Direct Reductive Amination of Biobased Furans to N-Substituted Furfurylamines by Engineered Reductive Aminase

Article abstract of DOI:10.1002/adsc.202001495

Furfurylamines are important building blocks for the synthesis of many pharmacologically active compounds and polymers. In this work, direct reductive amination of biobased furans to N-substituted furfurylamines by reductive aminase from Aspergillus oryzae (AspRedAm) was reported. Besides the reductive aminase activity, AspRedAm also showed a promiscuous, yet low alcohol dehydrogenase activity. The variant W210F proved to be a good catalyst for the synthesis of N-substituted furfurylamines. Furans were transformed to the target products with the conversions up to >99% and selectivities up to >99%. In addition, N-substituted furfurylamines were synthesized in the total turnover number (TTN) up to 3200 on a preparative scale, indicating the applicability of this biocatalytic route in synthetic chemistry. (Figure presented.).

Full text of DOI:10.1002/adsc.202001495

COMMUNICATIONS
DOI: 10.1002/adsc.202001495
Direct Reductive Amination of Biobased Furans to N-Substituted
Furfurylamines by Engineered Reductive Aminase
Zi-Yue Yang,a Ya-Cheng Hao,a Song-Qing Hu,a Min-Hua Zong,a Qi Chen,b,* and
Ning Lia,*
a
School of Food Science and Engineering, South China University of Technology, 381 Wushan Road, Guangzhou 510640,
People’s Republic of China
E-mail: lining@scut.edu.cn
b
State Key Laboratory of Bioreactor Engineering, Shanghai Collaborative Innovation Center for Biomanufacturing, East China
University of Science and Technology, 130 Meilong Road, Shanghai 200237, People’s Republic of China
E-mail: chenq@ecust.edu.cn
Manuscript received: December 1, 2020; Revised manuscript received: December 18, 2020;
Version of record online: January 7, 2021
Supporting information for this article is available on the WWW under https://doi.org/10.1002/adsc.202001495
efforts have been devoted to the production of furan
carboxylic acids (e.g., 2,5-furandicarboxylic acid
Abstract: Furfurylamines are important building
blocks for the synthesis of many pharmacologically
(FDCA) and 2-furoic acid) and alcohols (e.g., 2,5-bis
active compounds and polymers. In this work, direct
(hydroxymethyl)furan (BHMF) and furfurylalcohol);
reductive amination of biobased furans to N-
substituted furfurylamines by reductive aminase
from Aspergillus oryzae (AspRedAm) was reported.
Besides the reductive aminase activity, AspRedAm
significant advances in catalytic upgrading of biobased
furans have been achieved recently.[3] The synthesis of
furfurylamines via reductive amination is of great
interest,[4] since the furfurylamine moieties constitute
also showed a promiscuous, yet low alcohol
the key structural units of polymers[5] and pharmaco-
dehydrogenase activity. The variant W210F proved
to be a good catalyst for the synthesis of N-
substituted furfurylamines. Furans were transformed
to the target products with the conversions up to
>99% and selectivities up to >99%. In addition, N-
substituted furfurylamines were synthesized in the
logically active compounds such as antihypertensives
and histamine H2-receptor antagonist.[6]
A one-pot, two-step route starting with the imine
formation followed by the reduction with excess
NaBH4 was reported for the synthesis of N-substituted
furfurylamines from biobased furans.[7] It not only
total turnover number (TTN) up to 3200 on a
resulted in the generation of copious amounts of waste,
preparative scale, indicating the applicability of this
but also involved the use of the costly and strong
biocatalytic route in synthetic chemistry.
reductant and volatile organic solvents (e.g., CH2Cl2).
