Cross-Linked Artificial Enzyme Crystals as Heterogeneous Catalysts for Oxidation Reactions

Article abstract of DOI:10.1021/jacs.7b09343

Designing systems that merge the advantages of heterogeneous catalysis, enzymology, and molecular catalysis represents the next major goal for sustainable chemistry. Cross-linked enzyme crystals display most of these essential assets (well-designed mesoporous support, protein selectivity, and molecular recognition of substrates). Nevertheless, a lack of reaction diversity, particularly in the field of oxidation, remains a constraint for their increased use in the field. Here, thanks to the design of cross-linked artificial nonheme iron oxygenase crystals, we filled this gap by developing biobased heterogeneous catalysts capable of oxidizing carbon-carbon double bonds. First, reductive O₂ activation induces selective oxidative cleavage, revealing the indestructible character of the solid catalyst (at least 30 000 turnover numbers without any loss of activity). Second, the use of 2-electron oxidants allows selective and high-efficiency hydroxychlorination with thousands of turnover numbers. This new technology by far outperforms catalysis using the inorganic complexes alone, or even the artificial enzymes in solution. The combination of easy catalyst synthesis, the improvement of "omic" technologies, and automation of protein crystallization makes this strategy a real opportunity for the future of (bio)catalysis.

Full text of DOI:10.1021/jacs.7b09343

Subscriber access provided by READING UNIV
Article
Cross-linked artificial enzyme crystals as
heterogeneous catalysts for oxidation reactions
Sarah Lopez, Laurianne Rondot, Chloé Leprêtre, Caroline
Marchi-Delapierre, Stéphane Ménage, and Christine Cavazza
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.7b09343 • Publication Date (Web): 17 Nov 2017
Downloaded from http://pubs.acs.org on November 17, 2017
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of the American Chemical Society is published by the American Chemical
Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 20
Journal of the American Chemical Society
1
2
3
Cross-linked artificial enzyme crystals as heterogeneous catalysts for oxidation reactions
4
5
by Sarah Lopez,1,2,3 Laurianne Rondot, 1,2,3 Chloé Leprêtre, 1,2,3 Caroline MarchiꢀDelapierre,
6
7
1,2,3
Stéphane Ménage*,1,2,3 Christine Cavazza 1,2,3
*
8
9
1 Univ. Grenoble Alpes, Fꢀ38000 Grenoble, France.
2 CEA, BIG, Laboratory of Chemistry and Biology of Metals, BioCE and BioCat group Fꢀ
8054 Grenoble, France.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3 CNRS, UMR5249, Fꢀ38054 Grenoble, France
Abstract: Designing systems that merge the advantages of heterogeneous catalysis,
enzymology and molecular catalysis represents the next major goal for sustainable chemistry.
Crossꢀlinked enzyme crystals display most of these essential assets (wellꢀdesigned
mesoporous support, protein selectivity, molecular recognition of substrates). Nevertheless, a
lack of reaction diversity, particularly in the field of oxidation, remains a constraint for their
increased use in the field. Here, thanks to the design of crossꢀlinked artificial nonꢀheme iron
oxygenase crystals, we filled this gap by developing bioꢀbased heterogeneous catalysts
capable of oxidizing carbonꢀcarbon double bonds. First, reductive O2 activation induces
selective oxidative cleavage, revealing the indestructible character of the solid catalyst (at
least 30 000 turnover numbers without any loss of activity). Second, the use of 2ꢀelectron
oxidants allows selective and highꢀefficiency hydroxychlorination with thousands of turnover
numbers. This new technology by far outperforms catalysis using the inorganic complexes
alone, or even the artificial enzymes in solution. The combination of easy catalyst synthesis,
the improvement of “omic” technologies and automation of protein crystallization makes this
strategy a real opportunity for the future of (bio)catalysis.
1
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 20
1
2
3
Introduction
4
5
6
7
8
9
Catalysis represents a cornerstone of chemical synthesis in a sustainable world.1 Tremendous
achievements have been made by using biocatalysts, which combine mild conditions, high
catalytic cycle number and selectivity, in particular for reduction or oxidation processes.2 To
ensure their relevance for industry, enzyme immobilization is often a preꢀrequisite,
maximizing reusability, avoiding enzyme precipitation and improving the biocatalyst’s
performance.3 Basically, three traditional methods of enzyme immobilization can be
distinguished: covalent (in)organic material, entrapment (e.g. silica gel) and crossꢀlinking.
