Evaluation of Chemical Diversity of Biotinylated Chiral 1,3-Diamines as a Catalytic Moiety in Artificial Imine Reductase

Article abstract of DOI:10.1002/cctc.201600116

The possibility of obtaining an efficient artificial imine reductase was investigated by introducing a chiral cofactor into artificial metalloenzymes based on biotin-streptavidin technology. In particular, a chiral biotinylated 1,3-diamine ligand in coord

Full text of DOI:10.1002/cctc.201600116

DOI: 10.1002/cctc.201600116
Full Papers
Evaluation of Chemical Diversity of Biotinylated Chiral
1,3-Diamines as a Catalytic Moiety in Artificial Imine
Reductase
Michela Pellizzoni,[b] Giorgio Facchetti,[a] Raffaella Gandolfi,[a] Marco Fus›,[a]
Alessandro Contini,[a] and Isabella Rimoldi*[a]
The possibility of obtaining an efficient artificial imine reduc-
tase was investigated by introducing a chiral cofactor into arti-
ficial metalloenzymes based on biotin–streptavidin technology.
In particular, a chiral biotinylated 1,3-diamine ligand in coordi-
nation with iridium(III) complex was developed. Optimized
chemogenetic studies afforded positive results in the stereose-
lective reduction of a cyclic imine, the salsolidine precursor, as
a standard substrate with access to both enantiomers. Various
factors such as pH, temperature, number of binding sites, and
steric hindrance of the catalytic moiety have been proved to
affect both efficiency and enantioselectivity, underlining the
great flexibility of this system in comparison with the achiral
system. Computational studies were also performed to explain
how the metal configuration, in the proposed system, might
affect the observed stereochemical outcome.
Introduction
Considering that transition-metal complexes and enzymes pos-
sess unique and often complementary properties as catalysts,
in the last few decades researches have tried to combine the
structure and the reactivity of metals and peptides, in a so-
called hybrid metal–peptide catalysts[1] for the synthesis of
enantiomerically enriched compounds in aqueous media. In
more general terms, the development of a hybrid catalyst re-
sults from the combination of a biological scaffold (e.g., pro-
teins,[2] DNA,[3] or peptides[4]) with an active catalytic moiety.
Among several anchoring strategies exploited to embed chem-
ical complexes within biomolecules (dative, covalent, or supra-
molecular interactions), the use of biotin–streptavidin technol-
ogy in forming artificial metalloenzymes has been explored ex-
tensively. The success of this approach is related to the ease of
self-assembly of these artificial metalloenzymes, which allows
for rapid optimization. To improve the performance of hybrid
catalysts, the chemogenetic approach was found to be the
most suitable strategy.[5] This strategy concerns two distinct
optimizations: 1) genetic modification of the protein scaffold,
based on computational calculations and X-ray structures; par-
ticular attention is given to the active site of enzyme (<10 Š
around the catalytic metal center);[6] 2) chemical fine-tuning of
the cofactor to adjust the localization of the catalytic moiety
inside the funnel-shaped cavity of the protein.[7]
The Ward group[1j,2a,8] has relied on synthesizing several bio-
tinylated ligands to provide chemical diversity by using achiral
five-membered ring chelates. With the aim of studying the ro-
bustness of this strategy and to underline a different type of
interaction between the protein scaffold and metal catalytic
environment, chiral 1,3-diamines were also investigated for the
enantioselective reduction of cyclic imines.
Results and Discussion
Based on the biotin–streptavidin technology studied in depth
by the Ward group, we synthesized six different biotinylated li-
gands starting from simple chiral 1,3-diamines, previously re-
ported by our group, which are able to be used as ligands in
RuII complexes for the asymmetric transfer hydrogenation
(ATH) of ketones with interesting results regarding the differen-
ces in reactivity, especially with respect to different types of re-
action media.[9]
Relying on the biotin anchor and the spacer, the catalytic
moiety was redesigned by substituting classical 1,2-diamines
ligand with enantiopure 1,3-diamine, thus generating a six-
membered chelating ring and inserting a bulky substituent at
the chiral center. The introduction of these two features could
allow us to evaluate the versatility of this supramolecular an-
choring approach and see how the change in chiral environ-
ment might affect the position of the catalytic moiety within
the protein (chiral host), underlining the importance of secon-
dary interactions to achieve selectivity.