Chemically catalytic routes involving noble metals
(e.g., Ru, Pd, Au, and Ag) were constructed for
Keywords: amines; biobased furans; enzyme catal-
reductive amination of furans with H2, CO/H2O, and
ysis; reductive aminases; site-directed mutagenesis
electricity as the reductants.[4e,8] Chemical catalysis is
still suffering from some issues such as use of metals,
safety concerns (due to high pressure H2), and over-
Due to the concerns on fossil fuel resources crisis and alkylation (producing tertiary amines). Biocatalysis is a
environmental issues (e.g., global warming), the promising option to supplement and even substitute
production of biobased fuels and chemicals from chemical catalysis for upgrading the inherently unsta-
abundant, renewable, and carbon-neutral biomass has ble biobased furans,[9] since the former is usually
attracted increasing interest in the last decades.[1] associated with mild reaction conditions, exquisite
Biobased furans such as 5-hydroxymethylfurfural selectivity and environmental friendliness.[10] There are
(HMF) and furfural are versatile platform chemicals few reports on biocatalytic reductive amination of
bridging the gap between biomass and biobased biobased furans. Dunbabin et al. presented the syn-
chemicals as well as between biomass and biofuels.[2] thesis of primary furfurylamines via ω-transaminase-
The high reactivity allows the facile valorization of catalyzed transamination between furans and amines
these biobased furans into value-added products. Great (i.e., isopropylamine and α-methylbenzylamine) with
Adv. Synth. Catal. 2021, 363, 1033–1037
1033
© 2021 Wiley-VCH GmbH

COMMUNICATIONS
asc.wiley-vch.de
pyridoxal-5-phosphate (PLP) as cofactor.[4d] Other yl)-furan (HPAF) and N-propylfurfurylamine was
sources of ω-transaminases were exploited for the conducted.
synthesis of furfurylamines.[11] However, ω-transami-
AspRedAm-catalyzed reductive amination of HMF
nase-catalyzed reactions only provide the primary with propylamine was conducted (Figure 1), in which
amines, instead of the secondary amines. To our the regeneration of NADPH was realized via glucose
knowledge, biocatalytic synthesis of biobased N- dehydrogenase (GDH cell-free extract from Bacillus
substituted furfurylamines has not been reported yet. megaterium, 20 U/mg)[13]-catalyzed oxidation of glu-
Turner and co-workers reported an NADPH-dependent cose. As shown in Figure 1, HMF of approximately
reductive aminase from Aspergillus oryzae (AspRe- 27% was converted by this enzyme after the reaction
dAm) capable of catalyzing direct reductive amination of 24 h, and its conversions slightly increased with the
of various carbonyl compounds with amines, which elongation of the reaction periods. It indicates the low
showed broad substrate specificity.[12] We envisioned catalytic activity of AspRedAm toward HMF, which is
that this enzyme might accept biobased furans as consistent with the fact that the enzyme preferred
substrates. In this work, therefore, AspRedAm was ketones to aldehydes as substrates.[12] Unexpectedly,
examined for the synthesis of N-substituted furfuryl- the resulting products contained BHMF as well as the
amines from biobased furans via direct reductive desirable product HPAF. The selectivities toward
amination (Scheme 1). Structure-guided semi-rational HPAF (55–74%) were moderate. The formation of
mutation strategy was applied for improving its BHMF might be attributed to the promiscuous activity
catalytic performances. Besides, the preparative-scale of AspRedAm rather than the catalytic behaviors of the
synthesis of 2-(hydroxymethyl)-5-(propylaminometh- impure proteins such as alcohol dehydrogenases
(ADHs). To evidence the abovementioned assumption,
high purity of AspRedAm was prepared by nickel
affinity chromatography followed by size exclusion
chromatography (Figure S1), and its ADH activity was
determined with HMF as substrate (Figure S2). Indeed,
the reduction of HMF occurred in the presence of
AspRedAm especially with NADH as cofactor com-
pared to the controls in the absence of enzyme. So, it
was confirmed that AspRedAm had a promiscuous, yet
low ADH activity. This information may be useful to
unveil the evolutionary history of this enzyme.[14] Also,
AspRedAm may be a potential platform to create novel
ADHs by protein engineering for synthesis
Scheme 1. Reductive amination of biobased furans by AspRe-
applications.[15] According to Aleku et al.,[12] AspRe-
dAm and its variants.
dAm is representative of a subclass of imine reductases
(IREDs) based on the structural studies. Likewise,
Lenz et al. reported that imine reductases were capable
of promiscuously reducing a highly reactive ketone
2,2,2-trifluoroacetophenone.[16] Recently Nestl and co-
workers proposed that there might exist a common
evolutionary origin between IREDs and β-hydroxyacid
dehydrogenases based on the similarity in their gene
sequences as well as in NAD(P)H binding motifs.[17]
Therefore, the promiscuous ADH activity of AspRe-
dAm is assumed as the reversed function of the
progenitor enzyme during evolution.
