Prodrug Activation by Gold Artificial Metalloenzyme-Catalyzed Synthesis of Phenanthridinium Derivatives via Hydroamination

Article abstract of DOI:10.1002/anie.202100369

An emerging approach in the field of targeted drug delivery is the establishment of abiotic metal-triggered prodrug mechanisms that can control the release of bioactive drugs. Currently, the design of prodrugs that use abiotic metals as a trigger relies heavily on uncaging strategies. Here, we introduce a strategy based on the gold-catalyzed activation of a phenanthridinium-based prodrug via hydroamination under physiological conditions. To make the prodrug strategy biocompatible, a gold artificial metalloenzyme (ArM) based on human serum albumin, rather than the free gold metal complex, was used as a trigger for prodrug activation. The albumin-based gold ArM protected the catalytic activity of the bound gold metal even in the presence of up to 1 mM glutathione in vitro. The drug synthesized via the gold ArM exerted a therapeutic effect in cell-based assays, highlighting the potential usefulness of the gold ArM in anticancer applications.

Full text of DOI:10.1002/anie.202100369

Angewandte
Research Articles
Chemie
How to cite: Angew. Chem. Int. Ed. 2021, 60, 12446–12454
International Edition:
German Edition:
doi.org/10.1002/anie.202100369
doi.org/10.1002/ange.202100369
Drug Release
Prodrug Activation by Gold Artificial Metalloenzyme-Catalyzed
Synthesis of Phenanthridinium Derivatives via Hydroamination
Tsung-Che Chang, Kenward Vong, Tomoya Yamamoto, and Katsunori Tanaka*
Abstract: An emerging approach in the field of targeted drug
delivery is the establishment of abiotic metal-triggered prodrug
mechanisms that can control the release of bioactive drugs.
Currently, the design of prodrugs that use abiotic metals as
a trigger relies heavily on uncaging strategies. Here, we
introduce a strategy based on the gold-catalyzed activation of
a phenanthridinium-based prodrug via hydroamination under
physiological conditions. To make the prodrug strategy
biocompatible, a gold artificial metalloenzyme (ArM) based
on human serum albumin, rather than the free gold metal
complex, was used as a trigger for prodrug activation. The
albumin-based gold ArM protected the catalytic activity of the
bound gold metal even in the presence of up to 1 mM
glutathione in vitro. The drug synthesized via the gold ArM
exerted a therapeutic effect in cell-based assays, highlighting the
potential usefulness of the gold ArM in anticancer applications.
improve specificity. One strategy for improving drugs by
eliminating their undesirable properties is based on the
concept of prodrugs, which are inactive or less active
derivatives of drugs that undergo chemical transformation[3,4]
to regenerate the active forms (Figure 1a). In general,
a prodrug is composed of a drug and a masking group that
reduces the therapeutic effect of the drug. These prodrugs can
be triggered via specific chemical reactions involving various
components of the cancer microenvironment, such as thiols,[5]
reactive oxygen species,[6] and acidity,[7] which cleaves disul-
fide, boronate ester, and hydrazone bonds, respectively, in the
prodrug linker. Over the past few decades, click-to-release
chemistry[8] has emerged as a powerful approach for the study
of controlled drug release because of the use of small abiotic
molecules to mask drugs, thereby enhancing the specificity of
drug release. For instance, on the basis of the inverse electron
demand Diels–Alder (iEDDA) reaction, tetrazine moieties
have been used for drug release using dienophiles like trans-
cyclooctene[8] (Figure 1a).
Introduction
Chemotherapy is one of the four “pillars” of cancer
treatment, next to surgery, radiotherapy and immunother-
apy.[1] Often, chemotherapy is prescribed in combination with
surgery. However, because chemotherapy is known for being
notoriously unspecific, it has significant side effects on rapidly
dividing healthy cells in the digestive tract, hair follicles, bone
marrow, and lymphatic system.[2] Immunotherapy and radio-
therapy are alternatives to chemotherapy that offer decent
specificity but also suffers from other disadvantages and are
not always a viable option. To improve the specificity of drugs,
controlled drug-release is one of many possible ways to
Recently, bioorthogonal reactions that rely on abiotic
metals to catalyze new-to-nature reactions in living cells and
organisms have received more attention.[9] By incorporating
metals into nanoparticles or using homogeneous metal
complexes to overcome the inherent cytotoxicity of the
abiotic catalyst, several groups have demonstrated methods
for metal-based uncaging of drugs applicable in living
systems. Employing metals like iron, palladium, copper,
ruthenium, platinum and gold, several studies focused on
release of amine/hydroxyl groups through protecting groups
such as phenyl azide,[10] allyl,[11] propargyl,[12] allyloxycarbon-
yl,[13] dual-substituted propargyloxycarbonyl,[14] 2-alkynlbenz-
amide,[15] as well as a strategy based on ring-closing meta-
thesis.[16]
As mentioned previously, the design of a prodrug based
on the cancer microenvironment has the potential to achieve
site-specific activation. However, certain masking groups
used for the strategy are unstable and may lead to premature
drug release, resulting in off-target side effects before delivery
to the desired site.[17] Click-to-release chemistry could be used
to prevent premature drug release due to its relative stability
in living systems and controllable drug-release system via
inducible chemical reaction. However, the stoichiometric
character of click-to-release reactions might limit their
application for drug release at higher concentrations because
higher doses of trigger would be necessary.
[*] Dr. T.-C. Chang, Dr. K. Vong, Dr. T. Yamamoto, Prof. K. Tanaka
Biofunctional Synthetic Chemistry Laboratory
RIKEN Cluster for Pioneering Research, RIKEN
2-1 Hirosawa, Wako-shi, Saitama, 351-0198 (Japan)
E-mail: tanaka.k.dg@m.titech.ac.jp
Dr. K. Vong, Prof. K. Tanaka
GlycoTargeting Research Laboratory
RIKEN Baton Zone Program, RIKEN
2-1 Hirosawa, Wako-shi, Saitama, 351-0198 (Japan)
Prof. K. Tanaka
Department of Chemical Science and Engineering
School of Materials and Chemical Technology
Tokyo Institute of Technology
2-12-1 Ookayama, Meguro-ku, Tokyo, 152-8552 (Japan)
and
Biofunctional Chemical Laboratory
A. Butlerov Institute of Chemistry, Kazan Federal University
18 Kremlyovskaya Street, 420008 Kazan (Russia)
Abiotic metal-catalyzed bioorthogonal reactions can be
used to enhance specificity and efficiency of drug release due
to their bioorthogonal character and the outstanding catalytic
activity of transition metals. Recently, our group developed
the 2-alkynylbenzamide (ayba) moiety for gold-dependent
Supporting information and the ORCID identification number for
one of the authors of this article can be found under:
https://doi.org/10.1002/anie.202100369.