[a] Dr. G. Facchetti, Dr. R. Gandolfi, Dr. M. Fus›, Dr. A. Contini, Dr. I. Rimoldi
Dipartimento di Scienze Farmaceutiche
Università degli Studi di Milano
Via Venezian 21, 20133 Milano (Italy)
E-mail: isabella.rimoldi@unimi.it
[b] Dr. M. Pellizzoni
Department of Chemistry
University of Basel
Spitalstrasse 51, 4056 Basel (Switzerland)
The synthesis of the starting amino alcohols was achieved as
previously reported.[9] The reaction of para-, meta-, or ortho-ni-
trobenzenesulfonyl chloride in the presence of triethylamine
Supporting information and the ORCID identification number(s) for the
author(s) of this article can be found under http://dx.doi.org/10.1002/
cctc.201600116.
ChemCatChem 2016, 8, 1665 – 1670
1665
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Scheme 1. Synthesis of biotinylated cofactors 7.
(TEA) produced the corresponding nitrobenzenesulfonamides.
Then, the corresponding azido derivatives were achieved after
the addition of PPh3 and NaN3.[10] In our case, a strong depend-
ence on the solvent did not allow us to directly obtain the cor-
responding amine in the 3-position. The selective reduction of
the azido moiety, without effecting the nitro group, was ach-
ieved by using a Staudinger reduction[11] to yield compound 3.
The amino function was protected by employing di-tert-butyl
dicarbonate and finally the nitro group was reduced by using
molecular hydrogen in the presence of Pd/C, thus obtaining
compound 5. A condensation reaction with biotin[12] was per-
formed and the amino function at the chiral center was depro-
tected by using trifluoroacetic acid (Scheme 1).
complex in 1 mL of MOPS (3-(N-morpholino)propanesulfonic
acid) buffer at pH 7 in the presence of HCOONa (3m) as the
hydride source. After 10 h, the conversion was complete but
only 5–6% enantioselectivity was achieved. As shown previ-
ously for reactions catalyzed by ruthenium(II) complexes,[9] the
conformational freedom of the chelating six-membered ring
might negatively affect enantioselectivity. However, by insert-
ing a bulky substituent at the 3-position and introducing these
complexes in a rigid supramolecular system (Sav/biotinylated
metal complex), we expect to increase the energy barrier for
conformational change.
Nevertheless, considering that the reduction of these sub-
strates is often not easy to obtain, our preliminary results in
terms of reaction conversion could be considered a good start-
ing point to study the development of artificial imine reductas-
es by using these chiral 1,3-diamines as ligands in transition-
metal complexes.[14]
The biotinylated ligands were used for the preparation of
the corresponding IrIII d6 piano-stool precatalysts, in quantita-
tive yield, as yellow–orange powders.
The salsolidine precursor was selected as a model substrate
(Figure 1). Indeed, given the prevalence of the chiral 1,2,3,4-tet-
rahydroisoquinoline motif in natural alkaloids and synthetic
In this study, streptavidin WT and mutant S112X were select-
ed for screening. Indeed, previous computational studies and
the crystal structure of Sav/biotinylated iridium(III) hybrid com-
plexes, in which the ligand was achiral ethylenediamine,[8a,12]
indicated residue 112 as one of the most critical in determining
the final stereochemical outcome.
As an initial screening, we investigated the catalytic activity
of the d6 IrIII piano-stool bearing biotinylated aminosulfona-
mide ligands (7) as bare catalysts and their combination with
wild-type streptavidin (Sav WT, hereafter) for the production of
salsolidine, taking account of different reaction parameters
such as pH, temperature, substrate concentration, and catalyst
loading (Table 1).
Figure 1. Artificial imine reductases for ATH with chiral 1,3-diamine scaffolds.