The process optimization was carried out to
improve the synthesis of HPAF (Figure S3). Intrigu-
ingly, its synthesis was significantly enhanced upon
optimization. The substrate conversions up to >99%
were achieved within 12 h, and HPAF was obtained
with the selectivities up to 91% (Figure S3b). To
demonstrate the applicability of AspRedAm for the
valorization of HMF, the synthesis of HPAF was
performed on a preparative scale under the optimal
conditions (Figure S4). It was found that both HMF
conversion and selectivity toward the target product
Figure 1. Reductive amination of HMF with propylamine by
AspRedAm. Reaction conditions: 5 mM HMF, 100 mM propyl-
amine, 30 mM glucose, 0.2 mM NADPH, 0.2 mg/mL AspRe-
dAm, 0.5 mg/mL GDH, 1 mL Tris-HCl buffer (0.1 M, pH 8),
°
150 rpm, 25 C; tuning final pH to 9 after the addition of
propylamine.
Adv. Synth. Catal. 2021, 363, 1033–1037
1034
© 2021 Wiley-VCH GmbH

COMMUNICATIONS
asc.wiley-vch.de
were up to 96%. Upon extraction with ethyl acetate, (Table S3). Both variants did not exhibit the signifi-
the crude product was furnished with an isolated yield cantly improved activities. Fortunately, the improved
of 92% and the purity of around 90% (based on HPLC activities were observed with furfural as substrate
analysis).
(Figure 2). Especially, the activity of variant W210F
Although the proof-of-the-concept of reductive was increased over 3 times in reductive coupling of
amination of HMF using AspRedAm was demon- furfural with propylamine, compared to that of
strated, its catalytic activity remained very low. So, a AspRedAm. To get deep insights into the increased
structure-guided semi-rational engineering strategy catalytic activities upon mutation, molecular docking
was applied for improving its catalytic performances. was performed for both the parent and engineered
Based on the molecular docking study (Figure S5), the enzymes in complex with the product N-propylfurfur-
residues located within 4 Å around substrate (i.e., ylamine and NADPH, respectively (Figure S7). Com-
N93, I118, M119, A120, V121, D169, L173, and pared to Trp, Phe having a smaller side chain may
W210) were identified. It was found that the hydrogen allow the substrate to form more reasonable binding
bond interactions occurred between HMF and the mode in the binding pocket, resulting in the increased
residues N93, I118, and D169, whereas substrate docking energy (À 4.02 vs À 4.22 kcal/mol) as well as
formed hydrophobic interactions with M119, V121, the reduced distance between CNADPH and NPRODUCT (4.3
L173, W210, and NADPH. Site-directed mutation on vs 3.2 Å), which contribute to the enhanced activities.
the five residues including N93, I118, V121, L173, and
Finally, the variant W210F was harnessed for the
W210 was conducted, and their catalytic performances synthesis of N-substituted furfurylamines (Table 1).
were compared (Figure S6). It was observed that the Interestingly, the variant W210F displayed excellent
variants W210F and L173N provided slightly positive selectivities (>99%) in almost all the cases, except for
results in reductive amination of HMF. Then, their reductive amination of HMF (entry 5); N-substituted
reductive aminase activities toward cyclohexanone furfurylamines were solely produced. A high substrate
(Table S2), HMF (Table S3), and furfural (Figure 2) conversion as well as a high selectivity was obtained
were determined. Table S2 shows that improved in reductive amination of furfural with propylamine
activities are observed upon W210F mutation. The (Table 1, entry 1). Slight elongation of the chain of the
activity up to 13.5 U/mg was recorded in reductive amines resulted in a lower conversion (90%, entry 2).
coupling of cyclohexanone with cyclopropylamine. In addition, the introduction of functional groups to the
However, the activities of the parent enzyme toward amines exerted a significantly negative effect on the
HMF were very low, ranging from 9.3 to 21.6 mU/mg substrate conversions, but not on the selectivity
(entries 3 and 4). With HMF as the amine accept, the
conversion of 78% was obtained within 12 h, with the
selectivity of 91%. Like furfural, 5-methylfurfural and
5-methoxymethylfurfural are good substrates of the
variant W210F.