12446
ꢀ 2021 Wiley-VCH GmbH
Angew. Chem. Int. Ed. 2021, 60, 12446 –12454

Angewandte
Research Articles
Chemie
greatly beneficial to develop a suitable prodrug strategy for
such aromatic drugs.
Here, we report a new strategy for prodrug design using
2’-alkynyl-N-methyl-2-biphenylamine as framework for the
prodrug. Using this method, it is possible to synthesize 5-
methyl phenanthridinium derivatives via gold-catalyzed hy-
droamination (Figure 1b). The prodrug strategy not only can
generate 5-methyl phenanthridinium derivatives to mimic
naturally occurring alkaloids bearing a 5-methyl phenanthri-
dinium core (i.e. sanguinarine),[19] but can also synthesize
quaternary ammonium cation compounds. Nitrogen hetero-
aromatic cations are interesting due to their reactivity and
biological properties.[20] In particular, the phenanthridinium
moiety has attracted a great deal of attention because of its
presence in the scaffold of several DNA-intercalating agents
with antitumor properties.[21]
Results and Discussion
Development of a gold-mediated reaction for synthesis of
the 5-methyl phenanthridinium derivatives. The Song group
reported that 2’-alkynyl-biphenylamine could be converted to
a phenanthridine structure in the presence of gold ions.[22] To
confer a quaternary ammonium moiety on phenanthridine, 2’-
alkynyl-N-methyl-2-aminobiphenyl 1a was used to construct
a 5-methyl phenanthridinium moiety under gold catalysts via
hydroamination. As shown in Table 1, the cyclization of 1a to
produce the 5-methyl phenanthridinium derivative 2a was
first monitored under various conditions. Unexpectedly,
Table 1: Investigation of the cyclization of 1a in the presence of a gold
salt under various conditions.
Entry
Au-a
[mol%]
Solvent
Time
[min]
Temp. Yield of
[oC]
2a [%][a]
Figure 1. Strategy of prodrug design. a) General design of prodrugs
based on drug-release strategy mediated by the cancer microenviron-
ment, click chemistry, or transition metal-catalyzed reactions. b) This
work represents a new prodrug design based on the synthesis of 5-
methyl phenanthridinium derivatives that mimic naturally occurring
alkaloids via gold-triggered hydroamination of 2’-alkynyl-N-methyl-2-
biphenylamine derivatives.
1
2
3
4
5
6
7
8
10
25
50
75
100
10
10
10
10
10
10
10
10
10
10
–
10% dioxane/PBS
10% dioxane/PBS
10% dioxane/PBS
10% dioxane/PBS
10% dioxane/PBS
DMF
THF
CHCl3
dioxane
hexanes
15
15
15
15
15
15
15
15
15
15
15
15
60
120
15
120
37
37
37
37
37
37
37
37
37
37
37
37
37
37
25
37
79
74
64
62
56
N.R.[b]
14
37
4
release of secondary amines under physiological conditions.
The major benefit of the ayba protecting group is that the
decaging of drugs for cancer therapy can be triggered via gold
complexes that proceed orthogonally in the presence of
transition metals such as palladium and ruthenium.[15] Never-
theless, new strategies using abiotic metal as a trigger for
prodrug design are still in high demand because current
prodrug design relies heavily on the uncaging strategy
(Figure 1a). As a result, only drugs that have amine or
hydroxyl groups are suitable for this prodrug design. How-
ever, many drugs contain N-heterocyclic rings and polycyclic
aromatic hydrocarbon structures.[18] Hence, it would be
9
10
11
12
13
14
15
16
N.R.[b]
72
MeOH
10% dioxane/DMEM
10% dioxane/PBS
10% dioxane/PBS
10% dioxane/PBS
10% dioxane/PBS
67
89
92
35
N.R.[b]
[a] Yields determined by HPLC (peak retention times relative to product
standards, followed by MS analysis for confirmation, and calculation of
yields based on product standard curves). [b] N.R.: no reaction. All
reactions were standardized to 5 mmol of 1a in 0.5 mL of solvent
(10 mM).
Angew. Chem. Int. Ed. 2021, 60, 12446 –12454
ꢀ 2021 Wiley-VCH GmbH
www.angewandte.org 12447

Angewandte
Research Articles
Chemie
Table 2: Investigation of the catalytic activity of gold complexes with 2’-alkynyl-N-
methyl-2-aminobiphenyl derivatives.
increasing the level of the inorganic Au-a catalyst
under aqueous solution resulted in a reduced yield
of 2a (79% to 56%, entries 1–5). The use of
10 mol% Au-a achieved good yield within 15 min
(entry 1), whereas treatment of 1a with a larger
amount of Au-a (25–100 mol%) resulted in gener-
ation of a side product (see Supplementary Fig-
ure 22), decreasing the yield of 2a. Next, we
performed further investigations in which we
changed the solvent, time, and temperature. As
depicted in entries 6–11, reactions carried out in
organic solvent gave poor yields except in the case
of methanol (72%, entry 11). One potential explan-
ation for this observation is that water can facilitate
the ionization of the Au-Cl bond in Au-a,[11b] thus
smoothly triggering hydroamination in aqueous
solution. Longer reaction times achieved excellent
reaction yield (92%, entry 14), but lower temper-
ature caused lower yield (entry 15). Certainly, 1a
was stable in phosphate buffer (pH 7.4) in the
absence of gold catalyst (entry 16, see Supplemen-
tary Figure 21). Finally, to mimic biological condi-
tion, the reaction was also tested in DMEM culture
media (67%, entry 12). The results highlight the
potential biocompatibility of 5-methyl phenanthri-
dinium-based drug synthesis through gold-catalyzed
hydroamination.
Catalytic reactivity of gold complexes with 2’-
alkynyl-N-alkyl-2-biphenylamine derivatives. To
further probe the factors that contribute to the
cyclization reaction and investigate the catalytic
reactivity with various gold catalysts, substrates 1b–
h were prepared, and gold complexes (Au-b,c) were
applied. As summarized in Table 2, substrates 1a–h
were incubated with gold complex Au-b,c for 3 h at
378C, and the yield and turnover number (TON) of
their respective products 2a–h were recorded.
The electronic effects of N-substitution were
critical for facilitating the gold-mediated reaction.
