A novel chimeric amine dehydrogenase shows altered substrate specificity compared to its parent enzymes

Article abstract of DOI:10.1039/c4cc06527a

We created a novel chimeric amine dehydrogenase (AmDH) via domain shuffling of two parent AmDHs ('L- and F-AmDH'), which in turn had been generated from leucine and phenylalanine DH, respectively. Unlike the parent proteins, the chimeric AmDH ('cFL-AmDH') catalyzes the amination of acetophenone to (R)-methylbenzylamine and adamantylmethylketone to adamantylethylamine.

Full text of DOI:10.1039/c4cc06527a

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
A novel chimeric amine dehydrogenase shows
altered substrate specificity compared to its
parent enzymes†
Cite this: Chem. Commun., 2014,
50, 14953
Received 19th August 2014,
Accepted 10th October 2014
Bettina R. Bommarius,a Martin Schurmannb and Andreas S. Bommarius*ac
¨
DOI: 10.1039/c4cc06527a
www.rsc.org/chemcomm
We created a novel chimeric amine dehydrogenase (AmDH) via domain
shuffling of two parent AmDHs (‘L- and F-AmDH’), which in turn had
been generated from leucine and phenylalanine DH, respectively.
Unlike the parent proteins, the chimeric AmDH (‘cFL-AmDH’) catalyzes
the amination of acetophenone to (R)-methylbenzylamine and
adamantylmethylketone to adamantylethylamine.
Enantiomerically pure amines are sought-after building blocks of
active pharmaceutical ingredients (APIs) in pharma, as exemplified
by sitagliptin (Januvias), rasagiline (Azilects), and oseltamivir
(Tamiflus). Sitagliptin has recently been made accessible via
transaminase catalysis.1 Even more recently, our lab developed
direct amination of ketones with NH3, catalyzed by amine
dehydrogenases (AmDHs) with different substrate specificities from
either leucine DH (‘L-AmDH’)2 or phenylalanine DH (‘F-AmDH’).3
However, neither L- nor F-AmDH can convert benzylic ketones with
appreciable activity.
Based on similar results in amino acid dehydrogenases (AADHs),4
we have employed domain shuffling to generate a new chimeric
amine dehydrogenase, cFL1-AmDH, from F-AmDH and L-AmDH
using overlap PCR, which is described in detail in the Method
section of the ESI.†‡ The previously described amine dehydro-
genases from our lab served as the parental enzymes for this
chimera. Generation of chimeric proteins via domain shuffling
can lead to new enzymes with improved functionality or extended
Fig. 1 (A) Model of the chimeric enzyme based on the structure of the
related phenylalanine dehydrogenase from Rhodococcus sp. M4 (1BW9)
(green)9 and leucine dehydrogenase from Bacillus sphaericus (1LEH),
(blue).10 (B) Schematic outline of construction of the chimeric enzyme
(green F-AmDH, dark blue L-AmDH, teal loop overlap). The amino acid
sequence is listed in Fig. S2 (ESI†).
Even though the authentic ketone binding domain from
range of substrate specificity.5–8 Residues 1–149 were contributed by F-AmDH is present in the new enzyme, the chimeric amine
F-AmDH (F-AmDH numbering) and 140 to the terminus 366 by the dehydrogenase now accepts new substrates such as benzylic
L-AmDH (L-AmDH numbering) (Fig. 1). Thus, the cFL1-AmDH carbonyl substrates, in addition to maintaining the substrate
retains the ketone/amine binding pocket of F-AmDH and the specificity of its parent enzyme measured so far, F-AmDH
cofactor binding domain from L-AmDH.
(Table 1). Chiral GC analysis of cFL1-AmDH substrates, such as
p-F-phenylacetone (pFPA), with previously determined enantio-
selectivity with the parent enzyme F-AmDH3 exhibited the same
enantioselectivity towards (R)-amine.
Moreover, the chimeric enzyme features a shifted temperature
profile towards higher temperatures with a temperature of optimum
activity Topt of 4 60 1C compared to a Topt of 50 1C for F-AmDH.
Compared to its parent enzymes, this new chimera converts
benzylic ketones when probed for substrate activity at 60 1C.
a Georgia Institute of Technology, School of Chemical & Biomolecular Engineering,
Parker H. Petit Building, 315 Ferst Drive, Atlanta, GA 30332-0100, USA.
E-mail: andreas.bommarius@chbe.gatech.edu
b DSM Innovative Synthesis BV, PO Box 18, NL-6160, MD Geleen, Netherlands
c Georgia Institute of Technology, School of Chemistry and Biochemistry,
95 Atlantic Ave., Atlanta, GA 30332-0420, USA
† Electronic supplementary information (ESI) available. See DOI: 10.1039/
c4cc06527a
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 14953--14955 | 14953