Crossꢀlinking consists of attaching proteins to each other by a covalent bond, using a
bifunctional reagent, frequently glutaraldehyde which reacts preferentially with the εꢀamino
groups of lysine residues.3b,4 Among the different strategies tested to date, crossꢀlinked
enzyme aggregates (CLEAs) are used as a simpler alternative with a certain level of success,
mostly for hydrolytic reactions (lipase, invertase, hydroxynitrile lyase, cellulase) or reduction
reactions (alcohol dehydrogenase). In contrast, oxidation reactions ꢀ major transformations in
organic synthesis ꢀ are rather scarce, with the exception of a few heme enzymes.3a,5 An
alternative to CLEAs is the CrossꢀLinked Enzyme Crystals (CLEC) approach, which
combines the advantages of enzymology and heterogeneous catalysis.6 In CLEC, the high
chemoꢀ, regioꢀ and enantioꢀselectivity of the enzyme is combined with a high protein
concentration in mesoporous microcrystals. The high solvent content of protein crystals
allows mass transfer from the surface to the active sites, preserves the high catalytic
efficiency of the enzyme and affords remarkable stability, allowing the use of a large range of
solvents and thus substrates, without requiring expensive supports. CLEC technology has
been applied in chiral resolution, peptide synthesis (subtilisin, thermolysin)7 and biosensing
applications (Glucose oxidase)8, but only one example of an oxygen transfer reaction has been
described to date.9 This reaction relied on chloroperoxidase, and its activity was low. This
biotechnological method may be renewed through the development of artificial
metalloenzymes, a hybrid biomaterial formed by embedding an inorganic catalyst (playing the
role of the active site) within a protein scaffold (to drive reaction selectivity). This approach
could produce new biomimetic reactions to achieve nonꢀnatural catalysis or to propose higher
catalytic performance in terms of efficiency, stability and recycling.10 Up to now, only one
example of CLEC using an artificial metalloenzyme has been described. This system
catalyzes the moderate enantioselective reduction of ketones and was produced by soaking
hen egg white lysozyme crystals with Ru(benzene)Cl2 complexes.11
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Here, we report the design of unique heterogeneous catalysts, based on the use of
2
ACS Paragon Plus Environment

Page 3 of 20
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
crossꢀlinked NikA crystals. NikA is a 56 kDa periplasmic nickelꢀbinding protein involved in
nickel import in Escherichia coli. Like other extraꢀcytoplasmic binding proteins, NikA is
composed of two lobes connected by a hinge, in which the ligand is located (Figure S1 and
S2). Based on our previous studies showing that the protein is able to bind nonꢀphysiological
FeꢀEDTAꢀlike catalysts12, we synthesized a series of NikA/Fe-Lx (x= 0-2) CLEC. Our
results reveal the outstanding performance of NikA/Fe-L2 CLEC for oxidative alkene
cleavage by dioxygen, highlighting the high recycling properties of the catalytic biomaterial.
In addition, we demonstrated the versatility of NikA/Fe-L2 CLEC for a panel of oxidative
transformations of carbonꢀcarbon double bonds ꢀ a major transformation in organic
synthesis13 - including oxidative cleavage, enol ether oxidation and hydroxychlorination under
mild conditions.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. Activation of molecular O2 by NikA/Fe-L0 crystals and presentation of the
complexes used in this study.
Results
Design of ligands L1 and L2
In this study, we took advantage of the reactivity of the iron complex Fe-L0 (L0 = Nꢀ
benzylꢀN’ꢀ(2ꢀhydroxybenzyl)ꢀN,N’ꢀethylene diamine diacetic acid), bound to NikA, that
could perform a double intramolecular hydroxylation using dioxygen as the oxidant (Figure
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 20
1
2
3
4
5
6
7
8
9
1A). Our previous results showed that the binding of the reaction product – hydroxylated
ligand ꢀ to the metal inhibited further binding of dioxygen and prevented the catalytic
oxidation of exogenous substrates.14 To circumvent this event, new ligands were designed to
block these reactive positions in L0. Thus, the benzyl ring was replaced by a thiomethyl
(SMe) moiety, whereas a methoxy substituent was added to the leaving ortho position of the
phenol ring, generating the two ligands L1 and L2, respectively (Figure 1B).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Structural characterization of NikA/Fe-L1 and NikA/Fe-L2 and reactivity toward O2.
Xꢀray structures of NikA/Fe-L1, NikA/Fe-L2, NikA/Fe-L1-red, NikA/Fe-L2-red, NikA/Fe-
L1-O2 and NikA/Fe-L2-O2 were solved (Table S1A). The presence of Fe-L1 and Fe-L2
within the crystals was revealed by their red or purple color, respectively (Figure 2A). Xꢀray
fluorescence experiments attested the presence of iron (Figure S3). The crystal structures of
NikA/Fe-L1 and NikA/Fe-L2 showed that, in both cases, Fe(III) is hexacoordinated by three
oxygen atoms from one phenolate and two carboxylates, the two nitrogen atoms from the
ethyleneꢀ1,2ꢀdiamine moiety and a water molecule. Occupancies close to 1 could be assigned
for FeꢀL1 and FeꢀL2 in both structures (Figure 2A). Binding of Fe-L1 and Fe-L2 to NikA
involves a network of supramolecular interactions (Figure S1 and Table S1). The stronger
interaction is mediated by a salt bridge formed between a carboxylate group of the ligand and
Arginine 137. This carboxylate group is stabilized by an interaction with a structural water,
bound to Tyrosine 402. In addition, πꢀstacking interactions are possible with Tryptophan 398,
as previously observed with Fe-L0. Moreover, from a previous study,12 potential weak CH/π
interactions between L1 or L2 and Histidine 416, Tryptophan 100 and Tryptophan 398 would
help the stabilization of Fe-L1 and Fe-L2 in NikA’s binding pocket. To obtain the ferrous
state compatible with dioxygen activation,14 NikA/Fe-Lx crystals were soaked for 60 minutes
in 50 mM DTT in a glove box under anaerobic conditions. After this step, the crystals were
practically colorless (Figure 2B). The reduction step did not lead to the release of either the
iron or the complexes, since the occupancy for Fe-L1 and Fe-L2 remained close to 1. In both
structures, Fe(II) was found to be hexacoordinated with three oxygen atoms from one
phenolate and two carboxylates, the two nitrogen atoms from the ethyleneꢀ1,2ꢀdiamine
moiety and potentially, the sulfur atom of the SMe group (d FeꢀS = 2.4ꢀ2.6 Å) (NikA/Fe-Lx-
Red, Figure 2B). A similar methionine thioether coordination of iron, with a distance of
2.57 Å, was described in isopenicillin N synthase.15 This octahedral coordination resulted
from the Fe(III) to Fe(II) redox transition which induced a conformational change in the
ligand. Thus, the substituted phenyl ring moved to the apical position, previously occupied by
4
ACS Paragon Plus Environment

Page 5 of 20
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
a carboxylate group, which formed a new salt bridge with Arg97 to compensate for the loss of
charge of the complex. The conformational change observed was different to that in the
NikA/Fe-L0-Red structure, where the whole coordination sphere was affected, with the
phenyl ring replacing the phenol ring.14
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2. Crystal structures and dioxygen-reactivity of NikA/Fe-Lx hybrids.