The apparent evidence is substrate 1d bearing
a conjugate substituent group (R1 = 4-methylphen-
yl) in N-substitution, leading to weaken the nucleo-
philicity of the aniline moiety. Therefore, 1d gave
poor yields of the cyclization reaction whether we
used gold complex Au-b or Au-c (entry 4). In
contrast, substrate 1h bearing a 4-methoxyphenyl
Entry
1
Substrate
Au cat.
Product
Yield
[%][a]
TON[b]
Au-b
Au-c
99
88
396Æ9.2
351Æ7.0
2
Au-b
Au-c
99
97
396Æ10.2
388Æ3.2
3
4
Au-b
Au-c
99
63
396Æ6.1
254Æ2.2
Au-b
Au-c
7
42
29Æ0.5
166Æ3.7
5
6
7
Au-b
Au-c
87
50
343Æ1.4
199Æ4.2
Au-b
Au-c
50
99
202Æ3.1
396Æ9.5
Au-b
Au-c
69
41
275Æ5.8
164Æ3.7
8
Au-b
Au-c
23
55
91Æ0.8
218Æ1.2
[a] Yields determined by HPLC (peak retention times relative to product standards,
followed by MS analysis for confirmation and calculation of yields based on product
standard curves). [b] The TON was determined from product yields obtained by
HPLC. For all reactions, standardized concentrations of either Au-b or Au-c (2.5 mM,
0.25 mol%) and substrates 1a–h (1 mM) were used. Incubations were carried out
at 378C for 3 h, in triplicate. Error bars represent the s.d. of three independent
experiments.
group in the R1 position showed a higher conversion yield
compared to 1d due to enhancement of electronic effects in
N-substitution. Furthermore, N-methyl substitution 1a and
N-PEGylation 1c gave excellent yields using Au-b,c (entries 1
and 3) due to the effect of the electron-donating group.
Likewise, 1b bearing a methyl group in R2 position gave
a better result than 1a because the methyl group enhanced
the reactivity of N-methylaniline moiety (entry 2). Lower
overall yield of 1e relative to 1a was observed upon replace-
ment with amino acetic acid (entry 5) due to the inductive
effect of the carboxylic acid moiety. In the case of 1g, after the
alkyne group binds with gold complexes, the 1,3-dioxole
moiety might enhance electron density in the alkyne, thereby
decreasing the intramolecular hydroamination reactivity. On
the other hand, substrates 1 f and 1g, which bear an additional
benzo-annulation and 1,3-dioxole-annulation, respectively,
gave moderate yields using Au-b,c, highlighting the potential
for
synthesis
of
naturally
occurring
benzo[c]-
phenanthridinium alkaloids derivatives via this strategy.
Lastly, the steric effect of the ortho-substitution in the
biphenylic system also influence the cyclization yield via the
gold-mediated reaction (see Supplementary Table 3, sub-
strate S30).
In regard to the catalytic activity of gold complexes, on the
basis of yields and TONs in Table 2, Au-b (AuI complex)
could convert all substrates 1 into their respective products 2
12448 www.angewandte.org
ꢀ 2021 Wiley-VCH GmbH
Angew. Chem. Int. Ed. 2021, 60, 12446 –12454

Angewandte
Research Articles
Chemie
with 50–99% yields (except for substrates 1d,h, which were
Next, we carried out an experiment to determine the in
weakly reactive) and excellent TONs (over 200). By contrast,
Au-c [AuIII complex] only gave good yields and TONs for
substrates 1a,b,f. Likely, Au-b is more stable and more
reactive in phosphate buffer solution (pH 7.4) than Au-c
because the N-heterocyclic carbene ligand in Au-b has
a stronger metal-carbon s bond, which provides enhanced
stability and higher activities.[23] By contrast, Au-c complexes
possess high redox potential, leading to facile reduction of
AuIII complexes to metallic Au0 species in the presence of
electron-rich reagents.[24] Overall, the Au-b catalyst is a more
suitable trigger for the gold-mediated reaction for synthesis of
5-methyl phenanthridinium derivatives in aqueous solution.
Prodrug therapy using gold complex as a trigger. In the
next part of the study, we explored the applicability of the
drug-synthesis strategy mediated by gold catalysts to cancer
therapy. During the initial search for drug candidates,
reported compound 3 (Table 3) presented itself as a potential
drug candidate that could be synthesized by gold-mediated
hydroamination. According to the literature, 3 exerts cyto-
toxicity against various cancer cell lines.[25] Specifically, 3
causes cell-cycle arrest in cancer cells, increases the levels of
p53 protein, and apoptosis-specific fragmentation of PARP-1.
As depicted in Table 3, prodrug 4 could be converted to 5 by
a gold catalyst to mimic 3. Table 3 depicts our investigation of
the synthesis of drug 5 from prodrug 4 using a gold catalyst.
To completely convert prodrug 4 to drug 5, the amounts of
gold catalysts (Au-a, Au-b, and Au-c) were increased to
4 mol%. Initially, the use of inorganic Au-a and organome-
tallic complex Au-c for the reaction gave unsatisfying results.
Despite an increase in the reaction time, Au-a gave a low yield
(29%, entry 3), whereas Au-c showed poor reactivity (6–
12%, entries 7–9). By contrast, the Au-b [AuI complex]
exhibited excellent catalytic activity, achieving 5 at 84% yield
for 6 h (entry 6), further highlighting the high catalytic
reactivity of Au-b in this reaction.
cellulo activity of the prodrug against cancer cells (Figure 2).
To proceed with the cell-based experiment, we switched to
Au-1[15] because it was more soluble in DMSO than Au-b. In
an experiment, Au-1 could convert prodrug 4 to drug 5 within
6 h in DMEM, whereas any significant side product was not
observed (see Supplementary Figure 33). Later, the coumarin
moiety of Au-1 will be used to anchor the catalyst into human
serum albumin to prepare a gold artificial metalloenzyme.
Using a set concentration of prodrug 4 (8 mM for A549 cells),
we performed cytotoxicity studies by treating A549 cells with
Au-1 to convert 4 to the active drug 5. In control experiments,
cells were treated with DMSO, prodrug 4 (8 mM), or drug 5
(8 mM). The results revealed that a mixture of prodrug 4 with
an Au-1 concentration of 2.5 mM induced significant decrease
in cell growth relative to the individual effects of each
compound at the corresponding concentrations. The data
clearly showed that a prodrug strategy based on direct
synthesis of a phenanthridinium derivative by gold-catalyzed
hydroamination under mild conditions could be applied to
cancer therapy.