The results, displayed in Table 1, illustrate the very high ac-
tivity of the iridium piano-stool biotinylated aminosulfonamide
catalysts. In all cases, however, modest stereoselectivity oc-
curred. A similar behavior was observed for all diverse chemical
entities (para, meta, or ortho derivatives), suggesting the ineffi-
cacy of the chiral center on a flexible six-membered chelating
ring. Conversely, when the transition-metal catalyst is embed-
ded in Sav WT, a slight decrease in activity was observed
(Table 1, entries 2, 4, 8, 10, and 12). An improvement of stereo-
drugs,[13] the development of methods for its production in
enantiopure form is highly desirable.
A preliminary study with 1,3-diamines as ligands in d6 Ir
piano-stool complexes was realized by using a dihydroisoquino-
line analog in aqueous media. Reactions were carried out at
408C by using 5 mmol of substrate with 0.5 mol% of iridium
ChemCatChem 2016, 8, 1665 – 1670
www.chemcatchem.org
1666
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Table 1. Optimized screening with or without Sav WT.[a]
Entry Protein Complex pH T
Conversion ee (config)[b]
Table 2. Optimized screening with S112X and [Cp*Ir(Biot-7)Cl].
Entry Protein Complex
pH
T
Conversion ee (config)
[8C] [%] [%]
[8C] [%]
[%]
1
2
3
4
5
6
7
8
–
[Cp*Ir(Biot-p-(R)-7)Cl] 7.0 40 >99
6 (S)
1
2
3
4
5
6
7
8
S112K [Cp*Ir(Biot-p-(R)-7)Cl] 7.0 40 47[a]
S112R [Cp*Ir(Biot-p-(R)-7)Cl] 7.0 40 43[a]
S112C [Cp*Ir(Biot-p-(S)-7)Cl] 6.5 30 40[b]
S112C [Cp*Ir(Biot-p-(S)-7)Cl] 6.5 30 60[c]
S112Y [Cp*Ir(Biot-m-(R)-7)Cl] 7.0 40 72[c]
S112Y [Cp*Ir(Biot-m-(R)-7)Cl] 7.0 40 72[c,d]
S112C [Cp*Ir(Biot-m-(S)-7)Cl] 6.5 30 26[b]
S112E [Cp*Ir(Biot-m-(S)-7)Cl] 6.5 30 79[b]
S112M [Cp*Ir(Biot-o-(R)-7)Cl] 7.0 40 95[a]
S112Q [Cp*Ir(Biot-o-(S)-7)Cl] 6.5 30 78[b]
S112E [Cp*Ir(Biot-o-(S)-7)Cl] 6.5 30 73[b]
12 (S)
12 (S)
58 (R)
65 (R)
47 (S)
37 (S)
40 (R)
12 (R)
9 (R)
Sav WT[c] [Cp*Ir(Biot-p-(R)-7)Cl] 7.0 40
58
5 (R)
5 (R)
18 (R)
4 (S)
5 (R)
À
–
[Cp*Ir(Biot-p-(S)-7)Cl] 6.5 30 >99
Sav WT[c] [Cp*Ir(Biot-p-(S)-7)Cl] 6.5 30
83
–
[Cp*Ir(Biot-m-(R)-7)Cl] 7.0 40 >99
Sav WT[c] [Cp*Ir(Biot-m-(R)-7)Cl] 7.0 40 >99
[Cp*Ir(Biot-m-(S)-7)Cl] 6.5 30 >99
Sav WT[d] [Cp*Ir(Biot-m-(S)-7)Cl] 6.5 30
76
–
8 (R)
À
9
–
[Cp*Ir(Biot-o-(R)-7)Cl] 7.0 40 >99
9
10
11
10
11
12
Sav WT[c] [Cp*Ir(Biot-o-(R)-7)Cl] 7.0 40
90
À
19 (R)
18 (R)
–
[Cp*Ir(Biot-o-(S)-7)Cl] 6.5 30 >99
À
Sav WT[d] [Cp*Ir(Biot-o-(S)-7)Cl] 6.5 30
72
À
[a] Reactions were carried out at 408C for 24 h by using 1 mol% com-
plex 7 with or without 0.33 mol% tetrameric WT or mutant Sav in 1.2m
MOPS buffer, 3m HCOONa, pH 7.0, [sub]f =28 mm, [cat]f =0.28 mm.