Crystal structures, OMIT Fourier electron density maps and crystal images of NikA/Fe-Lx hybrids at different
reaction stages. (A) NikA/Fe-Lx. (B) NikA/Fe-Lx-red: Reduced NikA/Fe-Lx after soaking for 60 min in an
artificial mother liquor containing 50 mM DTT. (C) NikA/Fe-Lx-O2: NikA/Fe-Lx-red after exposure to air for
4 hours. Electron density maps (shown in green) are contoured to 3s. For clarity, in figure 2C, OMIT Fourier
electron density map was calculated with a model containing L1 to highlight the presence of the hydroxyl
group. Ligands Lx are depicted in orange, and iron is illustrated by a brown sphere. Water molecules are
depicted as red spheres. Aminoꢀacids constituting the complexꢀbinding site are shown as blue sticks. The
interaction distances between Fe and ligands are indicated in Table S1B).
When NikA/Fe-L1-red or NikA/Fe-L2-red crystals were subsequently exposed to air,
the color changed from colorless to purple in both cases (NikA/Fe-Lx-O2, Figure 2C). A
color transition from red to purple was also observed when NikA/Fe-L1 crystals were soaked
in a DTT solution under aerobic conditions, yielding NikA/Fe-L1-O2 crystals. Using the
same method, no color change was observed when NikA/Fe-L2 was transformed into NikA-
5
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 20
1
2
3
4
5
6
7
8
9
Fe-L2-O2. In the case of NikA/Fe-L1-O2, the crystal structure showed an additional electron
density modelled as a single intramolecular hydroxylation of the phenol group at the meta
position producing catechol as the sole product of the reaction.§ In contrast, the NikA/Fe-L2-
O2 structure showed no intramolecular hydroxylation (Figure 2C) and was therefore similar to
the initial NikA/Fe-L2 since, unlike L1, L2 lacks a CꢀH group close enough to be directly
attacked by the ironꢀbased oxidative species. In both NikA/Fe-L2-O2 and NikA/Fe-L1-O2
structures, an exchangeable position, necessary to dioxygen activation, is present in the
coordination sphere of the iron. This position was modeled as a water molecule. Therefore,
unlike NikA/Fe-L0, the two new artificial enzymes NikA/Fe-L1 and NikA/Fe-L2 can
potentially perform oxidation reactions on exogenous substrates.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Oxidation of carbon-carbon double bonds by NikA/Fe-L2 CLEC with O2 as the oxidant.
To increase the stability of the hybrids in organic solvents, CLEC technology was
applied, making it possible to use a large range of waterꢀinsoluble substrates for catalysis.
After crossꢀlinking, the CLEC were stable for several months in a mixture of water and
organic solvents. The integrity of NikA/Fe-Lx CLEC was verified by Xꢀray crystallography
and Xꢀray fluorescence (Figures S4 and S5, respectively) before they were used in catalysis.
Occupancy of 1 could be assigned for L0, L1, L2 and iron in the three protein structures.
Furthermore, proteinꢀbound iron content was determined under reducing conditions with
bathophenantroline disulfonate after acidic treatment of NikA/Fe-L2 CLEC,16 revealing a
close to 1:1 ratio between iron and the concentration of protein measured by SECꢀMALLS
(see SI).
We assessed the ability of NikA/Fe-L2 CLEC, the most stable hybrid under
dioxygen, to activate molecular dioxygen in the presence of DTT, and to oxidize the carbonꢀ
carbon double bond in βꢀmethoxystyrene, an enol ether of 2ꢀphenylacetaldehyde (Table 1).
Two products were obtained, benzaldehyde and 2ꢀmethoxyacetophenone, at yields of 38 and
36%, respectively, for a total TON of 550 (entry 1). In the absence of CLEC, no products
were detected (entry 4). Moreover, the use of NiꢀEDTA, a previously described ligand for
NikA,17 instead of Fe-L2 led to an inactive NikA/Ni-EDTA CLEC (entry 5). Consequently,
the efficiency of CLEC catalyst depends upon the iron complex inserted and the
glutaraldehyde treatment doesn’t affect the reactivity. Remarkably, NikA/Fe-L2 CLEC was
more efficient than either Fe-L2 or NikA/Fe-L2, which only performed 130 TON (entries 2ꢀ
3) over 90 minutes. We next expanded the range of substrate to alkenes, and exclusively
oxidative cleavage was obtained with yields of between 0 and 26% (entries 6ꢀ10ꢀ14ꢀ18ꢀ21). It
6
ACS Paragon Plus Environment

Page 7 of 20
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
is worth mentioning that with alkene substrates the reaction could only be catalyzed by the
CLEC system. Under the same conditions, cyclic aromatic double bonds in indene and 1,2ꢀ
dihydronaphthalene and aliphatic alkenes were not oxidized. Allowing a longer reaction time
(24 h) only benefited αꢀmethylstyrene and 4ꢀmethoxystyrene transformations (yield of 27%
9
and 51%, respectively, table 1, entries 6 and 10). The electronic effect of various substituents
on the phenyl ring of the styrenyl derivatives drove the reactivity of the double bond, in line
with an electrophilic addition.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 1. Oxidation of styrene derivatives by O2.