Table 3: Investigation of the cyclization of 4 in the presence of gold
complexes.
Figure 2. Cytotoxicity studies for prodrug activation via gold catalysis.
A viability assay was carried out using A549 cancer cells. The cells
were treated with prodrug 4 (8 mM), followed by immediate addition of
Au-1 at the indicated concentrations. In parallel experiments, the cells
were only treated with the indicated concentrations of Au-1. In control
experiments, cells were treated with DMSO, prodrug 4 (8 mM), or drug
5 (8 mM). To completely dissolve 4, 5, or Au-1 in media, the media
used in the viability assay contained 1% DMSO (v/v). Error bars
indicate s.d. of three replicate measurements.
Entry
Au cat.
Time [h]
Yield of 5 [%][a]
1
2
3
4
5
6
7
8
9
10
2
4
6
2
4
6
2
4
6
6
9
Au-a
11
29
42
56
84
6
Development of an albumin-based gold ArM as a trigger
for the prodrug strategy. Despite these promising results,
practical application of the gold-mediated prodrug strategy in
cancer treatment has been limited by a major obstacle,
namely, biocompatibility. Glutathione (GSH), a significant
metabolite in biological systems, is often present at physio-
logical concentrations in the range of 0.5–10 mM (intra-
cellular)[26] or 2–20 mM (extracellular).[27] Upon administering
gold(I) complexes to GSH, a ligand-exchange reaction with
GSH can occur rapidly,[28] abrogating the activity of gold
complexes.[29] Therefore, the next challenge is to make the
Au-b
Au-c
7
12
–
N.R.[b]
[a] Yields determined by HPLC (peak retention times relative to product
standards, followed by MS analysis for confirmation, and calculation of
yields based on product standard curves). All reactions were stand-
ardized to 1 mmol of 4 in 1.0 mL of solvent (1 mM). [b] N.R.: no reaction.
Angew. Chem. Int. Ed. 2021, 60, 12446 –12454
ꢀ 2021 Wiley-VCH GmbH
www.angewandte.org 12449

Angewandte
Research Articles
Chemie
prodrug strategy biocompatible with GSH, facilitating future
practical applications in cancer therapy.
deactivation of the gold complex by GSH (Figure 3a).
Furthermore, albumin contains more than 30 lysines that
can be conjugated to cancer-targeting ligands (e.g., anti-
bodies,[35] small peptides,[36] and carbohydrates[37]), thereby
minimizing unnecessary side effects on healthy cells.
Artificial metalloenzymes (ArMs) are created by incor-
porating synthetic metal complexes into protein scaffolds,
thereby combining the advantageous features of organome-
tallic and enzymatic catalysts and facilitating the design of
novel biocatalysts to perform new-to-nature reactions.[30]
Recently, our group used human serum albumin (HSA) as
a protein scaffold to design and develop ruthenium-based
ArMs as a potential anticancer prodrug therapy via ring-
closing metastasis.[31] In addition, we developed an ethylene
probe through cross metathesis with ethylene gas.[32] These
studies revealed two important biological effects of anchoring
a metal catalyst to albumin. First, an insoluble ruthenium
complex catalyst linked to a coumarin-ligand anchor can be
well dissolved in aqueous solution by addition of albumin.
This effect results from a binding interaction between the
coumarin moiety of the ruthenium catalyst and the hydro-
phobic binding pocket of albumin, which is well known to
bind many hydrophobic drug molecules.[33] In addition, due to
the negatively charged surface of albumin at physiological
pH,[34] albumin can prevent glutathione (GSH) from entering
the hydrophobic binding pocket to destroy the ruthenium
catalyst. On the basis of these results, development of an
albumin-based gold ArM as a trigger for a gold-mediated
drug synthesis strategy could be an ideal method for avoiding
Characterization of the albumin-based gold ArM. To
prepare and characterize the albumin-based gold ArM, we
followed a process that our group used previously to construct
ruthenium-based ArMs.[31] As depicted in Figure 3b, Au-
1 was anchored through the interaction of coumarin-based
derivatives (e.g., warfarin) to the drug site I binding pocket of
albumin[38] to generate albumin-Au-1. To confirm the for-
mation of the albumin-Au complex, we used a fluorometric
assay. The fluorescence emitted by coumarin-based molecules
is quite sensitive to the polarity of the surrounding environ-
ment. As shown in Supplementary Figure 15, the quantum
yield of Au-1 increased around 10-fold upon transition from
a polar environment (10% dioxane/H2O) to a non-polar
environment (60% dioxane/H2O). Therefore, formation of
the albumin-Au complex can be easily confirmed by measur-
ing the fluorescence emission of Au-1 (lEX = 425 nm/lEX
=
465 nm) upon binding to the hydrophobic area (pocket) of
albumin. The fluorescence intensities measured from 1:1
ligand-to-albumin incubations using catalyst Au-1 were sig-
nificantly higher than those of the corresponding individual
controls (Au-1 only and albumin only), indicating that the
Figure 3. Preparation and characterization of the albumin-based gold ArM generated in this study. a) Development of an albumin-based gold
ArM for drug synthesis via hydroamination to avoid deactivation by glutathione (GSH). b) The albumin-Au complex (albumin-Au-1) was prepared
by incubation of the gold catalyst Au-1 with human serum albumin. c) Crystal structure of HSA (PDB1AO6) with emphasis on two major binding
regions, drug site I and drug site II. d) To determine the binding sites, saturation binding analysis was carried out with HSA (20 mM) prebound
with known site markers warfarin (40 mM) and ibuprofen (40 mM). e) Binding between HSA (40 mM) and compound Au-1 (40 mM) monitored over
time. The horizontal dashed line represents the intrinsic fluorescence emitted by unbound HSA. Protein binding in (d) and (e) was determined by
monitoring the change in natural fluorescence emitted by albumin (mEX =280 nm/mEX =320 nm). Error bars indicate s.d. of three replicate
measurements.
12450 www.angewandte.org
ꢀ 2021 Wiley-VCH GmbH
Angew. Chem. Int. Ed. 2021, 60, 12446 –12454

Angewandte
Research Articles
Chemie
albumin-Au complex had formed (see Supplementary Fig-
ure 13).
