the Supplementary Table 1254
.
duplicate at 60 °C. A fixed concentration of 8a (15 mM; from 500 mM main stock
in DMSO, which was further diluted to 350 mM in buffer) and varied con-
centrations of product (R)-8b, (inhibitor, 0–25 mM; from 500 mM main stock in
DMSO, which was further diluted to 350 mM in buffer) were added to a pre-
incubated buffer (70 °C) and further incubated at 60 °C for 2 min in the
thermostatic-controlled cuvette holder of the UV-vis spectrophotometer. The
enzyme was added (2.1–10.7 µM for Ch1-AmDH and 4.9 µM for LE-AmDH-v1)
and the solution was incubated at the desired temperature for a further 30 s, after
which the reaction was initiated by the addition of coenzyme (250 µM; from
50 mM main stock in a 50 mM KPi buffer at pH 8.0, which was further diluted to
10 mM in a reaction buffer).
The initial velocities were calculated from the linear range of the fitted trend line
of the progress curve and were then plotted against the inhibitor concentration to
obtain IC50 at a fixed substrate concentration (this value represents the amount of
amine product that correlates with a decrease of 50% of the initial catalytic activity
due to product inhibition at 15 mM 8a) and is displayed in Fig. 4b. For details,
see Supplementary Methods and Supplementary Table 20.
Details on the procedure are reported in Supplementary Methods.
Molecular docking. The molecular dockings were performed using Autodock
Vina55, both as standalone engine or as tool incorporated into Yasara. The pre-
selection and analysis of the obtained docked conformations were executed using
the standard Vina’s scoring function implemented into in-house-scripts. This
analysis resulted in a number of selected docking poses that were finally inspected
visually. The refined analysis resulted in the selection of the reactive docking poses,
which were used for the molecular dynamics (MD) simulations. Details on the
procedure are reported in Supplementary Methods.
Molecular dynamics simulations. The MD simulations were performed using
Yasara as software52, and selecting AMBER 03 as force field53. MD simulations were
executed considering both the pro-R and pro-S binding conformations of the three
iminium substrate intermediates (3c, 4c and 8c). The following parameters were
selected for the MD simulations: (1) the minimum number of independent simula-
tions with random initial velocities was 6; (2) the time for each simulation was 500 ps;
(3) analysis was performed by taking one snapshot every 6.25 ps. Thus, 81 frames
(counting also the starting structure) were collected for each simulation and they were
analysed considering different dynamic properties as explained in more details in
the Supplementary Methods. Among all the inspected properties, the most important
ones resulted to be: (1) the distance between the departing hydride of the coenzyme
NADH and the prochiral carbon of the iminium intermediate; (2) a dihedral angle (χ)
that allows us to define the stereoselective outcome of the reaction based on the
binding mode of the substrate in the active site19. In the case of the first parameter, the
threshold distance for a productive hydride shift was set to the sum between the van
der Walls radii of the carbon and hydrogen atoms, which is equal to ca. 3.0 Å
Determination of the KI for (R)-8b. Varied concentrations of substrate 8a
(0–30 mM, from 500 mM main stock in DMSO, which was further diluted
to 350 mM in buffer) and inhibitor (R)-8b (0, 5 and 10 mM for LE-AmDH-v1
and 0, 0.2, and 0.4 mM for Ch1-AmDH, from 500 mM main stock in DMSO,
which was further diluted to 350 mM in buffer) were added to a preincubated
buffer (70 °C) and further incubated at 60 °C for 2 min in the thermostatic
controlled cuvette holder of the UV-vis spectrophotometer. The enzyme
was added (4.5–9 µM for LE-AmDH-v1 and 2.9–6.4 µM for Ch1-AmDH)
and the solution was incubated for a further 30 s, after which the reaction
was initiated by the addition of NADH (250 µM; from 50 mM main stock in
50 mM KPi buffer at pH 8.0, which was further diluted to 10 mM in a reaction
buffer).
The initial velocities were calculated from the linear range of the fitted trend
line. Primary double reciprocal Lineweaver-Burk plots were generated for each set
of experiments at different inhibitor concentrations (Fig. 4c, d). The slopes
obtained from these primary plots were plotted against the inhibitor concentration
to obtain KI (Fig. 4e)43. KMeff was determined according to the equation described
in Supplementary Methods41, and the results are summarised in Table 2 and
Supplementary Tables 21–23.