[b] Reactions were carried out at 308C for 48 h by using 1 mol% com-
plex 7 and 0.33 mol% tetrameric WT or mutant Sav in 1.2m MOPS buffer,
3m HCOONa, pH 6.5, [sub]f =35 mm, [cat]f =0.35 mm. [c] 1 mol% com-
plex 7 and 0.66 mol% tetrameric mutant Sav. [d] [sub]f =35 mm, [cat]f =
0.35 mm.
[a] All reactions were carried out for 24 h by using 1 mol% complex 7
and with or without 0.33 mol% tetrameric Sav WT in 1.2m MOPS buffer,
with 3m HCOONa [b] Conversion was obtained by HPLC by using a cor-
rection factor of 1.32 at l=283 nm. Enantiomeric excess was determined
by using HPLC equipped with a chiral OD-H column. Eluent: hexane/
EtOH=95:5, 0.1% diethanolamine (DEA), flow=0.8 mLminÀ1. [c] [sub]f =
28 mm, [cat]f =0.28 mm. [d] [sub]f =35 mm, [cat]f =0.35 mm.
selectivity, always leading to a reaction product in the (R)-con-
figuration, was obtained with the para iridium catalyst (Table 1,
entry 4, 18% ee). For [Cp*Ir(Biot-m-(R)-7)Cl] and [Cp*Ir(Biot-p-
(R)-7)Cl] (Cp*=pentamethylcyclopentadienyl) in the presence
of Sav WT, comparing the bare catalyst and the hybrid catalyst,
a modest inversion of product configuration was highlighted
(Table 1, entry 1 versus 2 and entry 5 versus 6).
metal configuration in the [Cp*Ir(Biot-p-(S)-7)H]ꢀSav S112C
and [Cp*Ir(Biot-m-(R)-7)H]ꢀSav S112Y (hereafter abbreviated as
p-S112C and m-S112Y, respectively) might affect the reaction
stereoselection.
By visually inspecting the (RIr)- and (SIr)-p-S112C models
(Figure 2), it was immediately evident that only the former
A genetic optimization of the Sav host protein was then per-
formed and different mutants at position 112 were tested (see
the table in the Supporting Information).
Selected results for the chemical and the genetic optimiza-
tion in the presence of [Cp*Ir(Biot-7)Cl] are summarized in
Table 2.
To provide a rationale behind the activity of the catalysts re-
ported here, a computational study was performed by evaluat-
ing two different mutants, namely S112C and S112Y, at position
112 (Table 2, entries 4 and 5).[8b] Indeed, this position is close to
the biotinylated metal when incorporated into streptavidin in
the presence of the achiral 1,2-diamine used as the ligand.
Models were generated starting from the tetrameric biological
assembly of the X-ray crystal structure reported in the litera-
ture (PDB: 3K2).[8a] Only monomers A and C were considered
to build a dimer with a fully functional binding site for one
biotinylated ligand. The Ser residue at position 112 was then
mutated to Cys or to Tyr and the sidechain positioning was
evaluated by a systematic conformational search using the
MOE software (see the Supporting Information for further com-
putational details).[15] The original ethylenediamine ligand was
then modified to obtain [Cp*Ir(Biot-p-(S)-7)Cl]ꢀSav S112C and
[Cp*Ir(Biot-m-(R)-7)Cl]ꢀSav S112Y. Finally, the chloride was sub-
stituted with a hydride to mimic the active catalyst, thus caus-
ing an inversion of configuration at the metal center according
to the Cahn–Ingold–Prelog rule.[16] For each complex, both
pseudoenantiomers at the metal center (hereafter indicated as
RIr and SIr) were generated with the aim to clarify how the
Figure 2. Computational models of A) (RIr)-[Cp*Ir(Biot-p-(S)-7)H]ꢀSav S112C
complex bound to the favored pro-R configuration of salsolidine precursor;
B) unbound (SIr)-[Cp*Ir(Biot-p-(S)-7)H]ꢀSav S112C complex; C) (SIr)-[Cp*Ir(Biot-
m-(R)-7)H]ꢀSav S112Y complex bound to the favored pro-S configuration of
the salsolidine precursor; D) unbound (RIr)-[Cp*Ir(Biot-m-(R)-7)H]ꢀSav S112Y.