Human serum albumin has two major hydrophobic bind-
ing regions, drug site I in subdomain IIA and drug site II in
subdomain IIIA (Figure 3c). To further confirm that Au-
1 binds to drug site I, we performed a saturation binding assay
with warfarin and ibuprofen, which both bind with low
micromolar KD to drug sites I and II, respectively. In the assay,
binding can be measured through the fluorescence quenching
of albumin (due to aromatic amino acids like tryptophan)
Au-1 (see Supplementary Table 4). The other significant
observation was the correlation between the hydrophobicity
of these substrates and their observed reactivities. In our
previous study using albumin-based ArM,[31] charged metabo-
lites were prevented from entering the hydrophobic binding
pocket due to the negatively charged surface of albumin at
physiological pH. On the basis of this result, substrate 1e gave
a lower TON (63) because it includes a carboxylate group,
making it difficult to enter the albumin binding pocket. By
contrast, substrates 1a–c,f do not include any physiologically
charged functional group to keep their hydrophobic charac-
ters so that they gave high TONs (over 100). In particular,
substrate 1c was converted with a TON of over 200. Although
substrate 1g does not have a charged functional group, it gave
a lower TON (71) since it had lower reactivity than 1a even
using free-in-solution Au-b (see data in Table 2). Importantly,
hydroamination is not catalyzed by any known naturally
occurring enzymes, highlighting the significance of gold ArM-
catalyzed hydroamination under physiological conditions.
Exploring the reactivity of the albumin-based gold ArM in
the presence of various bioadditives. Given that albumin-Au-
1 was active in catalyzing hydroamination, we next focused on
proving biocompatibility by the one-by-one addition of
several kinds of bioadditives to the gold-mediated reaction.
For this investigation, a known literature protocol was
followed,[39] where two kinds of amino acids (lysine and
arginine), sugars (glucose and sucrose), glutathione, and cell
lysate were used. Substrate 1c was chosen as the model
compound in this investigation. As depicted in Figure 4, using
0.25 mol% of free-in-solution Au-b or albumin-Au-1, 1c was
incubated with various bioadditives (1200 mM, 480 equiv to
Au-b or albumin-Au-1) for 3 h at 378C. Significant changes to
these measured TONs were not observed in the presence of
amino acids, sugars, and 0.5% cell lysates. The TON of 2c
using albumin-Au-1 was still > 100 when GSH was used. By
contrast, the reactivity of Au-b was completely quenched by
the same concentration of GSH. In the presence of 0.5% cell
lysate in the solvent, an approximate 50% decrease in the
reactivity of Au-b was observed. Overall, this suggested that
upon binding of ligand into drug site I (lEX = 280 nm/lEX
=
320 nm). As depicted in Figure 3d, binding of Au-1 was
unaffected in the presence of ibuprofen, but decreased
significantly in the presence of warfarin. This evidence
strongly indicated that drug site I is the main binding pocket
of Au-1 on HSA. To optimize the incubation time, we
monitored binding between HSA and Au-1 over several time
points (Figure 3e). Based on the fluorescence quenching of
albumin, an incubation time of 1 h was deemed sufficient to
ensure maximal binding.
Reactivity of the albumin-based gold ArM. Next, we
explored the ability of the albumin-Au complex to catalyze
hydroamination for substrates 1a–h and prodrug 4. Even
though a small amount of albumin-Au-1 (0.25 mol%) was
used, it still gave good conversion yields (31–53%) and
excellent TONs (over 140) with substrates 1a–c,f and prodrug
4 (Table 4). These data prove that albumin-Au-1 can perform
hydroamination to synthesize phenanthridinium derivatives
and could be used as a trigger for activation of prodrug 4. In
an additional investigation, we achieved the complete con-
version of 4 to drug 5 within 6 h using 20 mol% of albumin-
Table 4: Investigation of the catalytic activity of albumin-Au-1 with 2’-
alkynyl-N-methyl-2-aminobiphenyl derivatives.
Entry
Substrate
Product
Yield [%][a]
TON[b]
1
2
3
4
5
6
7
8
9
1a
1b
1c
1d
1e
1 f
1g
1h
4
2a
2b
2c
2d
2e
2 f
2g
2h
5
41
46
53
3
16
36
18
11
31
162Æ4.1
186Æ4.1
211Æ11.5
13Æ0.5
63Æ1.2
142Æ5.0
71Æ3.3
45Æ4.1
123Æ2.4
[a] Yields determined by HPLC (peak retention times relative to product
standards, followed by MS analysis for confirmation and calculation of
yields based on product standard curves). [b] TON was determined from
product yields obtained by HPLC. For all reactions, standardized
concentrations of albumin-Au-1 (2.5 mM, 0.25 mol%) and all substrates
(1 mM) were used. Incubations were carried out at 378C for 3 h, in
triplicate. Error represents s.d. of three independent experiments.
Figure 4. Investigating the reactivity of Au-b and albumin-Au-1 in the
presence of various bioadditives. We investigated the effect of each
bioadditive at a concentration 1200 mM or 0.5% (v/v) cell lysate in
total solvent on the reactivity of Au-b and albumin-Au-1 with substrate
1c (1 mM). Product formation was assessed by HPLC analysis and
subsequently converted to TON. Error bars represent s.d. of three
replicate measurements.
Angew. Chem. Int. Ed. 2021, 60, 12446 –12454
ꢀ 2021 Wiley-VCH GmbH
www.angewandte.org 12451

Angewandte
Research Articles
Chemie
albumin could protect the catalytic activity of the bound gold
metal from glutathione or a certain extent of cell lysate. Given
the promising results in vitro, albumin-Au-1 should be a more
suitable trigger than Au-1 for a prodrug strategy in the
context of cancer therapy.
Prodrug therapy using the gold ArM as a trigger. In the
final stage of the study, we investigated the feasibility of using
albumin-Au-1 as a trigger to activate prodrug 4 against cancer
cells. As shown in Figure 5, the mixture of albumin-Au-1 and
prodrug 4 were found to reduce A549 cell growth in
phenanthridinium core structures can be synthesized by
a readily available gold (I)-chloride complex and an albu-
min-based gold ArM via highly efficient hydroamination
under physiological conditions. The new reaction, which
involves a carbon-nitrogen bond cyclization in aqueous
solution using an abiotic metal catalyst (gold), is bioorthog-
onal. Finally, the reaction was used for prodrug design, and
ultimately applied to cancer therapy. Our promising results
imply that the gold-catalyzed cyclization reaction has great
potential for design of prodrugs based on phenanthridinium
structure, such as naturally occurring benzo[c]phenanthridine
alkaloids. Meanwhile, our prodrug strategy based on direct
drug synthesis also represents a significant addition to the
toolbox of prodrug strategy, which relies heavily on uncaging
strategies.