Entry
1
Substrate
Catalyst
Aldehyde[b] (%)
Ketone[b] (%)
TON[c]
550
130
130
0
38
36
NikA/Fe-L2 CLEC
2
8
9
Fe-L2
NikA/Fe-L2 hybrid[a]
3
11
6
4
0
0
-
5
0
0
0
NikA/Ni-EDTA CLEC
6
26(51[d])
0
200
5
NikA/Fe-L2 CLEC
7
2
6
0
0
0
0
0
22
0
0
0
2
0
0
0
0
Fe-L2
8
NikA/Fe-L2 hybrid[a]
0
50
0
9
ꢀ
0
10
11
12
13
14
15
16
17
18
19
20
21
7(27[d])
50
0
NikA/Fe-L2 CLEC
0
0
0
0
0
0
0
0
0
0
0
Fe-L2
NikA/Fe-L2 hybrid[a]
0
ꢀ
0
170
0
NikA/Fe-L2 CLEC
Fe-L2
NikA/Fe-L2 hybrid[a]
0
ꢀ
0
15
0
NikA/Fe-L2 CLEC
Fe-L2
0
NikA/Fe-L2 CLEC
Fe-L2
0
[a] Catalysis was performed in HEPES buffer 10 mM pH 7.5. [b] Yield calculated relative to initial DTT
concentration. [c] Errors evaluated to 14% (see experimental section). [d] Yield after 24 h.
7
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 20
1
2
3
4
5
Furthermore, the reaction was reductantꢀdependent. Thus, a direct correlation between
product yield and DTT concentration was observed, with a correlation factor close to one
6
7
(Figure S6) for βꢀmethoxystyrene oxidation. Other reductants, such as ascorbate or TCEP,
were ineffective under the experimental conditions tested. Reducing the metal prior to
substrate exposure with only 10 equivalents of DTT, to allow direct O2 activation, was also
incompatible with catalysis (less than 2% yield).
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Kinetics revealed a lag phase of about 30 minutes for NikA/Fe-L2 CLEC and Fe-L2
(Figure S7A and B). Preꢀincubation of both catalysts with DTT under anaerobic conditions
before adding substrate eliminated this lag phase and the reaction was completed, indicating
that reduction of the ferric state represents a rate limiting step for the reaction (Figure S8).
18Oꢀlabeled water was used to discriminate between potential oxidative pathways. None of the
products (aldehyde or ketone) produced from
β
ꢀmethoxystyrene or cisꢀstilbene were 18Oꢀ
labeled, indicating that the oxygen atoms inserted were derived from 16O2 (see SI).
Furthermore, addition of DMPO (75% vs. the substrate) completely inhibited the formation of
any products, whatever the substrate, demonstrating that the reaction involves radicals.
However, no DMPOꢀOH• or DMPOꢀOOH• radicals could be identified by EPR, suggesting
that free oxygenated radicals are not produced during the process.
Stability experiments were performed with the different catalysts. The activity of the
NikA/Fe-L2 decreased drastically in buffered medium with 210 TON performed over 3 runs
(Figure 3A). Fe-L2 was slightly more efficient, with 540 TON performed over 6 runs.
Remarkably, NikA/Fe-L2 CLEC remained fully active even after 50 successive runs, with
more than 28 000 TON performed with no loss of activity (Figure 3A). Xꢀray fluorescence
spectra showed the persistent presence of iron in NikA/Fe-L2 CLEC after 20 runs,
confirming that the catalyst was mostly retained in the crystal (Figure S5). Moreover, the
kinetics of the reaction were unaltered during successive runs, emphasizing the absence of
leaching of the catalyst (Figure S7C). In contrast, a steady loss of catalytic efficiency was
observed with NikA/Fe-L0 CLEC and NikA/Fe-L1 CLEC after 8 and 12 runs, respectively,
indicating that the catalytic reaction is centered on the iron complex and that the efficiency
(35% yield for NikA/Fe-L0 CLEC and 42% yield for NikA/Fe-L1 CLEC) and stability are
complexꢀdependent (Figure 3B).
8
ACS Paragon Plus Environment

Page 9 of 20
Journal of the American Chemical Society
1
2
3
Figure 3. Stability experiments relating to O2/DTT reactivity.
4
5
A. NikA/Fe-L2 CLEC vs. Fe-L2 and NikA/Fe-L2 hybrid.
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
B. NikA/Fe-L2 vs. NikA/Fe-L1 and NikA/Fe-L0.
Reaction performed with βꢀmethoxystyrene. Each point is a standard reaction, always using the same batch of
100 CLEC (see Experimental Section). Ratio Catalyst/Substrate/DTT: 1/1000/750, 3 h, [CLEC] = 31 ꢀM in
acetonitrile:water 1:1.
9
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 20
1
2
3
4
5
Optimization of the catalytic properties of NikA/Fe-L2 CLEC towards chlorohydrin
production.
6
7
8
To evaluate the catalytic efficiency of CLEC for double bond transformations, we
screened a range of oxidants to monitor the oxidation of enol ether and αꢀmethylstyrene, +/ꢀ
9
chloride for potential chlorohydrin formation (Table S2). H2O2, CH3CO3H and tꢀBuOOH
were inefficient, whereas KHSO5, NaOCl and mꢀCPBA provided reasonable conversion and
selectivity. KHSO5 and mꢀCPBA allowed production of equimolar amounts of ketone and
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
aldehyde from
β
ꢀmethoxystyrene, with a slightly lower conversion than during reductive
ꢀmethylstyrene, low conversion (less than 15% yield)
dioxygen activation. In the case of
α
and/or moderate chemoselectivity (around 75%) were observed. Conversely, in the presence
of chloride, KHSO5 was identified as the most promising oxidant as it produced a chemoꢀ and
regioꢀselective hydroxychlorination yielding 1ꢀchloroꢀ2ꢀphenylpropanꢀ2ꢀol as sole product
(Table 2, entry 1). As controls, Fe-L2 alone and NikA/Ni-EDTA CLEC were found quasi
inactive, demonstrating that the reaction was driven by NikA/Fe-L2 CLEC.