In addition, the preparative-scale synthesis of N-
propylfurfurylamine was conducted by using the
variant W210F (Figure S8). Similarly, a high substrate
conversion and an excellent selectivity were observed.
The total turnover number (TTN) for enzyme is
approximately 3200 in the preparative-scale synthesis.
The desired product was obtained with an isolated
yield of 98% and the purity of 93% after organic
solvent extraction. In terms of waste as well as easy
handling, the biocatalytic process appears to be
advantageous over the chemical route based on
stoichiometric strong reductant NaBH4,[7b] since mass
of waste borate is produced in the latter. Also, the
environmental footprint of this bioprocess was as-
sessed (Table 2), based on an E (environmental)
factor.[18] The total E factor is approximately 311.
Although this E factor appears to be high, the solvent
(water, E factor of 300) comprises most of the waste
generated in this bioprocess. In addition, reuse of
propylamine may result in a reduced E factor. Gluconic
acid generated in the bioprocess seems to be more
ecofriendly than borate produced in the abovemen-
Figure 2. Comparison of the activities of AspRedAm and its
variants in reductive amination of furfural. Reaction conditions:
15 mM furfural, 60 mM amines, 0.3 mM NADPH, appropriate
amount of enzyme, 0.4 mL borate buffer (100 mM, pH 8),
°
30 C; final pH of the reactant mixture was tuned to 9 before
addition of enzyme.
Adv. Synth. Catal. 2021, 363, 1033–1037
1035
© 2021 Wiley-VCH GmbH

COMMUNICATIONS
asc.wiley-vch.de
Table 1. Reductive amination of biobased furans by the variant W210F.
Entry
Furan
Amines
Product
Substrate conv. (%)
Select. (%) [a]
1
>99
>99
2
90�2
>99
3
4
51�2
69�2
>99
>99
5
78�1
91� 1
6
7
>99
>99
>99
>99
Reaction conditions: 20 mM furan, 200 mM amine, 30 mM glucose, 0.5 mg/mL (7.8 μM) variant W210F, 0.5 mg/mL GDH, 0.2 mM
°
NADPH, 2 mL Tris-HCl buffer (0.1 M, pH 8), 150 rpm, 25 C, 12 h; tuning final pH to 9 after the addition of propylamine. [a]
defined as the ratio of the amount of the formed furfurylamine to the sum of all the products (in mol).
Table 2. Estimation of the waste generated in the preparative-
scale synthesis (without consideration of purification process).
the engineered enzyme in synthetic chemistry was
demonstrated by the preparative-scale synthesis. A
Component
E factor contribution
modest TTN was obtained in the lab-scale preparation.
The desired products could be readily isolated from the
reaction mixtures by simple and scalable solvent
extraction, with high purity. However, the catalytic
performances of theseRedAms remain unsatisfactory
in the synthesis of N-substituted furfurylamines, and
they may be greatly improved by biocatalyst engineer-
ing strategies (protein engineering and directed
evolution).[20] Overall, biocatalysis proves to be a
promising approach for upgrading the inherently
unstable biobased furans, and its application in
(kg waste per kg product)
Water
300
4.7
4.0
1.8
0.07
0.3
Buffer salt
Propylamine
Gluconic acid
NADPH
Enzymes
Sum
310.9
tioned chemical method.[7b] Also, the biocatalytic route biorefineries would make great contribution to sustain-
seems to be more sustainable than chemically catalytic able circular bioeconomy.
approaches,[8a–d] since the former is totally free of use
of metals (especially noble metals such as Ru, Pd, and
Acknowledgements
Au) as well as high pressure gases (e.g., H2 and CO).
On the other hand, the TTN (3200) obtained in this
bioprocess is much higher than those (200–600)
reported in chemical processes.[4e,8b–d]
In summary, we constructed a novel biocatalytic
approach toward biobased N-substituted furfuryl-
amines by using reductive aminases AspRedAm and its
variant W210F in this work. From the green and
sustainable chemistry viewpoint, this bioprocess offers
attractive synthetic advantages over chemical counter-
parts such as excellent selectivities and bypassing use
of toxic catalysts and solvents.[19] The applicability of
This research was financially supported by the National Natural
Science Foundation of China (21676103 and 31971380), and
the Guangzhou Municipal Science and Technology Project
(201804010179 and 201803010080).
