602
A. Pordea et al. / Inorganica Chimica Acta 363 (2010) 601–604
L)Cl] (3.8 lL of a 0.0395 M stock solution in DMF, 0.15 lmol Ru)
and stirred at room temperature for 10 min. A degassed mixture
of HCOONa (1 M) and B(OH)3 (0.85 M) was added to each tube
(600
(200
l
l
L mixture, pH adjusted to 6.25), followed by MOPS buffer
L of a 1 M stock solution, pH adjusted to 6.25) and by the
corresponding substrate (15
lL of a 1 M stock solution in DMF,
15 mol). The test tubes were placed in a magnetically stirred mul-
l
tireactor, purged several times with nitrogen and heated at 55 °C
for 64 h. After completion, the reaction mixture was extracted with
Et2O (4 Â 1 mL) and dried over Na2SO4. The organic solution was
filtered through a short silicagel plug that was thoroughly washed
with Et2O, concentrated and subjected to HPLC analysis.
For 1-(p-methylphenyl)ethanol, the conversions and ee’s were
determined using an (S,S)-ULMO column (Regis Technologies Inc.,
IL, USA) with hexane:1,2-dimethoxyethane 100:1.5 at 1 mL/min;
tR = 19.5 min, tS = 16.6 min (UV detection at 215 nm, absolute con-
figuration assigned by comparison with published results [11]).
For 1,2,3,4-tetrahydro-1-naphthol, the conversions and ee’s
were determined using an OD-H column (Daicel Chemical Indus-
tries, Tokyo) with hexane:isopropanol 100:2 at 0.9 mL/min;
tR = 18.5 min, tS = 17.2 min (UV detection at 215 nm, absolute con-
figuration assigned by comparison with a commercial enantiopure
sample).
3. Results and discussion
The crystal structure of the
[g
6-(benzene)Ru(Biot-p-
L)Cl]ꢀS112K Sav tetramer (the inclusion symbol ꢀ refers to the
supramolecular incorporation of the biotinylated ruthenium com-
plex into streptavidin) revealed that the residues of the L7,8 loop
(residues 112–121) of both streptavidin monomers A and B were
situated in proximity to the catalytic site (Fig. 1a). Therefore,
site-directed mutations in this region were expected to influence
catalysis by interactions between the protein and the substrate
or the ruthenium complex. Previous results allowed the identifica-
tion of [
g
6-(p-cymene)Ru(Biot-p-L)Cl]ꢀS112A as a highly (R)-
enantioselective artificial enzyme [11]. In order to perturb the loop
structure, three neighboring polar amino acid residues from this
loop were mutated with glycine or alanine residues and the new
Sav variants, T114G Sav, T115A Sav and E116A Sav, were tested
in catalysis in the presence of the two biotinylated Ru complexes.
The T114G Sav mutation has a significant influence on the reaction
Fig. 1. Close-up views of the crystal structure of
[g
6-(benzene)Ru(Biot-p-
L)Cl]ꢀS112K Sav tetramer [12]. Only monomer A is occupied by the biotinylated
ruthenium complex. (a) The L7,8 loops of monomers A (blue) and B (yellow) are
highlighted in green; residues S112, T114, T115 and E116 of both monomers are
situated in the proximity of the catalytic site. (b) The neighboring monomers A
(blue) and D (orange) are highlighted; the L4,5 loop (red) of monomer D, containing
the P64 residue, is situated in the proximity of the L7,8 loop (green) of monomer A;
the L124 residue of monomer A is also highlighted. (For interpretation of the
references to colour in this figure legend, the reader is referred to the web version of
this article.)
outcome: in combination with [g
6-(p-cymene)Ru(Biot-p-L)Cl],
this mutation appears as beneficial for enantioselectivity when
compared to WT Sav (Table 1, entries 1–4), while the (S)-enanti-
oselectivity for the reduction of
a-tetralone is increased to 84%
ee (S) when using
[g
6-(benzene)Ru(Biot-p-L)Cl]ꢀT114G Sav
(Table 1, entries 5–6). The T115A and E116A mutations were not
as effective as T114G Sav in terms of enantioselectivity improve-
ment (Table 1, entries 7–8). To further increase the selectivity,
we combined the S112A and the T114G mutations, but the result-
ing double mutant afforded lower conversion and selectivity than
the corresponding single mutants (Table 1, entries 9–12). This
antagonistic behavior could be due to the fact that the 112 and
114 amino acids lie too close to each other and thus the two
modifications might drastically perturb the enantioselective
environment around the catalytic site.
2. Experimental
2.1. Protein expression and purification
Streptavidin mutants were produced, purified and quantified
according to Ref. [13].
2.2. Catalytic runs
Preliminary experiments investigating the effect of genetic
modifications of the host protein on the reduction of acetophenone
derivatives showed that the streptavidin mutant with the most
remote site of mutation (P64G Sav) significantly increased
enantioselectivity, while the closest-lying mutation (S112G Sav)
afforded a marked decrease in enantioselectivity, although with
higher conversions (Table 1, entries 13–14). Interestingly, an
enhanced activity and selectivity was observed upon combination
The substrates were commercially available. The biotinylated
ruthenium complexes, [g -
6-(p-cymene)Ru(Biot-p-L)Cl] and [g6
(benzene)Ru(Biot-p-L)Cl], were synthesized as previously
published [11]. Lyophilized streptavidin was dissolved in milliQ
water (100
thoroughly degassed (nitrogen flushed during 2 h). The degassed
protein (450 L solution, 0.045 mol of tetrameric Sav) was mixed
in a test tube with the precursor complex [
6-(arene)Ru(Biot-p-
lM tetrameric concentration) and the solution was
l
l
g
Pordea