For all models, chains A and C are represented as green or blue molecular
surfaces, respectively.
structure is suitable for the hydride transfer. Indeed, with the
metal center in the RIr configuration the reactive center is ex-
posed outside of the biotinylated ligand binding site (Fig-
ure 2A). Conversely, when the metal adopts the SIr configura-
tion, the cyclopentadienyl moiety is solvent-exposed and the
ChemCatChem 2016, 8, 1665 – 1670
www.chemcatchem.org
1667
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
reactive hydride atom is completely hindered by the protein
loop spanned by residues 112–122 (Figure 2B).
biotin binding sites) yields the best results (entries 3 versus
4 and entry 5).
Similarly, when considering the (SIr)- and (RIr)-m-S112Y
models, it can be observed that in the former complex the hy-
dride atom can be easily accessed by the salsolidine precursor
(Figure 2C), whereas in the latter, the reaction center is hin-
dered by the sidechain of Lys121 in protein chain C (Fig-
ure 2D).
4) When [Cp*Ir(Biot-(R)-7)Cl] was incorporated into Sav mu-
tants, the preferred chemical diversity was the meta posi-
tion of the linker; conversely, the iridium precatalyst bear-
ing the S diamine matched better with Sav when the para
position of this one was realized. These data confirm the
role of the bulky phenyl group on the stereocenter in influ-
encing the ability of the iridium biotinylated catalyst to be
deeply anchored into the biotin-binding cavity in accord-
ance with our computational findings and “induced lock-
and-key hypothesis”.[16]
Finally, we also evaluated the potential binding mode of the
cyclic imine 1-methyl-6,7-dimethoxy-3,4-dihydroquinoline,
which was used as a standard substrate (see the Supporting
Information). The precursor was manually docked in the (RIr)-p-
S112C and (SIr)-m-S112Y models, according to the indication re-
ported in ref. [17], evaluating several possible substrate orien-
tations in both the pro-R and pro-S configurations. Concerning
the (RIr)-p-S112C complex, the docked pro-R configuration (Fig-
ure 2A) was favored over the pro-S configuration (Figure S1A,
in the Supporting Information) by 0.9 kcalmolÀ1, in agreement
with experiments, even if this result should be treated with
caution as the energy difference is beyond the accuracy limit
of the method. In both cases, the substrate is stabilized by hy-
drogen-bonds between the two methoxy groups and the
5) The best results in terms of enantioselectivity were ob-
tained with [Cp*Ir(Biot-p-(S)-7)Cl]ꢀSav S112C (entry 4, 66%
(R) ee and 60% yield) and with [Cp*Ir(Biot-m-(R)-
7)Cl]ꢀS112Y (entry 5, 47% (S) ee and 72% yield). A rationale
for this finding was provided by our computational study.
6) Generally, the stereoselection in favor of the S enantiomer
product was accomplished with a lower ee, just like in the
results obtained with ethylenediamine as ligand.[8a] The un-
fulfilled enantioselectivity in these cases was probably due
to the preference for the R enantiomer as evident in all
cases with the Sav WT system (Table 1 versus Table 2).
7) When diamine ligands were in the R configuration, the re-
duction product salsolidine was obtained in the S configu-
ration; an exception was with the [Cp*Ir(Biot-o-(R)-7)Cl] cat-
alyst, which afforded the product in R configuration albeit
in a very modest 9% ee (entry 9). This last catalyst was re-
vealed as the worst of the series (see the Supporting Infor-
mation). Conversely, the S configuration of diamines led to
the R enantiomer of salsolidine. As in the previous biotin–
streptavidin technology system, this behavior was strictly
related to the enantiodiscrimination generated by the pro-
tein at the racemic iridium center; therefore, the enantiose-
lective reduction of the considered substrate seems to rely
on the chirality at the metal center. The nearby bulky
phenyl residues appear to be important in modulating pro-
tein–catalyst interactions. However, they do not seem to be
relevant in determining the stereoselectivity by guiding the
approach of the substrate to the catalyst.