References
[1] a) R. A. Sheldon, Green Chem. 2014, 16, 950–963;
b) R. A. Sheldon, Curr. Opin. Green Sustain. Chem.
Adv. Synth. Catal. 2021, 363, 1033–1037
1036
© 2021 Wiley-VCH GmbH

COMMUNICATIONS
asc.wiley-vch.de
2018, 14, 89–95; c) R. A. Sheldon, Catal. Today 2011,
167, 3–13.
J. D. Vidal, M. J. Climent, P. Concepción, A. Corma, S.
Iborra, ACS Sustainable Chem. Eng. 2019, 7, 6243–
6250; d) M.-M. Zhu, L. Tao, Q. Zhang, J. Dong, Y.-M.
Liu, H.-Y. He, Y. Cao, Green Chem. 2017, 19, 3880–
3887; e) J. J. Roylance, K.-S. Choi, Green Chem. 2016,
18, 5412–5417.
[2] a) M. J. Climent, A. Corma, S. Iborra, Green Chem.
2014, 16, 516–547; b) R.-J. van Putten, J. C. van der
Waal, E. de Jong, C. B. Rasrendra, H. J. Heeres, J. G.
de Vries, Chem. Rev. 2013, 113, 1499–1597; c) J.-P.
Lange, E. van der Heide, J. van Buijtenen, R. Price,
ChemSusChem 2012, 5, 150–166; d) J. J. Bozell, G. R.
Petersen, Green Chem. 2010, 12, 539–554.
[3] a) L. Hu, J. Xu, S. Zhou, A. He, X. Tang, L. Lin, J. Xu,
Y. Zhao, ACS Catal. 2018, 8, 2959–2980; b) Z. Zhang,
K. Deng, ACS Catal. 2015, 5, 6529–6544; c) H. Yuan,
H. Liu, J. Du, K. Liu, T. Wang, L. Liu, Appl. Microbiol.
Biotechnol. 2020, 104, 527–543; d) Y.-Z. Qin, Y.-M. Li,
M.-H. Zong, H. Wu, N. Li, Green Chem. 2015, 17,
[9] a) P. Domínguez de María, N. Guajardo, ChemSusChem
2017, 10, 4123–4134; b) L. Hu, A. He, X. Liu, J. Xia, J.
Xu, S. Zhou, J. Xu, ACS Sustainable Chem. Eng. 2018,
6, 15915–15935; c) X.-Y. Zhang, M.-H. Zong, N. Li,
Green Chem. 2017, 19, 4544–4551; d) X.-Y. Zhang, X.-
Y. Ou, Y.-J. Fu, M.-H. Zong, N. Li, J. Biotechnol. 2020,
307, 125–130; e) X.-Y. Zhang, X. Wang, N.-W. Li, Z.-W.
Guo, M.-H. Zong, N. Li, ChemCatChem 2020, 12, 3257–
3264.
3718–3722; e) Y.-M. Li, X.-Y. Zhang, N. Li, P. Xu, W.- [10] S. Wu, R. Snajdrova, J. C. Moore, K. Baldenius, U.
Y. Lou, M.-H. Zong, ChemSusChem 2017, 10, 372–378;
f) H.-Y. Jia, M.-H. Zong, G.-W. Zheng, N. Li, Chem-
Bornscheuer, Angew. Chem. Int. Ed. 2020, 10.1002/
anie.202006648, https://doi.org/10.1002/anie.202006648.
SusChem 2019, 12, 4764–4768; g) S.-S. Shi, X.-Y. [11] a) A. Petri, G. Masia, O. Piccolo, Catal. Commun. 2018,
Zhang, M.-H. Zong, C.-F. Wang, N. Li, J. Mol. Catal.
2019, 469, 68–74; h) X. Wang, X.-Y. Zhang, M.-H.
Zong, N. Li, ACS Sustainable Chem. Eng. 2020, 8,
4341–4345.
114, 15–18; b) P. Zhang, X. Liao, C. Ma, Q. Li, A. Li, Y.