A series of styrene derivatives were effectively hydroxychlorinated with exclusive
Markovnikov regioselectivity (Table 2). However, chemoselectivity varied between 29% and
100% as a dichlorinated product was also detected (for cisꢀstilbene and 4ꢀmethoxystyrene,
respectively). Styrenyl substrates with an electronꢀdonating moiety were attacked to give
quasiꢀquantitative yields; for example 4ꢀmethoxystyrene was transformed into 2ꢀchloroꢀ1ꢀ(4ꢀ
methoxyphenyl)ꢀethanol in 30 min (480 TON, entry 4). However, when substrates contained
an electronꢀwithdrawing substituent such as 4ꢀbromostyrene, the reaction was drastically
slowed, producing only 4% of chlorohydrin product after 30 min, and 26% in 14 hours (entry
5). These results indicate that the reaction is favored in the case of electronꢀrich double bonds,
hinting at the electrophilic nature of the oxidative species. As a consequence, the reaction of
aliphatic cyclic alkenes such as methylcyclohexene resulted in a low yield of chlorohydrin
(only 5%, entry 9). E-Z stereoisomerism did not impact the yield of the reaction, but did affect
chemoselectivity, since transꢀßꢀmethylstyrene gave 92% while the cisꢀßꢀmethylstyrene only
produced 79% (entries 2 and 3). An effect of steric hindrance was also observed, with
stilbenes only providing low yields (41% for the cisꢀ and 19% for the transꢀ isomers, entries 7
and 8) and rather poor chemoselectivity. NikA/Fe-L2 CLEC led to very high
diastereoselectivity (100:0 in favor of the trans diastereomeric pair), but poor
enantioselectivity as hydroxychlorination of the most rigid substrate, 1,2ꢀdihydronaphthalene
(entry 6), produced only 10% ee (Figure S9).
10
ACS Paragon Plus Environment

Page 11 of 20
Journal of the American Chemical Society
1
2
3
4
5
Table 2. Range of substrates tested with NikA/Fe-L2 CLEC
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Entry
1
Substrate
Yield[a] (%)
98
TON[b]
490
Selectivity[c] (%)
80
2
3
4
5
75
72
380
360
480
20
79
92
96
100
100
4(26[d])
6
7
8
9
97[e]
41
490
200
100
30
100
29
19
37
5f
100
Ratio Catalyst/Substrate/Oxidant/Chloride: 1/1000/750/1500, 30 min, [CLEC] = 31 ꢀM in acetonitrile:water 1:1.
[a] Yield calculated relative to initial KHSO5 concentration. [b] Errors evaluated to 14% (see experimental
section). [c] % of chlorohydrin relative to dichlorinated product. [d] Reaction time = 14 h. [e] ee = 10%. [f]
When methylcyclohexene is used product is 2ꢀchloroꢀ1ꢀmethylcyclohexanꢀ1ꢀol.
The conversion yield of 4ꢀmethoxystyrene was complexꢀdependent (Figure 4). The
yield was significantly decreased with NikA/Fe-L0 CLEC (17% yield i.e., 130 TON vs. 84%
yield of chlorohydrin product i.e., 630 TON with NikA/Fe-L1 CLEC). Although the yield
and TON were similar with NikA/Fe-L2 CLEC, the kinetics of 2ꢀchloroꢀ1ꢀ(4ꢀ
methoxyphenyl)ꢀethanol production by NikA/Fe-L1 CLEC was slightly faster, with a TOF of
6500 hꢀ1 (4600 hꢀ1 for NikA/Fe-L2 CLEC) (Figure S10). Increasing the temperature from 20
to 70 °C resulted in a sevenꢀfold increase in yield after only one minute of reaction (Figure
S11A).
The stability of the three CLEC systems was tested during conversion of 4ꢀ
methoxystyrene (Figure 4). With NikA/Fe-L0 CLEC, the reactivity rapidly decreased during
11
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 12 of 20
1
2
3
4
5
6
7
8
9
the three first catalytic runs (only 4% yield during the third run), with a total TON of 270.
NikA/Fe-L1 CLEC retained full activity over 3 catalytic runs but then decreased drastically;
a total of 3200 TON were performed over 13 catalytic runs. In comparison, NikA/ Fe-L2
CLEC retained full activity over 7 catalytic runs (and 3 runs at 70 °C, Figure S11B) and then
decreased progressively, performing 5900 TON over 13 catalytic runs. Activity decreased as
the crystals were bleached from purple to light yellow, corresponding to the loss of iron, as
shown by Xꢀray fluorescence (Figure S5). In a second experiment, NikA/Fe-L2 CLEC were
soaked in Fe-L2 at runs 7, 9, 11 and 13. In these conditions, the catalytic efficiency remained
stable for 9 runs, with yields decreasing slightly from 84% to 70% in the tenth run for a total
of 6600 TON over 13 runs (Figure S12). Beyond 13 catalytic runs, NikA/Fe-L2 CLEC were
practically inactive despite additional Fe-L2 soaking. This result suggests that the full system
underwent some oxidative damage. Soaking NikA/Fe-L2 CLEC in FeCl3 at runs 7, 9, 11 and
13 had no effect on activity (Figure S13), indicating that the decrease in activity following
several runs was related to loss or degradation of the full complex.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4. Stability experiments for hydroxychlorination of 4-methoxystyrene by
NikA/Fe-Lx CLEC.
Reaction performed on 4ꢀmethoxystyrene. Each point is a standard reaction always using the same batch of 100
CLEC (see Experimental Section). Ratio Catalyst/Substrate/Oxidant/Chloride: 1/1000/750/1500, 30 min,
[CLEC] = 31 ꢀM in acetonitrile:water 1:1.