Notable cases of ArMs obtained from non-metalloprotein
sources have integrated metals such as copper,[40] rhodium,[41]
iridium,[42] and ruthenium[43] to achieve new-to-nature reac-
À
tions including Diels–Alder reaction, C H activation, trans-
fer hydrogenation, and olefin metathesis. Very recently,
a study by Vornholt et al reported a reaction-independent,
automation-compatible platform to effectively discover ac-
tive ArM variants for new-to-nature reactions. They system-
atically assess 400 ArM mutants for five bioorthogonal
transformations involving different metal cofactors, which
include two first gold-based ArMs for hydroamination and
hydroarylation.[44] However, current examples highlighting
the therapeutic potential of ArMs have been limited, with
very few distinguished exceptions.[31,45] According to the
results described above, this study reports a novel gold-based
ArM for hydroamination that is capable of synthesizing
phenanthridinium derivatives in high TON (over 200). In
addition, the gold ArM did not significantly decrease hydro-
amination reactivity even in the presence of 1 mM gluta-
thione. This study also demonstrates that a gold-based ArM
can be used as a trigger for the synthesis of phenanthridinium-
based drug via hydroamination, and that the resultant agent
can be applied to cancer therapy, further highlighting the
potential of albumin-based gold ArM to be used therapeuti-
cally.
Figure 5. Cytotoxicity studies for prodrug activation by albumin-Au-
1 catalysis. A viability assay was carried out using A549 cancer cells.
The cells were treated with prodrug 4 (8 mM), followed by the
immediate addition of albumin-Au-1 at the indicated concentrations.
In parallel experiments, the cells were only treated with the indicated
concentrations of albumin-Au-1. In control experiments, cells were
treated with DMSO, prodrug 4 (8 mM), or drug 5 (8 mM). To
completely dissolve 4 or 5, the media used in the viability assay
contained 1% DMSO (v/v). Error bars, s.d. of three replicate measure-
ments.
a concentration-dependent manner. The other significant
observation was that albumin-Au-1 was non-toxic at 10 mM,
whereas the gold complex Au1 exerted obvious cytotoxicity at
this concentration (Figure 2). On the basis of the data, the
usage of albumin-Au-1 is less cytotoxic than Au-1, implying
that this agent could avoid unwanted side effects. These
observations highlight the potential importance of albumin-
based gold ArMs as triggers for prodrug strategies.
Acknowledgements
A comparison of cytotoxicity data between Figure 2 and
Figure 5 indicated that albumin-Au-1 could synthesize active
drug 4 more efficiently than Au-1 in a cell-based experiment;
using 0.63 mM albumin-Au-1 decreased cell growth by 50%,
whereas 2.5 mM Au-1 was required to achieve the same level
of growth inhibition. These results show that albumin could
protect the catalytic reactivity of Au-1 from cell metabolites.
This work was financially supported by the AMED Grant
JP15KM0908001, a research grant from the Astellas Founda-
tion, and JSPS KAKENHI Grant Numbers JP18K14347 and
JP19K15708. This work was also funded by a subsidy
allocated to Kazan Federal University for state assignment
in the sphere of scientific activities (0671–2020-0063), with the
support of the Russian Government Program of Competitive
Growth of Kazan Federal University.
Conclusion
Conflict of interest
Taken together, the work conducted in this study will
significantly advance two research fields: metal-mediated
prodrug strategy for cancer therapy and research focused on
development of therapeutic ArMs. Our results show that
The authors declare no conflict of interest.
12452 www.angewandte.org
ꢀ 2021 Wiley-VCH GmbH
Angew. Chem. Int. Ed. 2021, 60, 12446 –12454

Angewandte
Research Articles
Chemie
16783 –16790; h) M. Sancho-Albero, B. Rubio-Ruiz, A. M.
Keywords: antitumor agents ·gold catalysts ·metalloenzymes ·
PØrez-López, V. Sebastiµn, P. Martín-Duque, M. Arruebo, J.
Santamaría, A. Unciti-Broceta, Nat. Catal. 2019, 2, 864 –872; i) P.
Destito, A. Sousa-Castillo, J. R. Couceiro, F. López, M. A.
Correa-Duarte, J. L. MascareÇas, Chem. Sci. 2019, 10, 2598 –
2603; j) C. Torres-Sµnchez, A. M. PØrez-López, M. N. Alqahtani,
A. Unciti-Broceta, B. Rubio-Ruiz, New J. Chem. 2019, 43, 1449 –
1458; k) B. L. Oliveira, B. J. Stenton, V. B. Unnikrishnan,
C. R. D. Almeida, J. Conde, M. Negr¼o, F. S. S. Schneider, C.
Cordeiro, M. G. Ferreira, G. F. Caramori, J. B. Domingos, R.
Fior, G. J. L. Bernardes, J. Am. Chem. Soc. 2020, 142, 10869 –
10880.
phenanthridinium ·prodrugs
[1] D. D. Luo, A. K. Carter, D. Miranda, J. F. Lovell, Adv. Sci. 2017,
4, 1600106.
[2] E. Petru, D. Schmahl, Neoplasma 1991, 38, 147 –155.
[3] V. Abet, F. Filace, J. Recio, J. Alvarez-Builla, C. Burgos, Eur. J.
Med. Chem. 2017, 127, 810 –827.
[4] J. Delahouse, C. Skarbek, A. Paci, Cancer Chemother. Pharma-
col. 2019, 84, 937 –958.
[5] B. Sun, C. Luo, H. Yu, X. Zhang, Q. Chen, W. Yang, M. Wang, Q.
Kan, H. Zhang, Y. Wang, Z. He, J. Sun, Nano Lett. 2018, 18,
3643 –3650.
[6] M. Ye, Y. Han, J. Tang, Y. Piao, X. Liu, Z. Zhou, J. Gao, J. Rao, Y.
Shen, Adv. Mater. 2017, 29, 1702342.
[7] S. J. Sonawane, R. S. Kalhapure, T. Govender, Eur. J. Pharm. Sci.
2017, 99, 45 –65.
[8] X. Ji, Z. Pan, B. Yu, L. K. De La Cruz, Y. Zheng, B. Ke, B. Wang,
Chem. Soc. Rev. 2019, 48, 1077 –1094.
[9] a) J. J. Soldevila-Barreda, P. J. Sadler, Curr. Opin. Chem. Biol.
2015, 25, 172 –183; b) Y. Bai, J. Chen, S. C. Zimmerman, Chem.
Soc. Rev. 2018, 47, 1811 –1821; c) M. Martínez-Calvo, J. L.
MascareÇas, Coord. Chem. Rev. 2018, 359, 57 –79; d) J. G.