He, ACS Sustainable Chem. Eng. 2019, 7, 17636–17642;
c) S. A. Kelly, T. S. Moody, B. F. Gilmore, Eng. Life Sci.
2019, 19, 721–725.
[4] a) K. Zhou, B. Chen, X. Zhou, S. Kang, Y. Xu, J. Wei, [12] G. A. Aleku, S. P. France, H. Man, J. Mangas-Sanchez,
ChemCatChem 2019, 11, 5562–5569; b) H. Yuan, J.-P.
S. L. Montgomery, M. Sharma, F. Leipold, S. Hussain,
Li, F. Su, Z. Yan, B. T. Kusema, S. Streiff, Y. Huang, M.
G. Grogan, N. J. Turner, Nat. Chem. 2017, 9, 961–969.
Pera-Titus, F. Shi, ACS Omega 2019, 4, 2510–2516; [13] H.-Y. Jia, M.-H. Zong, G.-W. Zheng, N. Li, ACS Catal.
c) M. Chatterjee, T. Ishizaka, H. Kawanami, Green 2019, 9, 2196–2202.
Chem. 2016, 18, 487–496; d) A. Dunbabin, F. Subrizi, [14] a) K. Hult, P. Berglund, Trends Biotechnol. 2007, 25,
J. M. Ward, T. D. Sheppard, H. C. Hailes, Green Chem.
2017, 19, 397–404; e) V. V. Karve, D. T. Sun, O.
231–238; b) G. Yang, C. Miton, N. Tokuriki, Protein Sci.
2020, n/a, DOI: 10.1002/pro.3901.
Trukhina, S. Yang, E. Oveisi, J. Luterbacher, W. L. [15] a) U. T. Bornscheuer, R. J. Kazlauskas, Angew. Chem.
Queen, Green Chem. 2020, 22, 368–378.
Int. Ed. 2004, 43, 6032–6040; b) R. J. Kazlauskas, Curr.
Opin. Chem. Biol. 2005, 9, 195–201; c) R. B. Leveson-
Gower, C. Mayer, G. Roelfes, Nat. Chem. Rev. 2019, 3,
687–705.
[5] a) X. He, A. H. Conner, J. A. Koutsky, J. Polym. Sci.
Part A 1992, 30, 533–542; b) C. Wang, J. Sun, X. Liu,
A. Sudo, T. Endo, Green Chem. 2012, 14, 2799–2806;
c) X. Shen, J. Dai, Y. Liu, X. Liu, J. Zhu, Polymer 2017, [16] M. Lenz, J. Meisner, L. Quertinmont, S. Lutz, J. Kästner,
122, 258–269. B. M. Nestl, ChemBioChem 2017, 18, 253–256.
[6] a) D. L. Crookes, US, 1985; b) L. H. Schlager, US, [17] M. Lenz, S. Fademrecht, M. Sharma, J. Pleiss, G.
1991.
Grogan, B. M. Nestl, Protein Eng. Des. Sel. 2018, 31,
[7] a) Y. Yang, C. Fei, K. Wang, B. Liu, D. Jiang, B. Yin,
109–120.
Org. Lett. 2018, 20, 2273–2277; b) A. Cukalovic, C. V. [18] R. A. Sheldon, Green Chem. 2017, 19, 18–43.
Stevens, Green Chem. 2010, 12, 1201–1206. [19] P. Jessop, Green Chem. 2020, 22, 13–15.
[8] a) Z. Xu, P. Yan, W. Xu, S. Jia, Z. Xia, B. Chung, Z. C. [20] a) F. H. Arnold, Angew. Chem. Int. Ed. 2018, 57, 4143–
Zhang, RSC Adv. 2014, 4, 59083–59087; b) S. Jiang, E.
Muller, F. Jerôme, M. Pera-Titus, K. De Oliveira Vigier,
Green Chem. 2020, 22, 1832–1836; c) A. García-Ortiz,
4148; b) A. S. Bommarius, J. K. Blum, M. J. Abraham-
son, Curr. Opin. Chem. Biol. 2011, 15, 194–200.
Adv. Synth. Catal. 2021, 363, 1033–1037
1037
© 2021 Wiley-VCH GmbH