+
NH3 group of Lys121 (2.40 and 2.00 Š for the pro-R, and 2.63
and 2.01 Š for the pro-S configuration). Moreover, it was also
observed that the H···C=N distance in the pro-R configuration
was 1.14 Š shorter than that in the pro-S (Table S4, in the Sup-
porting Information).
Conversely, when considering the (SIr)-m-S112Y model as the
receptor, a lower docking energy was obtained for the pro-S
configuration. By inspecting the geometry of the two com-
plexes, a shorter H···C=N distance was found for the pro-S,
whereas in both complexes we observed the possibility for
Lys121 to act as an acidic catalysis trough a hydrogen-bond
+
between the NH3 group of Lys121 and the dihydroquinoline
nitrogen (1.82 and 1.87 Š for the pro-S and pro-R configura-
tions, respectively; Figure 2C and Figure S1B, in the Support-
ing Information).
Finally, compared with WT Sav, both S112C and S112Y muta-
tions concur in creating a hydrophobic pocket, in chain C,
which accommodates well the phenyl moiety on the six-mem-
bered ring (Figure 2). Thus, in both the meta or para substitu-
tion patterns, the ring is forced into a single conformation,
with an equatorial orientation of the phenyl group, and conse-
quently its flexibility is reduced.
With the aim of clarifying and confirming our assertions
about the behavior of the here-reported chiral 1,3 biotinylated
diamine ligands, the synthesis of achiral 1,3-ligands and tests
of the catalytic performances of their Ir(Cp) complexes were
performed.
From the experimental results and from the computational
study, different trends emerged:
The synthesis proceeded starting from 1,3-diaminopro-
pane,[18] as shown in Scheme 2, mirroring the synthesis of the
chiral ligands.
1) In all cases, incorporation of biotinylated metal complexes
within mutant Sav S112X leads to decreased conversion.
2) The optimal pH was 6.50 for [Cp*Ir(Biot-(S)-7)Cl]ꢀSav S112X
and pH 7.0 for [Cp*Ir(Biot-(R)-7)Cl]ꢀSav S112X; changes in
temperature were not as significant, even if the best results
were obtained at 308C and 408C, respectively.
3) Streptavidin is a homo-tetramer protein but only three
biotin binding sites were determined by using an assay
with biotin-4-fluorescein;[17] nevertheless, different [Cp*Ir-
(Biot-7)Cl]/Sav S112X ratios confirmed that 1:2 (versus
The resulting achiral o-, m-, and p-biotinylated diamines
were used for preparing the corresponding iridium(III) com-
plexes, then used in ATH of the salsolidine precursor under the
same reaction conditions as were used for the chiral ligands.
The results are reported in the Supporting Information (see
Table S2) and the optimized data are summarized in Table 3.
When the achiral system containing ligands 10 was used,
the product was mainly obtained in the R configuration, with
ChemCatChem 2016, 8, 1665 – 1670
www.chemcatchem.org
1668
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Scheme 2. Synthesis of biotinylated cofactors 10.
uct under ATH conditions when the chiral diamine moiety was
in the S configuration, although the best result was obtained
by using the homologue achiral ligand with a 83% ee. In
regard to the formation of S-salsolidine, the meta diversity of
the linker position at the stereocenter in the R configuration of
the ligand allowed us to obtain 47% ee with the Sav S112Y
mutant. Although the chiral d6 IrIII piano-stool biotinylated
complexes worked in the reduction of cyclic amine with very
low enantioselectivity, the presence of the biological scaffold
allowed us to introduce a new source of chirality with good ac-
tivity and enantioselectivity depending, as confirmed by com-
putational studies, on the enantiodiscrimination at the metal
center operated by the protein. In other words, although the
bare catalysts or their associated he complexes within Sav WT
revealed a lack of enantiodiscrimination, the combination of
chemical and genetic optimization showed a synergistic effect
in terms of enantioselectivity in both the achiral system and in
the chiral one.
Table 3. Optimized screening with S112X and [Cp*Ir(Biot-10)Cl].