Raising the pH from 2.5 (pH value in the standard catalytic conditions) to 6.0 inhibited
the hydroxychlorination reaction, thus only alcohols and carboxylic acids could be used as
nucleophiles (Figure S14). Labeling experiments using H218O confirmed the existence of a
12
ACS Paragon Plus Environment

Page 13 of 20
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
nucleophilic attack (the final product contained 85% oxygenꢀlabeled 2ꢀchloroꢀ1ꢀ(4ꢀ
methoxyphenyl)ꢀethanol (Figure S15). Thus, the primary step in the reaction involves an
electrophilic addition of a chloronium into the double bond produced by the metalꢀbased
activation by KHSO5. This addition is compatible with the absence of inhibition by the radical
scavenger DMPO, which excludes the formation of Cl•, SO4• and OH• as reactive
intermediates.18 Monitoring the reaction of a mixture of KHSO5 and NaCl by UVꢀvisible in
the absence of substrate (Figure S16) revealed a slow buildup of a unique transition at 310
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
nm, suggesting the presence of Cl2;19 free HOCl ( max = 240 nm)20 or OClꢀ( max = 296 nm)19
λ λ
were undetectable. Cl2 forms from the conversion of HOCl in acidic medium, and its
appearance (as indicated by the 310 nm transition) was inhibited by the presence NikA/Fe-L2
CLEC.
Discussion
The CLEC reported here presents considerable assets for heterogeneous catalysis. The
unexpected efficiency under oxidative conditions, even using KHSO5 or NaOCl, are
unprecedented in (bio)catalysis, except with some heme catalysts or isolated hemoenzymes.21
Moreover, selective oxidation processes relying on O2 are problematic since they involve
radical chemistry, and synthetic metallic catalysts are inherently susceptible to degradation by
oxygenꢀbased radicals. However, with TOF values ranging from 2 to 30 minꢀ1, our hybrid
systems are comparable with natural enzymes.22 This efficiency can be explained by two main
factors: first, the crossꢀlinking method produces a robust NikAꢀbased hybrid crystal that is
highly stable even under strong acidic conditions and high temperatures. Second, the
mesoporous character of the highꢀcatalystꢀcontent crystals considerably benefits the reactivity
(Figure S2). Indeed, the easy diffusion of molecules within a crystal in buffered solution or
organic solvents has been demonstrated. Moreover, glutaraldehyde treatment did not appear
to dramatically alter the crystal packing or the network of solvent channels. Thus, the
diffusion of substrates (DTT, oxidants, and Fe-Lx) in soaking experiments could be
demonstrated by the high catalytic efficiencies of NikA/Fe-Lx CLEC. In addition, the solid
phase should influence the kinetics of the catalysis, and consequently the efficiency, for
several reasons: i) the active species is locally highly concentrated while dispersed in the
solution either for the complex alone or the hybrid; ii) the use of enzymes is often limited by
their poor stability under reaction conditions, especially in the field of oxidation. In this study,
we showed that NikA/Fe-Lx hybrids in solution were poorly active because of the
13
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 14 of 20
1
2
3
4
5
6
7
8
requirement of organic solvent for the reaction. In the same way, the use of KHSO5 under
acidic conditions led to the immediate degradation of the hybrids; iii) in some cases, the
complex may be activated by the protein scaffold.14,23 The case of O2/DTT is symptomatic of
another aspect of the influence of the solid state, i.e., the compartmentalization of the catalyst
9
with respect to its substrates and reactants. For example, in solution, cis-stilbene and
αꢀ
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
methylstyrene were not transformed by O2 and only minor activity was observed with 4ꢀ
methoxystyrene or ßꢀmethoxystyrene (yield 0 vs. 11%, 0 vs. 7%, 17 vs. 26% and 17 vs. 78%,
respectively). The yield for competitive oxidation of DTT was clearly enhanced for
homogeneous catalysts compared to CLEC catalysts when performing oxidative cleavage of
alkenes (Figure S17). The better synchronization between O2 reduction steps and alkene
oxidation suggests controlled traffic of both substrate and DTT inside the crystal. The CLEC
system also provided better chemoꢀ and stereoꢀselectivity compared to inꢀsolution Fe-Lx
complexes, mainly by restricting intermolecular side reactions and the influence of the protein
scaffold.
For cleavage of nonꢀaromatic double bonds, O2 is rarely used, particularly with ironꢀbased
catalysts.24 In nature, all but one25 of the nonꢀheme or heme enzymes display a dioxygenaseꢀ
type behavior.26 In contrast, the oxidative catalytic cleavage induced by NikA/Fe-Lx CLEC
resembles a monooxygenase behavior since it is reductantꢀdependent and a 2eꢀ oxidant can
replace dioxygen (Table S2B). However, its expected products, epoxides or diols,25 were not
detected. Moreover, pure samples of diols or epoxides remained inert under our oxidative
conditions. Altogether, these results suggest a novel mechanistic pathway. O2 activation by
the reduced form of NikA/Fe-Lx CLEC initially led to the formation of a transient iron
superoxo species, prior to forming the peroxo adduct structurally characterized in our
previous study.14 Only the involvement of a peroxo adduct is compatible with both the radical
nature and chemoselectivity of the reaction if an ‘in cage’ hydroxyl radical resulting from a
homolytic cleavage of the OꢀO bond (no aromatic oxidation was detected and the substrate
promiscuity is too low for free OH•).27 Moreover, peroxo iron species should not develop
radical chemistry.28 High valent species such as Fe(IV)=O can be eliminated based on the
absence of labeling scrambling.29 Conversely, superoxo adducts have been identified as
intermediates in radicalꢀmediated oxidative alkene cleavage and were shown to be capable of
electrophilic addition to double bonds.24a,28a In our case, the superoxo iron species or the “in
cage” hydroxyl radical should be added to a carbon on the double bond to produce a benzyl
radical at the
β
position that further reacts with dioxygen prior to cleavage. Identification of
14
ACS Paragon Plus Environment

Page 15 of 20
Journal of the American Chemical Society
1
2
3
4
5
the radical species and the reaction steps following radical formation on the substrate will
need to be clarified, as will the mechanism for ketone formation from the enol.