Rebelein, T. R. Ward, Curr. Opin. Biotechnol. 2018, 53, 106 –
114; e) J. Chen, J. Wang, Y. Bai, K. Li, E. S. Garcia, A. L.
Ferguson, S. C. Zimmerman, J. Am. Chem. Soc. 2018, 140,
13695 –13702; f) J. Chen, J. Wang, K. Li, Y. Wang, M. Gruebele,
A. L. Ferguson, S. C. Zimmerman, J. Am. Chem. Soc. 2019, 141,
9693 –9700; g) S. R. Thomas, A. Casini, Curr. Opin. Chem. Biol.
2020, 55, 103 –110; h) P. Destito, C. Vidal, F. López, J. L.
MascareÇas, Chem. Eur. J., DOI: 10.1002/chem.202003927;
i) D. C. Luther, R. Huang, T. Jeon, X. Zhang, Y. W. Lee, H.
Nagaraj, V. M. Rotello, Adv. Drug. Delivery Rev. 2020, 156, 188 –
213; j) Y. Liu, Y. Bai, ACS Appl. Bio Mater. 2020, 3, 4717 –4746;
k) M. O. N. van de L’lsle, M. C. Ortega-Liebana, A. Unciti-
Broceta, Curr. Opin. Chem. Biol. 2021, 61, 32 –42.
[10] a) P. K. Sasmal, S. Carregal-Romero, A. A. Han, C. N. Streu, Z.
Lin, K. Namikawa, S. L. Elliott, R. W. Kçster, W. J. Parak, E.
Meggers, ChemBioChem 2012, 13, 1116 –1120; b) K. K. Sadhu,
T. Eierhoff, W. Rçmer, N. Winssinger, J. Am. Chem. Soc. 2012,
134, 20013 –20016; c) R. Huang, C. H. Li, R. Cao-Milan, L. D.
He, J. M. Makabenta, X. Zhang, E. Yu, V. M. Rotello, J. Am.
Chem. Soc. 2020, 142, 10723 –10729; d) J. Chen, K. Li, J. S. L.
Shon, S. C. Zimmerman, J. Am. Chem. Soc. 2020, 142, 4565 –
4569.
[13] a) R. M. Yusop, A. Unciti-Broceta, E. M. V. Johansson, R. M.
Sµnchez-Martín, M. Bradley, Nat. Chem. 2011, 3, 239 –243; b) J.
Li, J. Yu, J. Zhao, J. Wang, S. Zheng, S. Lin, L. Chen, M. Yang, S.
Jia, X. Zhang, P. R. Chen, Nat. Chem. 2014, 6, 352 –361; c) T.
Vçlker, F. Dempwolff, P. L. Graumann, E. Meggers, Angew.
Chem. Int. Ed. 2014, 53, 10536 –10540; d) M. I. Sµnchez, C.
Penas, M. E. Vµzquez, J. L. MascareÇas, Chem. Sci. 2014, 5,
1901 –1907; e) M. A. Miller, B. Askevold, H. Mikula, R. H.
Kohler, D. Pirovich, R. Weissleder, Nat. Commun. 2017, 8,
15906; f) M. Martínez-Calvo, J. R. Couceiro, P. Destito, J.
Rodríguez, J. Mosquera, J. L. MascareÇas, ACS Catal. 2018, 8,
6055 –6061.
[14] X. Wang, Y. Liu, X. Fan, J. Wang, W. S. C. Ngai, H. Zhang, J. Li,
G. Zhang, J. Lin, P. R. Chen, J. Am. Chem. Soc. 2019, 141, 17133 –
17141.
[15] K. Vong, T. Yamamoto, T.-C. Chang, K. Tanaka, Chem. Sci. 2020,
11, 10928 –10933.
[16] V. Sabatino, J. G. Rebelein, T. R. Ward, J. Am. Chem. Soc. 2019,
141, 17048 –17052.
[17] K. M. Huttunen, H. Raunio, J. Rautio, Pharmacol. Rev. 2011, 63,
750 –771.
[18] Z. Hosseinzadeh, A. Ramazani, N. Razzaghi-Asl, Curr. Org.
Chem. 2018, 22, 2256 –2279.
[19] V. Bisai, M. K. S. Shaheeda, A. Gupta, A. Bisai, Asian J. Org.
Chem. 2019, 8, 946 –969.
[20] D. Utreja, S. Sharma, A. Goyal, K. Kaur, S. Kaushal, Curr. Org.
Chem. 2019, 23, 2271 –2294.
´
[21] L.-M. Tumir, M. R. Stojkovic, I. Piantanida, Beilstein J. Org.
Chem. 2014, 10, 2930 –2954.
[22] J.-B. Wang, Q.-Q. Wu, Y.-Z. Min, Y.-Z. Liu, Q.-H. Song, Chem.
Commun. 2012, 48, 744 –746.
[23] X.-T. Tang, F. Yang, T.-T. Zhang, Y.-F. Liu, S.-Y. Liu, T.-F. Su, D.-
C. Lv, W.-B. Shen, Catalysts 2020, 10, 350.
[24] L. Rocchigiani, M. Bochmann, Chem. Rev. 2020, ASAP article.
´
[25] P. Lasµk, K. Motyka, V. Krystof, J. Styskala, Molecules 2018, 23,
[11] a) M. Tomµs-Gamasa, M. Martínez-Calvo, J. R. Couceiro, J. L.
MascareÇas, Nat. Commun. 2016, 7, 12538; b) C. Vidal, M.
Tomµs-Gamasa, P. Destitio, F. López, J. L. MascareÇas, Nat.
Commun. 2018, 9, 1913.
2155.
[26] C. R. Corso, A. Acco, Crit. Rev. Oncol. Hematol. 2018, 128, 43 –
57.
[27] G. Wu, Y.-Z. Fang, S. Yang, J. R. Lupton, N. D. Turner, J. Nutr.
2014, 134, 489 –492.
[28] C. F. Shaw III, Chem. Rev. 1999, 99, 2589 –2600.
[29] S. J. Berners-Price, A. Filipovska, Metallomics 2011, 3, 863 –873.
[30] M. Jeschek, S. Panke, T. R. Ward, Trends Biotechnol. 2018, 36,
60 –72.
[31] S. Eda, I. Nasibullin, K. Vong, N. Kudo, M. Yoshida, A.
Kurbangalieva, K. Tanaka, Nat. Catal. 2019, 2, 780 –792.
[32] K. Vong, S. Eda, Y. Kadota, I. Nasibullin, T. Wakatake, S.
