S. Tanaka, H. Kojima, S. Takeda et al.
Tetrahedron Letters 77 (2021) 153249
ee), although the yields were low (38–39%; Table 1, entries 10 and
1
2). Thus, a larger effect on the stereoselectivity was observed for
solvent molecules with a benzene or cyclohexane ring. However,
hexane only had a small effect on the stereoselectivity (7% ee (S);
Table 1, entry 11). Dodecane, a hydrophobic solvent (log P = 6.6)
with a straight long chain, did not affect the stereoselectivity at
all (79% ee (R); Table 1, entry 13).
Scheme 1.
including dimethyl sulfoxide (DMSO), ethanol, tetrahydrofuran
THF), pyridine, chloroform, benzene, toluene, ethylbenzene, cyclo-
Next, we investigated the effect of light conditions on the
reduction reaction. Reactions of 1a (10 lg/mL) with Synechocystis
(
hexane, hexane, methylcyclohexane, and dodecane, were exam-
ined. The yields and ee values were determined by gas
chromatography (GC).
sp. PCC 6803 (Abs680-750 = 0.33) in the absence or presence of
toluene (1% (v/v)) were carried out under illumination with red
À2 À1
LED light (660 nm, 10
l
mol photons m
s ) and in the dark at
The results in Table 1 show that the addition of polar organic
solvents had little effect on the stereoselectivity. Thus, solvents
with low log P values (log P < 0) such as DMSO and ethanol did
not affect the stereochemical course of the reduction, and tert-
butyl (R)-3-hydroxybutanoate ((R)-2a) was obtained (79–85% ee;
Table 1, entries 1–3). In the case of less polar solvents such as
THF and pyridine, both yields and ee decreased (20–44%, 42–56%
ee (R); Table 1, entries 4 and 5). Next, solvents with higher log P
values (log P = 2.0–6.6) were tested. Intriguingly, chloroform and
benzene, a slightly polar solvent (log P = 2.0) afforded (S)-3-
hydroxybutanoate ((S)-2a), the (S)-isomer of the product, with an
excellent ee (>99%; Table 1, entries 6 and 7). Thus, the stereochem-
ical course of the reduction was changed from (R) to (S) only by
adding only 1% chloroform or 1% benzene. Although the yield of
25 °C for 24 h. The yields of (R)- and (S)-2a and the ee values are
listed in Table 2.
The production of (R)-alcohol decreased in the dark (from 72%
to 21%; Table 2, entries 1 and 2). We previously reported that R-
selective reduction by the cyanobacterium was activated by light
illumination and that R-selective reductase required NADPH recy-
cling through photosynthesis [10]. Under dark conditions, NADPH
regeneration was suppressed, which may have reduced the forma-
tion of (R)-alcohol. When 1% toluene was added, (R)-alcohol was
not formed at all, not only under dark conditions, but also under
light conditions. (Table 2, entries 3 and 4). Therefore, we hypothe-
size that toluene inhibited NADPH regeneration through
photosynthesis.
To determine whether the photosynthetic system of the
cyanobacteria was working properly, the absorbance at 680 nm
(Abs680-750) in the UV–vis spectra of the cyanobacteria was mea-
sured in the presence of an organic solvent. Abs680-750 represents
the amount of chlorophyll a—one of the most important light-har-
vesting pigments in the photosynthetic system. The Abs680-750 and
ee values are listed in Table 3. The UV–vis spectra (400–750 nm) of
the cell suspension after the reaction with/without organic sol-
vents are shown in Fig. S1. In the case of DMSO, THF, and dodecane,
almost no change in absorbance was observed (Table 3, entries 1–3
and 7). In the case of benzene, toluene, and hexane, Abs680-750
decreased and the S-selectivity was improved (Table 3, entries
(
S)-2a was low (17–25%) in the case of chloroform and benzene,
the use of toluene dramatically improved the yield of (S)-2a
87%) with an extremely high ee (>99%; Table 1, entry 8). Similar
(
results were obtained for ethylbenzene (>99% ee (S)), although
the yield of (S)-2a was slightly lower (66%; Table 1, entry 9). Cyclo-
hexane and methylcyclohexane yielded (S)-2a with a high ee (>99%
Table 1
Asymmetric reduction of b-keto ester 1a using cyanobacterium Synechocystis sp. PCC
6
803.a
Organic solventb
Yield, %c
ee, %c
4–6). The decrease in Abs680-750 was the greatest in the case of
toluene (Abs680-750 = 0.15; Table 3, entry 5). These results suggest
Entry
log P
1
2
3
4
5
6
7
8
9
1
1
1
1
None
DMSO
Ethanol
THF
Pyridine
Chloroform
Benzene
–
81
77
69
44
20
17
25
87
66
39
29
38
94
79 (R)
84 (R)
85 (R)
56 (R)
42 (R)
>99 (S)
>99 (S)
>99 (S)
>99 (S)
>99 (S)
7 (S)
À1.3
À0.24
0.49
0.71
2.0
2.0
2.5
3.2
3.2
Table 3
Abs680 and ee with some organic solvents.a
Toluene
Entry
Organic solvent
Abs680-750
ee, %
Ethylbenzene
Cyclohexane
Hexane
Methylcyclohexane
Dodecane
0
1
2
3
1
2
3
4
5
6
7
None
DMSO
THF
Benzene
Toluene
Hexane
Dodecane
0.34
0.33
0.33
0.24
0.15
0.26
0.33
79 (R)
84 (R)
56 (R)
>99 (S)
>99 (S)
7 (S)
3.5
3.6
6.6
>99 (S)
79 (R)
a
Reaction conditions: 1a (10
4 h; red LED light (660 nm, 10
Organic solvent: 1% (v/v).
Determined by GC. All values are the mean of three results (n = 3).
l
l
g/mL); cyanobacteria (Abs680-750 = 0.33); 25 °C,
À2 À1
2
mol photons m
s ).
79 (R)
b
c
a
Reaction conditions: 1a (10
l
g/mL); cyanobacteria (Abs680-750 = 0.33); organic
À2 À1
solvent (1% (v/v)); 25 °C, 24 h; red LED light (660 nm, 10
l
mol photons m
s ).
Table 2
Effect of light conditions.a,b
Entry
Toluene, %
Light cond.
(R)-2a, %b
(S)-2a, %b
ee, %b
1
2
3
4
0
0
1
1
Light
Dark
Light
Dark
72
21
0
8.5
13
87
75
79 (R)
25 (R)
>99 (S)
>99 (S)
0
a
À2 À1
Reaction conditions: 1a (10
Determined by GC. All values are the mean of three results (n = 3).
l
g/mL); cyanobacteria (Abs680-750 = 0.33); toluene (0 or 1% (v/v)); 25 °C, 24 h; red LED light (660 nm, 10
l
mol photons m
s
).
b
2