Entry Protein Complex
pH
T
Conversion ee (R config)
[8C] [%]
[%]
1
2
3
S112C [Cp*Ir(Biot-p-10)Cl] 7.0 40 25
S112C [Cp*Ir(Biot-m-10)Cl] 7.0 40 15
S112C [Cp*Ir(Biot-o-10)Cl] 7.0 40 15
83
52
56
All reactions were carried out at 408C for 48 h by using 1 mol% com-
plex 10 with or without 0.33 mol% tetrameric WT or mutant Sav ([pro-
tein]f =0.28 mm) in 1.2m MOPS buffer, 3m HCOONa, pH 7.0, [sub]f =
10 mm, [cat]f =0.14 mm.
good ee, only in the presence of the S112C mutant. The in-
creased enantiodifferentiation ability shown by the achiral
ligand might be due to the presence of the chiral pocket close
to the reactive iridium catalytic site. In this case, in fact, the
lack of an additional bulky group weakly affects both the bind-
ing of the substrate with the catalytic complex and the interac-
tion of the biotinylated catalysts at the interface of chain A
and C (Figure 2A, B). Conversely, in the presence of mutant
S112Y, the phenyl ring at position 1 might increase the stabili-
zation of the complex by interacting with Y112 in chain C (Fig-
ure 2C, D), with a positive effect on both yield and ee, in this
case for the S enantiomer. Thus, the presence of a bulky chiral
center, as in diamines 7, especially with meta chemical diversi-
ty, offers the chance of modulating the stereoselectivity at the
metal, allowing the preparation of the S enantiomer also, al-
though with moderate enantioselectivity.
Acknowledgments
The authors would like to thank Prof. T. R. Ward and his group
for the gift of streptavidin mutants.
Keywords: asymmetric transfer hydrogenation
·
chiral
diamines · imino reductase · metalloenzymes · salsolidine
[1] a) M. T. Reetz, J. Am. Chem. Soc. 2013, 135, 12480–12496; b) C. M.
Thomas, T. R. Ward, Appl. Organomet. Chem. 2005, 19, 35–39; c) J. Pod-
tetenieff, A. Taglieber, E. Bill, E. J. Reijerse, M. T. Reetz, Angew. Chem. Int.
Ed. 2010, 49, 5151–5155; Angew. Chem. 2010, 122, 5277–5281; d) P. J.
Deuss, G. Popa, C. H. Botting, W. Laan, P. C. J. Kamer, Angew. Chem. Int.
Ed. 2010, 49, 5315–5317; Angew. Chem. 2010, 122, 5443–5445; e) T.
Heinisch, T. R. Ward, Curr. Opin. Chem. Biol. 2010, 14, 184–199; f) R.
Ricoux, M. Allard, R. Dubuc, C. Dupont, J.-D. Marechal, J.-P. Mahy, Org.
Biomol. Chem. 2009, 7, 3208–3211; g) A. Chevalley, M. V. Cherrier, J. C.
Fontecilla-Camps, M. Ghasemi, M. Salmain, Dalton Trans. 2014, 43,
5482–5489; h) F. Yu, V. M. Cangelosi, M. L. Zastrow, M. Tegoni, J. S. Ple-
Conclusions
The combination of chelating six-membered-ring transition-
metal complexes containing para biotinylated aminosulfona-
mide with Sav S112C in the formation of imino reductases was
able to reduce cyclic 1-methyl-6,7-dimethoxy-3,4-dihydroqui-
noline, a precursor of salsolidine, with 66% ee for the R prod-
ChemCatChem 2016, 8, 1665 – 1670
www.chemcatchem.org
1669
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
garia, A. G. Tebo, C. S. Mocny, L. Ruckthong, H. Qayyum, V. L. Pecoraro,
Chem. Rev. 2014, 114, 3495–3578; i) M. Hoarau, C. Hureau, E. Gras, P.
Faller, Coord. Chem. Rev. 2016, 308, 445–459; j) T. Quinto, F. Schwizer,
J. M. Zimbron, A. Morina, V. Kçhler, T. R. Ward, ChemCatChem 2014, 6,
1010–1014; k) M. Genz, V. Kçhler, M. Krauss, D. Singer, R. Hoffmann,
T. R. Ward, N. Sträter, ChemCatChem 2014, 6, 736–740; l) T. K. Hyster, L.
Knçrr, T. R. Ward, T. Rovis, Science 2012, 338, 500–503.