6
7
8
9
Catalytic hydroxychlorination activity has rarely been observed with natural enzymes
(mainly as a byꢀproduct of chloroperoxidases).30 Here again, the NikA/Fe-L2 CLEC
displayed a unique reactivity. Three mechanistic activation scenarii are possible (Scheme 1):
i) activation of KHSO5 by Fe-L2 following its binding and heteroꢀ or homoꢀlytic cleavage of
the OꢀO(S)O2 bond. Homolytic cleavage should lead to radical chemistry and is therefore
excluded in this case. (ii) FeL2 plays the role of a Lewis acid activating the proximal oxygen
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ꢀ
(then being more positive) of an interacting SO5 . A direct attack of the chloride anion would
generate the (H)OCl species either free in acidic medium or coordinated with the iron. The
absence of reactivity when OClꢀ was substituted for KHSO5 precludes the existence of this
iron species. HOCl has generally been proposed to be the oxidizing species with KHSO5, but
in our case the chemoselectivity observed was different (4ꢀbromostyrene and cyclic alkenes
were inert in this study).31 iii) Metalꢀbased polarization of Cl2 produced from HOCl, which is
the pathway we favor.
Scheme 1. Possible pathways of KHSO5 activation to generate Cl2 as the chloronium
source.
Conclusions
In this article we describe and characterize the first CLEC with artificial enzymes
capable of performing an oxidation reaction. The stability of the biomaterial enables very
complex reactions, such as selective catalysis by O2 activation. The final asset of our
approach lies in the straightforward the preparation of our hybrid based on supramolecular
interactions, allowing easy loading or exchange of the inorganic catalyst. With this
technology, new ways to use modular NikAꢀbased CLEC are opened, with a broad scope of
15
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 16 of 20
1
2
3
4
5
6
7
complexes to perform a wide range of reactions in organic solvents. More importantly, the
increased stability compared to enzymes in solution (pH, temperature, oxidation conditions)
should extend the field for catalysis as an alternative to biocatalysis.
8
9
Experimental Section
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Synthesis of ligands, complexes and products are described in SI.
Purification of NikA, crystallization and synthesis of cross-linked NikA/Fe-Lx crystals
(NikA/Fe-Lx CLEC)
Cytoplasmic apoꢀNikA was purified as previously described.32 NikA/Fe-L1, NikA/Fe-L2,
NikA/Fe-L0 and NikA/Ni-EDTA crystals were obtained as previously described for
NikA/Fe-L0.14
NikA/Fe-Lx and NikA/Ni-EDTA crystals were crossꢀlinked in 5% glutaraldehyde, 2.1 M
ammonium sulfate and 100 mM TrisꢀHCl pH 7.0 for 5 h. Typically, crystals were soaked by
addition of 5 μL of a solution of 10% glutaraldehyde in 2.1 M ammonium sulfate and 100
mM TrisꢀHCl pH 7.0. Then, NikA/Fe-Lx CLEC were transferred to a 50%, (v/v)
CH3CN/H2O solution for catalysis experiments. For crystallographic studies, NikA/Fe-Lx
CLEC were transferred to a 30% (v/v) glycerol/H2O solution before flash cooling in liquid
nitrogen.
Crystallographic studies of NikA/Fe-Lx in the presence of DTT and O2
NikA/Fe-L1 and NikA/Fe-L2 crystals were transferred to an anaerobic glove box (Jacomex)
and soaked in an artificial mother liquor containing 100 mM TrisꢀHCl pH 7.0, 2.1 M
ammonium sulfate, mixed with 50 mM DTT. After soaking for 60 min, the colorless reduced
NikA/Fe-Lx-red crystals were exposed to air for four hours to yield NikA/Fe-Lx-O2 crystals.
NikA/Fe-Lx crystals were also soaked four hours in the mother liquor containing 50 mM
DTT in the presence of air to yield NikA/Fe-Lx-O2 crystals. All crystals were cryoꢀprotected
by adding 25% (v/v) glycerol to the mother liquor before flash cooling in liquid nitrogen.
Representative procedure for O2 activation for a 1/1000/750 (catalyst/substrate/DTT)
ratio with CLEC.
100 NikA/Fe-Lx CLEC, fished directly from the soaking solution, were washed twice in 1
mL of acetonitrile/water 1:1 solution. The supernatant was removed, leaving a final volume of
37 µL. Then, 1.5 µL of a 0.82 M solution of substrate (1.23 µmol) in 100% acetonitrile and
16
ACS Paragon Plus Environment

Page 17 of 20
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
1.5 µL of a 0.62 M DTT solution (0.92 µmol) were added successively, resulting in a final
volume of 40 µL. The reaction was stirred at room temperature for 3 hours, 1.5 µL of
benzophenone 0.1 M was added and the reaction mixture was extracted with AcOEt (60 µL).
The organic phase was analyzed by GC/MS. The final concentration of CLEC was 31 ± 5 µM
based on MALLS measurements.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Representative procedure for hydroxychlorination catalysis for a 1/500/600/1000
Catalyst/Substrate/Oxidant/Chloride (C/S/O/Cl) ratio with CLEC
100 NikA-Fe-Lx CLEC were washed twice in 1 mL of acetonitrile/water 1:1 solution. The
supernatant was removed, leaving a final volume of 30 µL. Then 1.5 µL of a 0.41 M solution
of substrate (0.615 µmol) in 100% acetonitrile, 4 µL of a 0.31 M sodium chloride solution
(1.23 µmol) and 5 µL of a 0.29 M KHSO5 solution (0.725 µmol) were added successively to
give a final volume of 40.5 µL. The final CLEC concentration was 31 ± 5 µM. The reaction
was stirred at room temperature for 30 min, then 1.5 µL of benzophenone 0.1 M was added
and the reaction mixture was extracted with AcOEt (60 µL). The organic phase was analyzed
by GC/MS. A catalyst concentration of 31 µM was used to calculate ratios and determine
TON and TOF. To monitor the kinetics, a reaction mixture was prepared for each point and
was measured in duplicate. Reaction mixtures were prepared according to the above
procedure. The TOF was calculated from the linear slope of the curve of the TON as a
function of time between 0 and 3 minutes.