Yokoshima, K. Shirasu, K. Tanaka, Nat. Commun. 2019, 10,
5746.
[12] a) J. T. Weiss, J. C. Dawson, K. G. Macleod, W. Rybski, C. Fraser,
C. Torres-Sµnchez, E. E. Patton, M. Bradley, N. O. Carragher, A.
Unciti-Broceta, Nat. Commun. 2014, 5, 3277; b) G. Y. Tonga, Y.
Jeong, B. Duncan, T. Mizuhara, R. Mout, R. Das, S. T. Kim, Y. C.
Yeh, B. Yan, S. Hou, V. M. Rotello, Nat. Chem. 2015, 7, 597 –603;
c) B. Rubio-Ruiz, J. T. Weiss, A. Unciti-Broceta, J. Med. Chem.
2016, 59, 9974 –9980; d) A. M. PØrez-López, B. Rubio-Ruiz, V.
Sebastian, L. Hamilton, C. Adam, T. L. Bray, S. Irusta, P. M.
Brennan, G. C. Lloyd-Jones, D. Sieger, J. Santamaria, A. Unciti-
Broceta, Angew. Chem. Int. Ed. 2017, 56, 12548 –12552; e) J.
Clavadetscher, E. Indrigo, S. V. Chankeshwara, A. Lilienkampf,
M. Bradley, Angew. Chem. Int. Ed. 2017, 56, 6864 –6868; f) T. L.
Bray, M. Salji, A. Brombin, A. M. PØrez-López, B. Rubio-Ruiz,
L. C. A. Galbraith, E. E. Patton, H. Y. Leung, A. Unciti-
Broceta, Chem. Sci. 2018, 9, 7354 –7361; g) C. Adam, A. M.
PØrez-López, L. Hamilton, B. Rubio-Ruiz, T. L. Bray, D. Sieger,
P. M. Brennan, A. Unciti-Broceta, Chem. Eur. J. 2018, 24,
[33] F. F. An, X. H. Zhang, Theranostics 2017, 7, 3667 –3689.
[34] D. Malamud, J. W. Drysdale, Anal. Biochem. 1978, 86, 620 –647.
[35] T. M. Allen, Nat. Rev. Cancer 2002, 10, 750 –763.
[36] D. E. Large, J. R. Soucy, J. Hebert, D. T. Auguste, Adv. Ther.
(Weinheim, Ger.) 2019, 2, 1800091.
[37] K. Vong, T. Yamamoto, K. Tanaka, Small 2020, 16, 1906890.
Angew. Chem. Int. Ed. 2021, 60, 12446 –12454
ꢀ 2021 Wiley-VCH GmbH
www.angewandte.org 12453

Angewandte
Research Articles
Chemie
[38] J. Ghuman, P. A. Zunszain, I. Petitpas, A. A. Bhattacharya, M.
Otagiri, S. Curry, J. Mol. Biol. 2005, 353, 38 –52.
Rebelein, H. Mallin, C. Trindler, M. M. Pellizzoni, T. R. Ward, J.
Am. Chem. Soc. 2018, 140, 13171 –13175.
[43] a) C. Mayer, D. G. Gillingham, T. R. Ward, D. Hilvert, Chem.
Commun. 2011, 47, 12068 –12070; b) T. Matsuo, C. Imai, T.
Yoshida, T. Saito, T. Hayashi, S. Hirota, Chem. Commun. 2012,
48, 1662 –1664; c) M. Jeschek, R. Reuter, T. Heinisch, C.
Trindler, J. Klehr, S. Panke, T. R. Ward, Nature 2016, 537, 661 –
665; d) M. Szponarski, F. Schwizer, T. R. Ward, K. Gademann,
Commun. Chem. 2018, 1, 84; e) T. Heinisch, F. Schwizer, B.
Garabedian, E. Csibra, M. Jeschek, J. Vallapurackal, V. B.
Pinheiro, P. Marli›re, S. Panke, T. R. Ward, Chem. Sci. 2018, 9,
5383 –5388.
[44] T. Vornholt, F. Christoffel, M. M. Pellizzoni, S. Panke, T. R.
Ward, M. Jeschek, Sci. Adv. 2021, 7, eabe4208.
[45] M. Basauri-Molina, D. G. A. Verhoeven, A. J. V. Schaik, H.
Kleijn, R. J. M. K. Gebbink, Chem. Eur. J. 2015, 21, 15676 –
15685.
[39] L. Anhäuser, M. Teders, A. Rentmesister, F. Glorius, Nat.
Protoc. 2019, 14, 2599 –2626.
[40] W. Ghatts, V. Dubosclard, A. Wick, A. Bendelac, R. Guillot, R.
Ricoux, J.-P. Mahy, J. Am. Chem. Soc. 2018, 140, 8756 –8762.
[41] a) J. Zhao, D. G. Bachmann, M. Lenz, D. G. Grillingham, T. R.
Ward, Catal. Sci. Technol. 2018, 8, 2294 –2298; b) T. K. Hyster,
L. Knorr, T. R. Ward, T. Rovis, Science 2012, 338, 500 –503; c) H.
Yang, P. Srivastava, C. Zhang, J. C. Lewis, ChemBioChem 2014,
15, 223 –227; d) H. Yang, A. M. Swartz, H. J. Park, P. Srivastava,
K. Ellis-Guardiola, D. M. Upp, G. Lee, K. Belsare, Y. Gu, C.
Zhang, R. E. Moellering, J. C. Lewis, Nat. Chem. 2018, 10, 318 –
324; e) P. Srivastava, H. Yang, K. Ellis-Guardiola, J. C. Lewis,
Nat. Commun. 2015, 6, 7789; f) A. R. Grimm, ACS Catal. 2018,
8, 2611 –2614.
[42] a) V. Kçhler, Y. M. Wilson, M. Dürrenberger, D. Ghislieri, E.
Churakova, T. Quinto, L. Knçrr, D. Häussinger, F. Hollman, N. J.
Turner, T. R. Ward, Nat. Chem. 2012, 5, 93 –99; b) D. J. Raines,
J. E. Clarke, E. V. Blagova, E. J. Dodson, K. S. Wilson, A. K.
Duhme-Klair, Nat. Catal. 2018, 1, 680 –688; c) J. Zhao, J. G.
Manuscript received: January 9, 2021
Accepted manuscript online: March 14, 2021
Version of record online: April 6, 2021
12454 www.angewandte.org
ꢀ 2021 Wiley-VCH GmbH
Angew. Chem. Int. Ed. 2021, 60, 12446 –12454