[9] G. Facchetti, R. Gandolfi, M. Fuse, D. Zerla, E. Cesarotti, M. Pellizzoni, I.
Rimoldi, New J. Chem. 2015, 39, 3792–3800.
[10] G. V. Sagar Reddy, G. V. Rao, R. V. K. Subramanyam, D. S. Iyengar, Synth.
Commun. 2000, 30, 2233–2237.
[11] M. Vaultier, N. Knouzi, R. CarriØ, Tetrahedron Lett. 1983, 24, 763–764.
[12] C. Letondor, A. Pordea, N. Humbert, A. Ivanova, S. Mazurek, M. Novic,
T. R. Ward, J. Am. Chem. Soc. 2006, 128, 8320–8328.
[2] a) T. R. Ward, Acc. Chem. Res. 2011, 44, 47–57; b) H. Yamaguchi, T.
Hirano, H. Kiminami, D. Taura, A. Harada, Org. Biomol. Chem. 2006, 4,
3571–3573.
[3] S. T. R. Walsh, H. Cheng, J. W. Bryson, H. Roder, W. F. DeGrado, Proc. Natl.
Acad. Sci. USA 1999, 96, 5486–5491.
[4] R. P. Megens, G. Roelfes, Chem. Eur. J. 2011, 17, 8514–8523.
[5] a) C. Jäckel, D. Hilvert, Curr. Opin. Biotechnol. 2010, 21, 753–759; b) M.
Creus, T. R. Ward, Org. Biomol. Chem. 2007, 5, 1835–1844.
[6] a) K. L. Morley, R. J. Kazlauskas, Trends Biotechnol. 2005, 23, 231–237;
b) M. T. Reetz, J. D. Carballeira, J. Peyralans, H. Hçbenreich, A. Maichele,
A. Vogel, Chem. Eur. J. 2006, 12, 6031–6038.
[13] J. D. Scott, R. M. Williams, Chem. Rev. 2002, 102, 1669–1730.
[14] a) F. Rosati, G. Roelfes, ChemCatChem 2010, 2, 916–927; b) P. J. Deuss,
R. den Heeten, W. Laan, P. C. J. Kamer, Chem. Eur. J. 2011, 17, 4680–
4698.
[15] Molecular Operating Environment (MOE), 2014.0901, Chemical Comput-
ing Group Inc., Montreal, 2014.
[16] V. M. Robles, M. Dürrenberger, T. Heinisch, A. Lledós, T. Schirmer, T. R.
Ward, J.-D. MarØchal, J. Am. Chem. Soc. 2014, 136, 15676–15683.
[17] a) R. Mittal, M. P. Bruchez, Bioconjugate Chem. 2011, 22, 362–368; b) G.
Kada, H. Falk, H. J. Gruber, Biochim. Biophys. Acta Gen. Subj. 1999, 1427,
33–43.
[7] J. M. Zimbron, T. Heinisch, M. Schmid, D. Hamels, E. S. Nogueira, T.
Schirmer, T. R. Ward, J. Am. Chem. Soc. 2013, 135, 5384–5388.
[8] a) M. Dürrenberger, T. Heinisch, Y. M. Wilson, T. Rossel, E. Nogueira, L.
Knçrr, A. Mutschler, K. Kersten, M. J. Zimbron, J. Pierron, T. Schirmer, T. R.
Ward, Angew. Chem. Int. Ed. 2011, 50, 3026–3029; Angew. Chem. 2011,
123, 3082–3085; b) V. MuÇoz Robles, P. Vidossich, A. Lledós, T. R. Ward,
J.-D. MarØchal, ACS Catal. 2014, 4, 833–842.
[18] C. E. Jakobsche, C. G. Parker, R. N. Tao, M. D. Kolesnikova, E. F. Douglass,
D. A. Spiegel, ACS Chem. Biol. 2013, 8, 2404–2411.
Received: January 29, 2016
Published online on April 5, 2016
ChemCatChem 2016, 8, 1665 – 1670
www.chemcatchem.org
1670
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim