Angewandte
Communications
Chemie
Table 1: Light-catalyzed whole-cell biotransformation of different substrates using Synechocystis sp. PCC 6803 cells with recombinant enoate reductase
À1
YqjM at a cell density of 1.8 gL
.
[a]
Substrate
Substrate
concentration
Time
Specific
Max. conversion
Conversion
Space-time-yield
Product
Cell
productivity
[ggCDW
[
a,b]
activity
[UgCDW
rate
concentration
À1
À1 [a]
À1 À1
À1
À1
[
mm]
[h]
]
[mmh
]
[%]
[gL
h
]
[gL
]
]
[d]
1
2
3
4
5
6
6
7
a
a
a
a
a
a
a
a
15
15
10
15
15
20
10
20
4
24
24
24
24
3.3
1
39
21.1
6.2
25.6
53.2
90.9
123
99.5
4.1
2.2
0.7
2.7
5.7
9.6
10
70
42
57
99
94
99
99
99
0.03
0.03
0.04
0.5
0.64
0.67
1.1
0.1
0.71
0.9
0.06
0.4
0.5
0.72
0.46
1.24
0.62
1.05
1.2
0.83
2.24
1.1
[
c]
3
10.5
0.63
1.9
[
a] Determined by gas chromatography. [b] Determined after 1 h. [c] 100 mg scale. [d] 0.5% formation of cyclohexanol was observed.
smoothly by the system. In several cases, depletion of the
substrate was observed. A loss of up to 20% of the volatile
substances 1b and 2b is likely due to evaporation. It is worth
mentioning here that, with the exception of ketoisophorone
the effectiveness of NADPH formation, and side reactions
leading to the degradation of the substrate or product. The
maximum rate of formation is clearly influenced by the
enzyme concentration and its activity, the availability of the
reduced cofactor NADPH, and the efficiency of the transport
of the substrate and product across the cell membrane.
Biotransformation experiments at different substrate concen-
trations show an initial maximum rate of product formation,
thus indicating that the enzyme can work at full capacity
under the optimized conditions.
3
a, all the prochiral starting materials were converted with
high enantioselectivity (Table 1). Comparative experiments
with enantiopure levodione (R)-3b, however, showed that the
product readily racemizes under the reaction conditions (data
[
3c]
not shown).
À1
Volumetric productivities of up to 10 mm h and product
À1
titers of up to 2 gL were obtained. Although this is still too
At higher substrate concentrations, toxicity effects
decrease the viability of the cells, which slows down the
conversion. Furthermore, it is frequently observed in whole-
cell biotransformations that the fed substrates and products
are metabolized by the cellular metabolism. The targeted
low for industrial applications, it represents an encouraging
starting point for further optimization. More importantly, the
À1
specific activity of the biocatalyst reached 100 Ug , thereby
approaching activities required for industrial application. A
preliminary calculation showed that the observed rate of
[13]
deletion of degrading pathways or an optimization of the
reaction conditions (as reported here for 6a) can be used to
circumvent this challenge.
À1
À1
product formation of 2 mm h with 1.8 gCDW
L
corresponds
À1
to a turnover rate of 20000 s per cell. A comparison of
substrates 1a–3a shows that small differences in the substrate
structure have a strong impact on the specific activity.
Increasing the space-time-yield and widening the substrate
scope either by using different enoate reductases or by
reaction engineering and enzyme engineering will be required
to achieve a successful industrial application. His-tag purifi-
cation enabled 8.4 mg enzyme to be purified from 1 g cells.
The resulting total turnover number of 18750 per enzyme
molecule indicates that the enantioselective reduction is
indeed a catalytic process. To clarify the role of the amount of
active enzyme, we compared the conversion of 6a by two
YqjM variants bearing different protein tags. The fusion of
the N terminus of enoate reductases with different protein
To the best of our knowledge, this study presents the first
example of the use of the cyanobacterial photosystem to fuel
biotransformations with recombinant enzymes. Although
[
14]
visible-light-mediated photochemo-
tic
and photobiocataly-
[
5b,c,6a,15]
reactions are a very promising concept, the strength
of the reported approach is the simplicity and the ease of the
production of the biocatalyst. In principle, the approach
presented here can be extended to all oxidoreductases that
use NADPH, NADH, or ferredoxin as the electron donor.
The high conversion and enantioselectivity reported offer the
possibility of using cofactor regeneration with the combina-
tion of water and light. The tremendous progress in the
molecular biology of photosynthetic microorganisms provides
the tools for fine-tuning the expression and thus further
optimization. Despite very promising examples of the appli-
[12]
tags influences the folding and specific activity. Cells with
YqjM bearing a streptavidin-tag showed a specific activity
À1
[16]
towards 6a of 12 UgCDW , while His-tagged YqjM showed
cation of cyanobacteria as producers of organic molecules,
À1
À1
a specific activity of 91 UgCDW . This comparison shows that
product titers are usually in the mgL range. In contrast,
the amount of active enzyme in the cell is an important factor
for the activity of the biocatalyst.
whole-cell photo-biotransformations are possible with several
gL . Although the product concentrations and space-time-
À1
Finally, we performed one photobiocatalytic reaction
under semipreparative conditions. After three hours, a bio-
transformation of 100 mg 6a resulted in 81 mg of optically
pure (R)-6b (80% yield of isolated product, > 99% ee).
The data show that the productivity of the photoauto-
trophic production system depends on three substrate-depen-
dent factors, namely the activity of the overexpressed enzyme,
yields reported here are rather low for industrial processes,
the system has not yet been optimized. Standard molecular
biology tools such as controllable promoters that are cur-
rently developed for Synechocystis offer considerable poten-
tial for improvement. Moreover, cyanobacteria utilize only
a small part of the light energy because they prioritize
“photosynthetic robustness”, that is, protection strategies
5
584
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 5582 –5585