À
Á
threshold
CH
rHvdw ¼ 1:2 Å; rCvdw ¼ 1:7 Å; rHvdw þ rCvdw ¼ 2:9 Å; thus : d
¼ 3:0 Å 56. In
the case of the second parameter, the dihedral angle was defined using the
Cahn–Ingold–Prelog priority and considering the hydride of the NADH coenzyme
along with the three atoms of the iminium substrates that are bound to the prochiral
carbon. In our model system, an angle of 63.88 0.21° leads to the formation of the R-
configured amine product, whereas an angle of −63.97 0.29° leads to the formation
of the S-configured amine product. A detailed description is reported in Supplemen-
tary Methods.
Reporting summary. Further information on research design is available in
the Nature Research Reporting Summary linked to this article.
Determination of NADH saturation for steady-state kinetics. All kinetic
experiments were conducted using a Shimadzu UV‐1800 UV-vis spectro-
photometer in HCOONH4/NH3 buffer (4 M, pH 9) and following the oxidation of
NADH at λ of 360 nm (ε = 4250 M−1 cm−1) for 1 min (λ of 360 nm was selected Data availability
due to an overlap of the absorption of 8a and NADH at the most frequently
selected λ of 340 nm).
The saturation concentration of NADH for Ch1-AmDH and LE-AmDH-v1 was
determined with 8a as substrate. Measurements at 60 °C were performed in two
sets of independent experiments and using two different charges of purified
protein.
A fixed concentration of 8a (25 mM; from 500 mM main stock in DMSO which
was further diluted to 350 mM in buffer) was added to a preincubated buffer (70 °C)
and further incubated at 60 °C for 2 min in the thermostatic-controlled cuvette
holder of the UV-vis spectrophotometer. The enzyme was added (2.0 µM for Ch1-
AmDH and 5.2 µM for LE-AmDH-v1) and the solution was incubated for a further
20 s at the desired temperature. Then, the reaction was initiated by the addition of
coenzyme (0–270 µM; from 50 mM main stock in 50 mM KPi buffer at pH 8.0,
which was further diluted to 10 mM in a reaction buffer).
The raw data underlying Figs. 3, 4, 5, Supplementary Methods-Supplementary Figs. 24,
36–41, and Supplementary Methods-Supplementary Table 17 are provided as a Source
Data file. Note that Table 2 is generated from the same source data of Supplementary
Methods-Supplementary Figs. 37–41, whereas Supplementary Methods-Supplementary
Tables 18 and 21 are generated from the same source data of Supplementary Methods-
Supplementary Figs. 37 and 41, respectively. Figure 4e and Supplementary Methods-
Supplementary Fig. 42 consist in a graphical/mathematical elaboration of Source Data
derived from Supplementary Methods-Supplementary Fig. 41 and following the
procedure described in ref. 43. All other data are available from the corresponding
authors upon reasonable request.
Received: 3 May 2019 Accepted: 15 July 2019
The initial velocities were calculated from the linear range of the fitted trend line
of the progress curve and were then plotted against the NADH concentration to
obtain the KM values (Supplementary Methods and Supplementary Table 18). The
Michaelis-Menten equation fitted well with the experimental determination.
References
1. Constable, D. J. C. et al. Key green chemistry research areas-a perspective from
pharmaceutical manufacturers. Green. Chem. 9, 411–420 (2007).
2. Ghislieri, D. & Turner, N. J. Biocatalytic approaches to the synthesis of
enantiomerically pure chiral amines. Top. Catal. 57, 284–300 (2013).
3. Chiral Amine Synthesis: Methods, Developments and Applications (ed. Nugent,
T. C.) (Wiley-VCH, Weinheim, 2010).
4. Wang, C. & Xiao, J. Asymmetric reductive amination, in Stereoselective
Formation of Amines Vol. 343 (eds. Wei, L. & Xumu, Z.) 261–282 (Springer-
Verlag, Berlin, 2014).
5. Kohls, H., Steffen-Munsberg, F. & Hohne, M. Recent achievements in
developing the biocatalytic toolbox for chiral amine synthesis. Curr. Opin.
Chem. Biol. 19, 180–192 (2014).
6. Mata-Rodríguez, M. & Gotor-Fernández, V. Resolution of alcohols, amines,
acids, and esters by nonhydrolytic processes, in Science of Synthesis, Biocatalysis
in Organic Synthesis 1 (eds. Faber, K., Fessner, W.-D. & Turner, N. J.) 189–222
(Georg Thieme Verlag KG, Stuttgart, 2015).
Steady-state kinetics. Two sets of independent measurements were performed
using two different charges of purified protein. Varied substrate concentrations
(from 500 mM main stock in DMSO, which was further diluted to 350 mM in
buffer) were added to a preincubated buffer (with an incubation temperature 10 °C
higher than the measuring temperature) and further incubated at the desired
measuring temperature in the thermostatic controlled cuvette holder of the UV-vis
spectrophotometer for 2 min. The enzyme was added and the solution was incu-
bated for a further 20 s, after which the reaction was initiated by the addition of
NADH (250 µM; from 50 mM main stock in 50 mM KPi buffer at pH 8.0, which
was further diluted to 10 mM in a reaction buffer).
The initial velocities were calculated from the linear range of the fitted trend line
of the progress curve and were then plotted against the substrate concentration to
obtain the kinetic parameter.
A summary of the reaction conditions used for the assay is provided in
the Supplementary Methods and Supplementary Table 19. The results are
summarised in Table 2. The Michaelis–Menten equation fitted well with the
experimental determinations.
N
A
T
U
R
E
C
O
M
M
U
N
I
C
A
T
I
O
N
S
|
(
2
0
1
9
)
1
0
:
3
7
1
7
|
h
t
t
p
s
:
/
/
d
o
i
.
o
r
g
/
1
0
.
1
0
3
8
/
s
4
1
4
6
7
-
0
1
9
-
1
1
5
0
9
-
x
|
w
w
w
.
n
a
t
u
r
e
.
c
o
m
/
n
a
t
u
r
e
c
o
m
m
u
n
i
c
a
t
i
o
n
s
9

A R T I C L E
N
A
T
U
R
E
C
O
M
M
U
N
I
C
A
T
I
O
N
S
|
h
t
t
p
s
:
/
/
d
o
i
.
o
r
g
/
1
0
.
1
0
3
8
/
s
4
1
4
6
7
-
0
1
9
-
1
1
5
0
9
-
x
7. Pollegioni, L. & Molla, G. C-N oxidation with amine oxidases and amino acid
oxidases, in Science of Synthesis, Biocatalysis in Organic Synthesis 3 (eds. Faber,
K., Fessner, W.-D. & Turner, N. J.) 235–284 (Georg Thieme Verlag KG,
Stuttgart, 2015).
33. Chen, F.-F. et al. Reshaping the active pocket of amine dehydrogenases for
asymmetric synthesis of bulky aliphatic amines. ACS Catal. 8, 2622–2628
(2018).
34. Heydari, M., Ohshima, T., Nunoura-Kominato, N. & Sakuraba, H. Highly
stable l-lysine 6-dehydrogenase from the thermophile Geobacillus
stearothermophilus isolated from a Japanese hot spring: characterization, gene
cloning and sequencing, and expression. Appl. Environ. Microbiol. 2004,
937–942 (2004).
35. Knaus, T., Cariati, L., Masman, M. F. & Mutti, F. G. In vitro biocatalytic
pathway design: orthogonal network for the quantitative and stereospecific
amination of alcohols. Org. Biomol. Chem. 15, 8313–8325 (2017).
36. Rowland, R. S. & Taylor, R. Intermolecular nonbonded contact distances in
organic crystal structures: comparison with distances expected from van der
Waals radii. J. Phys. Chem. 100, 7384–7391 (1996).
37. Au, S. K., Bommarius, B. R. & Bommarius, A. S. Biphasic reaction system
allows for conversion of hydrophobic substrates by amine dehydrogenases.
ACS Catal. 4, 4021–4026 (2014).
38. Lowe, J., Ingram, A. A. & Groger, H. Enantioselective synthesis of amines via
reductive amination with a dehydrogenase mutant from Exigobacterium
sibiricum: Substrate scope, co-solvent tolerance and biocatalyst
immobilization. Bioorg. Med. Chem. 26, 1387–1392 (2018).
39. Shin, J.-S. & Kim, B.-G. Asymmetric synthesis of chiral amines with ω-
transaminase. Biotechnol. Bioeng. 65, 206–211 (1999).
40. Truppo, M. D., Rozzell, J. D. & Turner, N. J. Efficient production of
enantiomerically pure chiral amines at concentrations of 50 g/L using
transaminases. Org. Process Res. Dev. 14, 234–237 (2010).
8. Bartsch, S. & Vogel, A. Addition of ammonia and amines to C=C bonds, in
Science of Synthesis, Biocatalysis in Organic Synthesis 2 (eds. Faber, K., Fessner,
W.-D. & Turner, N. J.) 291–311 (Georg Thieme Verlag KG, Stuttgart, 2015).
9. Li, R. F. et al. Computational redesign of enzymes for regio- and
enantioselective hydroamination. Nat. Chem. Biol. 14, 664–670 (2018).
10. Ilari, A., Bonamore, A. & Boffi, A. Addition to C=N bonds. in Science of
Synthesis, Biocatalysis in Organic Synthesis 2 (eds. Faber, K., Fessner, W.-D. &
Turner, N. J.) 159–175 (Georg Thieme Verlag KG, Stuttgart, 2015).
11. Lichman, B. R., Zhao, J. X., Hailes, H. C. & Ward, J. M. Enzyme catalysed
Pictet-Spengler formation of chiral 1,1 ‘-disubstituted- and spiro-
tetrahydroisoquinolines. Nat. Commun. 8, 14883 (2017).
12. Schrittwieser, J. H. et al. Deracemization by simultaneous bio-oxidative
kinetic resolution and stereoinversion. Angew. Chem. Int. Ed. 53, 3731–3734
(2014).
13. Schrittwieser, J. H. et al. Biocatalytic enantioselective oxidative C-C coupling
by aerobic C-H activation. Angew. Chem. Int. Ed. 50, 1068–1071 (2011).
14. Prier, C. K., Zhang, R. K., Buller, A. R., Brinkmann-Chen, S. & Arnold, F. H.
Enantioselective, intermolecular benzylic C-H amination catalysed by an
engineered iron-haem enzyme. Nat. Chem. 9, 629–634 (2017).
15. Reznichenko, A. L., Nawara-Hultzsch, A. J. & Hultzsch, K. Asymmetric
hydroamination, in Stereoselective Formation of Amines (eds. Li, W. & Zhang, X.)
191–260 (Springer, Berlin, 2014).
16. Slabu, I., Galman, J. L., Lloyd, R. C. & Turner, N. J. Discovery, engineering,
and synthetic application of transaminase biocatalysts. ACS Catal. 7,
8263–8284 (2017).
17. Sharma, M., Mangas-Sanchez, J., Turner, N. J. & Grogan, G. NAD(P)H-
dependent dehydrogenases for the asymmetric reductive amination of
ketones: structure, mechanism, evolution and application. Adv. Synth. Catal.
359, 2011–2025 (2017).
41. Fox, R. J. & Clay, M. D. Catalytic effectiveness, a measure of enzyme
proficiency for industrial applications. Trends Biotechnol. 27, 137–140 (2009).
42. Eisenthal, R., Danson, M. J. & Hough, D. W. Catalytic efficiency and kcat/KM:
a useful comparator? Trends Biotechnol. 25, 247–249 (2007).
43. Segel, I. H. Enzyme Kinetics: Behavior and Analysis of Rapid Equilibrium and
Steady-State Enzyme Systems. 10 (Wiley, New York, 1993).
44. Pompeu, Y. A., Sullivan, B. & Stewart, J. D. X-ray crystallography reveals how
subtle changes control the orientation of substrate binding in an alkene
reductase. ACS Catal. 3, 2376–2390 (2013).
18. Aleku, G. A. et al. A reductive aminase from Aspergillus oryzae. Nat. Chem. 9,
961–969 (2017).
19. Tseliou, V., Masman, M. F., Böhmer, W., Knaus, T. & Mutti, F. G. Mechanistic
insight into the catalytic promiscuity of amine dehydrogenases: asymmetric
synthesis of secondary and primary amines. ChemBioChem 20, 800–812
(2019).
20. Fuchs, M., Farnberger, J. E. & Kroutil, W. The industrial age of biocatalytic
transamination. Eur. J. Org. Chem. 2015, 6965–6982 (2015).
21. Savile, C. K. et al. Biocatalytic asymmetric synthesis of chiral amines
from ketones applied to sitagliptin manufacture. Science 329, 305–309
(2010).
45. Parales, R. E. et al. Regioselectivity and enantioselectivity of naphthalene
dioxygenase during arene cis-dihydroxylation: control by phenylalanine 352
in the a subunit. J. Bacteriol. 182, 5495–5504 (2000).
46. Magnusson, A. O., Takwa, M., Hamberg, A. & Hult, K. An S-selective lipase
was created by rational redesign and the enantioselectivity increased with
temperature. Angew. Chem. Int. Ed. 44, 4582–4585 (2005).
47. Svedendahl, M., Branneby, C., Lindberg, L. & Berglund, P. Reversed
enantiopreference of an ω-transaminase by a single-point mutation.
ChemCatChem 2, 976–980 (2010).
22. Mutti, F. G., Sattler, J., Tauber, K. & Kroutil, W. Creating a biocatalyst for the
production of an optically pure sterically hindered amine. ChemCatChem 3,
109–111 (2011).
23. Knaus, T., Böhmer, W. & Mutti, F. G. Amine dehydrogenases: efficient
biocatalysts for the reductive amination of carbonyl compounds. Green. Chem.
19, 453–463 (2017).
24. Itoh, N., Yachi, C. & Kudome, T. Determining a novel NADq-dependent
amine dehydrogenase with a broad substrate range from Streptomyces
virginiae IFO 12827: purification and characterization. J. Mol. Cat. B 10,
281–290 (2000).
48. Pratter, S. M. et al. Inversion of enantioselectivity of a mononuclear non-heme
iron(II)-dependent hydroxylase by tuning the interplay of metal-center
geometry and protein structure. Angew. Chem. Int. Ed. 52, 9677–9681
(2013).
49. Sun, Z. et al. Reshaping an enzyme binding pocket for enhanced and inverted
stereoselectivity: use of smallest amino acid alphabets in directed evolution.
Angew. Chem. Int. Ed. 54, 12410–12415 (2015).
50. Reetz, M. T. Laboratory evolution of stereoselective enzymes: a prolific source
of catalysts for asymmetric reactions. Angew. Chem. Int. Ed. 50, 138–174
(2011).
25. Mayol, O. et al. Asymmetric reductive amination by a wild-type amine
dehydrogenase from the thermophilic bacteria Petrotoga mobilis. Catal. Sci.
Technol. 6, 7421–7428 (2016).
51. Jongejan, J. A. Effects of organic solvents on enzyme selectivity, in Organic
Synthesis with Enzymes in Non-Aqueous Media (eds. Giacomo, C. & Sergio, R.)
25-46 (Wiley, New York, 2008).
26. Mayol, O. et al. A family of native amine dehydrogenases for the asymmetric
reductive amination of ketones. Nat. Catal. 2, 324–333 (2019).
27. Abrahamson, M. J., Vazquez-Figueroa, E., Woodall, N. B., Moore, J. C. &
Bommarius, A. S. Development of an amine dehydrogenase for synthesis of
chiral amines. Angew. Chem. Int. Ed. 51, 3969–3972 (2012).
28. Abrahamson, M. J., Wong, J. W. & Bommarius, A. S. The evolution of an
amine dehydrogenase biocatalyst for the asymmetric production of chiral
amines. Adv. Synth. Catal. 355, 1780–1786 (2013).
29. Ye, L. J. et al. Engineering of amine dehydrogenase for asymmetric reductive
amination of ketone by evolving Rhodococcus phenylalanine dehydrogenase.
ACS Catal. 5, 1119–1122 (2015).
30. Chen, F.-F., Liu, Y.-Y., Zheng, G.-W. & Xu, J.-H. Asymmetric amination of
secondary alcohols by using a redox-neutral two-enzyme cascade.
ChemCatChem 7, 3838–3841 (2015).
52. Krieger, E., Koraimann, G. & Vriend, G. Increasing the precision of
comparative models with YASARA NOVA–a self-parameterizing force field.
Proteins 47, 393–402 (2002).
53. Oostenbrink, C., Villa, A., Mark, A. E. & van Gunsteren, W. F. A biomolecular
force field based on the free enthalpy of hydration and solvation: the
GROMOS force-field parameter sets 53A5 and 53A6. J. Comput. Chem. 25,
1656–1676 (2004).
54. Venselaar, H. et al. Homology modelling and spectroscopy, a never-ending
love story. Eur. Biophys. J. 39, 551–563 (2010).
