Inorganic Chemistry
Article
The synthesis of the ligands started from monoBoc-
ethylenediamine that was reacted with ortho-, meta-, or para-
substituted chlorosulfonyl-benzoic acids 2 (Scheme 1).
In the case of meta- and para-compounds 2, the desired
products 3 were obtained with good yield by reaction in DCM
and using TEA as the base. Conversely, the ortho-3 compound
was never obtained, also changing the reaction conditions and
smaller difference between the R R and S R diastereomers
Ir Ir Ir Ir
in the case of [IrCp*(p-I)H] suggests the presence of more
than one diastereomer in the reaction medium. Indeed, the
higher variability could be at the origin of a possible poor
enantioselectivity of the para-system compared to the meta-
system, as also confirmed by the catalytic results reported
below. Focusing on the [IrCp*(m-I)Cl]Cl ⊂ Van complex
the base, and in any case, only bicyclic sulfamido derivative 5
instead, the S configuration causes more relevant clashes
Ir
1
(
identified by H NMR analysis) was isolated.
between the Cp* rings of the iridium complex and the Van
Ir
spectroscopic investigations were performed on the [IrCp*(m-
I)Cl]Cl ⊂ Van complex expected to be the most effective
artificial metalloenzyme from the data obtained by the
computational study.
3
1
for details). Briefly, the DADA dipeptide was assembled on
32
Wang resin using N-Fmoc solid-phase standard protocols;
then, compounds 3 were made to react with NH -DADA using
2
HOBt and HBTU as coupling reagents and DIPEA as the base.
After cleavage from the resin, m-l and p-l ligands were isolated
with 40% overall yield. Considering the demonstrated
capability of Van to coordinate directly with the iridium
metal center, thus producing an effectively functioning
1
2D-DOSY- H NMR experiments, performed at a final
concentration of 33 mM m-I in D
O (1.0% DMSO-d ),
2
6
indicated that Van alone in water formed aggregates (Table
42−45
1).
The increase in the calculated hydrodynamic radius
3
3
catalyst, the complexes were synthesized and used in a
preformed state to avoid the possible interference derived from
the direct Van/Ir interaction. The precatalysts were synthe-
sized by dissolving the ligand m-I or p-I in 2-propanol in the
presence of [IrCp*Cl ] and TEA. A yellow precipitate was
Table 1. Estimation of the Diffusion Coefficient and
Hydrodynamic Radius for Van Alone and for [IrCp*(m-
I)Cl]Cl ⊂ Van by 2D DOSY-NMR Experiments (https://
a
www.fxsolver.com/solve/)
2
2
obtained for both ligands. The formation of the complexes was
were recorded in water (1% DMSO) at a concentration of 250
μM. For both [IrCp*(p-I)Cl]Cl and [IrCp*(m-I)Cl]Cl, a
change in the metal-to-ligand charge transfer (MLCT) band at
almost 390 nm should underline the interaction between Van
and the amino acid moiety of the complex observed both using
diffusion coefficient hydrodynamic radius
10
148,859 × 10−9
Van
1642 × 10−
3725 × 10−
2442 × 10
3126 × 10−
10
10
10
−10
[IrCp*(m-I)Cl]Cl
826,251 × 10
143,534 × 10
206,418 × 10
−
−9
−9
Van/[IrCp*(m-I)Cl]Cl 1:1
Van/[IrCp*(m-I)Cl]Cl 2:1
a
DOSY-NMR: [sample] = 33 mM in D O (1.0% DMSO-d ), little
2
6
delta: 5.000 m, large delta: 149.900 m.
1
:1 and 2:1 ratios. This shift should be attributable to the
noncovalent interactions (hydrogen bonding, π−π stacking,
electrostatic, hydrophobic, and van der Waals interactions)
that occurred between the metal center and the second
coordination sphere generated by Van around the catalytic
active portion of the complex. No evidence of possible
precipitation due to the increase in the amount of Van was
(Table 1) of the [IrCp*(m-I)Cl]Cl complex in the presence of
1 equiv of Van unequivocally established the formation of the
supramolecular interaction between the complex and Van
34
highlighted in the spectra.
CD experiments were carried out in pure water and in
NaOAc buffer (0.1 M, pH 5). In both cases, the presence of a
positive band at 350 nm, together with a slightly positive
shoulder at approximately 415 nm, both ascribable to iridium
energetic transitions, confirmed the induction of chirality at the
metal complex due to the presence of the second coordination
sphere provided by the chiral Van moiety (Figure 2). As
suggested by computational studies and by comparing the CD
spectra of the [IrCp*(m-I)Cl]Cl complex with the spectrum of
the [IrCp*(m-I)Cl]Cl ⊂ Van artificial metalloenzyme, this
induced chirality is ascribable to the binding of one of the
preferential configurations to Van during the supramolecular
interaction transition state. This consideration was confirmed
by catalysis data: when the [IrCp*(m-I)Cl] complex was used
alone as the catalyst without the addition of Van. The reaction
entries 6−11). Looking at the aromatic ring band at 285 nm, a
different behavior in the CD spectra is observed depending on
A preliminary computational study was performed by
evaluating the two meta- and para-systems in terms of the
stability of the metal complex that was obtained. A higher
stability is thus expected to lead to the favored thermodynami-
cally stable form that has been revealed as the catalytically
performing system. Models were generated starting from the
aggregates of the X-ray crystal structure reported in the
1
1,35
literature.
Four vancomycin and two DADA units were
considered to retain the 2:1 ratio based on the extensive work
by Chen and Loll, in which the structure of the antibiotic−
target interaction was reported to always result in a dimeric
3
6,37
ratio.
IrCp*(p-I)H] and [IrCp*(m-I)H]. The structures obtained
were then optimized, employing the tight-binding GFN2-xTB
The original DADA was then modified to obtain
[
3
8
method including the implicit solvation model (GBSA),
which has been proven to perform well in systems of this size
at an affordable computational cost.
39−41
For both systems, the
four possible diastereomers at the metal centers were
generated with the aim of evaluating the different stabilities.
46
the environment. (Figure 2) For Van alone, this band is
negative, with a different intensity depending on the solvent,
thus confirming the Van aggregation propensity observed in
1
the R R configuration is significantly more stable in the case
2D-DOSY- H NMR experiments (ratio 2:1). In the [IrCp*(m-
Ir Ir
of the meta-substituted catalyst, [IrCp*(m-I)H], whereas a
I)Cl]Cl ⊂ Van artificial metalloenzyme, this band became less
2
978
Inorg. Chem. 2021, 60, 2976−2982