View Article Online
Communication
ChemComm
Table 1 Substrate specificity of cFL1-AmDH
Table 3 Kinetic properties of cFL2-AmDHa
Activityb (mU mgÀ1
)
Substrate
KM (mM)
kcat (sÀ1
)
kcat/KM (MÀ1 sÀ1
)
Substratesa
cFL1c
Fd
Ld
pFPA
Acetophenone
NH3
3.4 Æ 0.17
2 Æ 0.24
2.52 Æ 0.07
0.1 Æ 0.002
2.82 Æ 0.04
741
50
2.82
Ketones
p-Fluorophenylacetone
Acetophenone
1000 Æ 28
1725
301
107
69
30
133
0
4000
0
a
o0.1
59
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
pH 9.6, T = 60 1C; data for NH3 and NADH measured with 15 mM
p-FPA, 5 M formate buffer; data for p-FPA and NH3 measured with 200 mM
NADH, all values apparent.
1-Tetralone
0
Adamantylmethylketone
3-Methyl-1-phenylbutanone
Pinacolone
2-Tetralone
Benzophenone
0
0
0
0
Table 4 Thermal optima of different amine dehydrogenases
0
n.d.
AmDH
L-
F-
cFL1-
cFL2-
70
Amines
50a
50b
480
(R)-Methylbenzylamine
(R,S)-Methylbenzylamine
(R,S)-MOIPA
19
21
40
n.d.
0.5
0
476
484
130
Topt (1C)
a
b
Ref. 2. Ref. 3.
a
Substrate concentration = 20 mM. b 5 M NH4Cl buffer pH 9.5. c Measured
at 60 1C. d Measured at 25 1C. n.d. = not determined.
While the apparent kcat value for pFPA was indeed increased in
cFL2-compared to cFL1-AmDH (Table 3), the value for acetophenone
was decreased, indicating that N270/N271 both play a role in specific
activity dependent on the substrate within the cFL2-AmDH.
Specifically, we demonstrate the novel transformation of aceto-
phenone to (R)-methylbenzylamine. In addition, conversion of
the non-aromatic adamantylmethylketone to (R)-1-(1-adamantyl)-
ethylamine and the aliphatic methoxyacetone to (R)-methoxy-
isopropylamine (R)-MOIPA was observed (Fig. S3, ESI†). Thus,
we find that the cofactor binding domain in dehydrogenases can
play a significant role in substrate specificity, reshaping and
extending the substrate pocket to allow conversion of aliphatic
as well as aromatic and bulky ketones.
cFL1-AmDH was characterized with respect to its kinetic
properties (Table 2) as well as thermal behaviour (Fig. S4, ESI†).
Kinetic analysis revealed a reduced apparent KM value for
pFPA and ammonia compared to its parent enzyme, F-AmDH,3
but also reduced kcat values.
Sequence comparison of the amino acid DHs revealed two
adjacent asparagines N270 and N271 (cFL numbering) (Fig. 2).
Upon analysis of the protein structure model we concluded that the
2nd asparagine might have additional influence on amination.
This finding lead to the creation of cFL2-AmDH in which both
asparagines are mutated to leucine (N270L/N271L).
As mentioned before, domain shuffling of two related amine
dehydrogenases L- and F-AmDH resulted in an altered thermal
profile (Table 4). cFL1-AmDH is hardly active at 30 1C, starts to
exhibit good activity at 60 1C and stays active beyond 70 1C,
(at 470 1C, cofactor stability starts to be impaired and our
UV-VIS instrument reaches its limitations (Fig. S4, ESI,† for details
see Method section of ESI† as well)), whereas cFL2-AmDH reaches
its temperature of maximum activity Topt at 70 1C. The apparent
activation energies Ea,app for acetophenone and p-F-phenylacetone
with cFL1-AmDH were determined to be 32.7 and 48.7 kJ molÀ1
,
respectively, (35 kJ molÀ1 with cFL2-AmDH for p-F-phenylacetone),
corresponding to 16.5, 10.9, and 15.4 1C temperature increases,
respectively, for doubling activity. cFL1-AmDH at 45 and 55 1C
(Fig. S5 and Table S6, ESI†) was found to be very stable (t1/2
500 min); only at 70 1C does half-life decrease to 40 min.
4
We find that the cofactor binding domain in amine dehydro-
genases can play a significant role in ketone specificity and that
domain shuffling can (i) alter substrate specificity at comparable
kinetic properties to the parents and (ii) strongly improve thermal
activity.
Table 2 Kinetic properties of cFL1-AmDHa
B.R.B. gratefully acknowledges support from the NSF I/UCRC
grant 0969003 to the Center for Pharmaceutical Development (CPD).
Substrate
KM (mM)
kcat (sÀ1
)
kcat/KM (MÀ1 sÀ1
)
pFPA
Acetophenone
NH3
1.1 Æ 0.05
5.2 Æ 1.03
350 Æ 133
0.04 Æ 0.004
1.24 Æ 0.02
0.24 Æ 0.01
1.09 Æ 0.01
0.92 Æ 0.03
1127
48
Notes and references
3
NADH
2 Â 104
‡ Standard molecular biology protocols were applied to generate the
chimeric enzyme. Fusion of the domains was achieved using overlap
PCR within a common loop region of the 2 parent enzymes. Enzymatic
properties were determined using UV-spectroscopy monitoring the
change in NADH absorption at 340 nm as well as GC and HPLC. The
3D model of the structure was generated using Pymol.
a
pH 9.6, T = 60 1C; data for NH3 and NADH measured with 15 mM
p-FPA, 5 M formate buffer; data for p-FPA and NH3 measured with 200 mM
NADH, all values apparent.
1 C. J. Savile, J. M. Janey, E. C. Mundorff, J. C. Moore, S. Tam, W. R. Jarvis,
J. C. Colbeck, A. Krebber, F. J. Fleitz, J. Brands, P. N. Devine, G. W.
Huisman and G. J. Hughes, Biocatalytic Asymmetric Synthesis of Chiral
Amines from Ketones Applied to Sitagliptin Manufacture, Science, 2010,
329(5989), 305–309.
2 M. J. Abrahamson, E. Vazquez-Figueroa, N. B. Woodall, J. C. Moore
and A. S. Bommarius, Development of an amine dehydrogenase for
synthesis of chiral amines, Angew. Chem., 2012, 51(16), 3969–3972.
Fig. 2 Amino acid DH positions 270–280.
14954 | Chem. Commun., 2014, 50, 14953--14955
This journal is ©The Royal Society of Chemistry 2014

View Article Online
ChemComm
Communication
3 M. W. Abrahamson, J. W. Wong and A. S. Bommarius, The Evolution
of an Amine Dehydrogenase Biocatalyst for the Asymmetric Production
of Chiral Amines, Adv. Synth. Catal., 2013, 355(9), 1780–1786.
4 K. Kataoka, H. Takada, K. Tanizawa, T. Yoshimura, N. Esaki,
T. Ohshima and K. Soda, Construction and Characterization of
Chimeric Enzyme Consisting of an Amino-Terminal Domain of
Phenylalanine Dehydrogenase and a Carboxy-Terminal Domain
of Leucine Dehydrogenase, J. Biochem., 1994, 116, 931–936.
5 M. Shimoji, H. Yin, L. Higgins and J. P. Jones, Design of a novel
P450: a functional bacterial-human cytochrome P450 chimera,
Biochemistry, 1998, 37(25), 8848–8852.
7 I. R. Wheeldon, E. Campbell and S. Banta, A chimeric fusion protein
engineered with disparate functionalities-enzymatic activity and
self-assembly, J. Mol. Biol., 2009, 392(1), 129–142.
8 H. Zhang, G. T. Lountos, C. B. Ching and R. Jiang, Engineering of
glycerol dehydrogenase for improved activity towards 1, 3-butanediol,
Appl. Microbiol. Biotechnol., 2010, 88(1), 117–124.
9 J. L. Vanhooke, J. B. Thoden, N. M. Brunhuber, J. S. Blanchard and
H. M. Holden, Phenylalanine dehydrogenase from Rhodococcus sp.
M4: high-resolution X-ray analyses of inhibitory ternary complexes
reveal key features in the oxidative deamination mechanism, Bio-
chemistry, 1999, 38(8), 2326–2339.
6 C. R. Otey, J. J. Silberg, C. A. Voigt, J. B. Endelman, G. Bandara and 10 P. J. Baker, A. P. Turnbull, S. E. Sedelnikova, T. J. Stillman and
F. H. Arnold, Functional evolution and structural conservation in
chimeric cytochromes p450: calibrating a structure-guided approach,
Chem. Biol., 2004, 11(3), 309–318.
D. W. Rice, A role for quaternary structure in the sub-
strate specificity of leucine dehydrogenase, Structure, 1995, 3(7),
693–705.
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 14953--14955 | 14955