55. Trott, O. & Olson, A. J. AutoDock Vina: improving the speed and accuracy of
docking with a new scoring function, efficient optimization, and
multithreading. J. Comput. Chem. 31, 455–461 (2010).
56. Bondi, A. van der Waals Volumes and radii of metals in covalent compounds.
J. Phys. Chem. 70, 3006–3007 (1966).
31. Pushpanath, A., Siirola, E., Bornadel, A., Woodlock, D. & Schell, U.
Understanding and Overcoming the limitations of Bacillus badius and
Caldalkalibacillus thermarum amine dehydrogenases for biocatalytic reductive
amination. ACS Catal. 7, 3204–3209 (2017).
32. Bommarius, B. R., Schürmann, M. & Bommarius, A. S. A novel chimeric
amine dehydrogenase shows altered substrate specificity compared to its
parent enzymes. Chem. Commun. 50, 14953–14955 (2014).
Acknowledgements
This project has received funding from the European Research Council (ERC) under the
European Union’s Horizon 2020 Research and Innovation programme (ERC-StG, Grant
Agreement No. 638271, BioSusAmin). Dutch funding from the NWO Sector Plan for
Physics and Chemistry is also acknowledged.
10
N A T UR E C O M M U N I C A TI O N S |
(2 0 1 9 ) 1 0 : 3 7 1 7 | h t t p s : / / d o i . o r g / 1 0 . 1 0 3 8 / s 4 1 4 6 7 - 0 1 9 - 1 1 5 0 9 - x | w ww .n a t u r e . c o m / n a t u re c o m m u n i c a t i o n s

N A T UR E C O M M UN I CA T I O NS | h t t p s : //d o i . o r g / 10 . 1 0 38 / s 4 1 4 6 7 - 01 9 - 1 1 5 0 9 - x
A R T I C L E
Peer review information: Nature Communications thanks Bian Wu and other
anonymous reviewer(s) for their contribution to the peer review of this work. Peer
reviewer reports are available.
Author contributions
All of the authors contributed to the design of the experiments. V.T. expressed and
purified the enzymes, performed and analysed the biocatalytic reactions by GC, opti-
mised the reaction conditions, performed thermal stability measurements, and per-
formed preparative scale amination. T.K. performed the tasks of molecular biology for
the creation of the gene libraries of variants, as well as the kinetics, inhibition, and long-
term stability studies, and analysed the kinetic data. M.F.M. performed the homology
modelling, the molecular modelling and the molecular dynamics simulations. V.T.,
M.F.M. and F.G.M. designed the focused library of variants. M.L.C. performed the
chemical synthesis and identification of 6-oxo carboxylic acids and their methyl esters, 6-
amino carboxylic acids methyl esters, and analysed biocatalytic reactions by RP-HPLC.
F.G.M. and T.K. directed the project. F.G.M. conceived the project and received the
funding. V.T., T.K., M.F.M. and F.G.M. prepared the paper and Supplementary Infor-
mation, which were revised and approved by all the authors.
Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in
published maps and institutional affiliations.
Open Access This article is licensed under
a Creative Commons
Attribution 4.0 International License, which permits use, sharing,
adaptation, distribution and reproduction in any medium or format, as long as you give
appropriate credit to the original author(s) and the source, provide a link to the Creative
Commons license, and indicate if changes were made. The images or other third party
material in this article are included in the article’s Creative Commons license, unless
indicated otherwise in a credit line to the material. If material is not included in the
article’s Creative Commons license and your intended use is not permitted by statutory
regulation or exceeds the permitted use, you will need to obtain permission directly from
the copyright holder. To view a copy of this license, visit http://creativecommons.org/
licenses/by/4.0/.
Additional information
Supplementary Information accompanies this paper at https://doi.org/10.1038/s41467-
019-11509-x.
Competing interests: The authors declare no competing interests.
© The Author(s) 2019
Reprints and permission information is available online at http://npg.nature.com/
reprintsandpermissions/
N A T U R E C OM M U N I C A T I ON S |
(
2
0
1
9
)
1
0
:
3
7
1
7
|
h
t
t
p
s
:
/
/
d
o
i
.
o
r
g
/
1
0
.
1
0
3
8
/
s
4
1
4
6
7
-
0
1
9
-
1
1
5
0
9
-
x
|
w
w
w
.
n
a
t
u
r
e
.
c
o
m
/
n
a
t
u
r
e
c
o
m
m
u
n
i
c
a
t
i
o
n
s
11