Stability experiments
For stability experiments, the standard procedures were applied for the first run. Then CLEC
were washed with 1 mL of acetonitrile:water 1:1 and reused for the second run. The same
procedure was applied for as many runs as possible. In the case of homogeneous catalysis, the
oxidant and substrate were added on three separate samples: first, with an extraction after 90
min; second with an extraction after two successive additions; third with an extraction after
three successive additions and so on. The mixture was then extracted with AcOEt and
analyzed for its content.
For the reloading experiments, after runs 7, 9, 11 and 13, NikA/Fe-L2 CLEC were soaked in
50 μL of 5 mM FeCl3 or Fe-L2 solution for 3 h before washing three times with 1 mL of an
acetonitrile/water 1:1 solution to remove excess free complex from the solution.
ACKNOWLEDGEMENTS
17
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 18 of 20
1
2
3
4
We thank the staffs from the BMꢀ30A and ID30Aꢀ3 beamlines of the European Synchrotron
Facility in Grenoble.
5
6
7
ASSOCIATED CONTENT
8
9
Supporting Information. Ligands and complexes synthesis and their physical
characterization, crystal structures, catalytic procedures and kinetic data are reported. This
material is available free of charge via the Internet at http://pubs.acs.org.”
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
AUTHOR INFORMATION
Corresponding Author
* To whom correspondence should be addressed: christine.cavazza@cea.fr,
stephane.menage@cea.fr
Funding Sources
This work was supported by the French National Agency for Research (grant ANRꢀ14ꢀCE06ꢀ
0005ꢀ01), Labex ARCANE (ANRꢀ11ꢀLABEXꢀ0003ꢀ1), and recurrent funding from CEA,
CNRS and Univ. GrenobleꢀAlpes.
References
(1)
a) Anastas, P. T.; Kirchhoff, M. M. Acc. Chem. Res. 2002, 35, 686; b) Poliakoff, M.;
Fitzpatrick, J. M.; Farren, T. R.; Anastas, P. T. Science 2002, 297, 807; c) Sheldon, R. A. Chem. Soc.
Rev. 2012, 41, 1437.
(2)
a) Bornscheuer, U. T.; Huisman, G. W.; Kazlauskas, R. J.; Lutz, S.; Moore, J. C.; Robins, K.
Nature 2012, 485, 185; b) Faber, K.; Kroutil, W. Curr. Opin. Chem. Biol. 2005, 9, 181; c) Nestl, B. M.;
Hammer, S. C.; Nebel, B. A.; Hauer, B. Angew. Chem. Int. Ed. 2014, 53, 3070; d) Breuer, M.; Ditrich,
K.; Habicher, T.; Hauer, B.; Keßeler, M.; Stürmer, R.; Zelinski, T. Angew. Chem. Int. Ed. 2004, 43,
788.
(3)
a) Tran, D. N.; Balkus, K. J. ACS Catal. 2011, 1, 956; b) Sheldon, R. A.; van Pelt, S. Chem.
Soc. Rev. 2013, 42, 6223; c) GarciaꢀGalan, C.; BerenguerꢀMurcia, Á.; FernandezꢀLafuente, R.;
Rodrigues, R. C. Adv. Synth. Cat. 2011, 353, 2885.
(4)
a) Yan, E.ꢀK.; Cao, H.ꢀL.; Zhang, C.ꢀY.; Lu, Q.ꢀQ.; Ye, Y.ꢀJ.; He, J.; Huang, L.ꢀJ.; Yin, D.ꢀC.
RSC Adv. 2015, 5, 26163; b) Bornscheuer, U. T. Angew. Chem. Int. Ed. 2003, 42, 3336.
(5)
(6)
Hanefeld, U.; Cao, L.; Magner, E. Chem. Soc. Rev. 2013, 42, 6211.
a) Margolin, A. L.; Navia, M. A. Angew. Chem. Int. Ed. 2001, 40, 2204; b) Roy, J. J.;
Abraham, E. T. Chem. Rev. 2004, 104, 3705; c) Quiocho, F. A.; Richards, F. M. PNAS 1964, 52, 833.
(7) Persichetti, R. A.; Clair, N. L. S.; Griffith, J. P.; Navia, M. A.; Margolin, A. L. J. Am. Chem.
Soc. 1995, 117, 2732.
(8)
a) Luiz de Mattos, I.; Lukachova, L. V.; Gorton, L.; Laurell, T.; Karyakin, A. A. Talanta 2001,
54, 963; b) Khalaf, N.; Govardhan, C. P.; Lalonde, J. J.; Persichetti, R. A.; Wang, Y.ꢀF.; Margolin, A.
L. J. Am. Chem. Soc. 1996, 118, 5494.
(9)
Ayala, M.; Horjales, E.; Pickard, M. A.; VazquezꢀDuhalt, R. Biochem. Biophys. Res. Commun.
2002, 295, 828.
(10)
a) Yu, F.; Cangelosi, V. M.; Zastrow, M. L.; Tegoni, M.; Plegaria, J. S.; Tebo, A. G.; Mocny, C.
S.; Ruckthong, L.; Qayyum, H.; Pecoraro, V. L. Chem. Rev. 2014, 114, 3495; b) MarchiꢀDelapierre,
C.; Rondot, L.; Cavazza, C.; Ménage, S. Isr. J. Chem. 2015, 55, 61; c) Schwizer, F.; Okamoto, Y.;
Heinisch, T.; Gu, Y.; Pellizzoni, M. M.; Lebrun, V.; Reuter, R.; Köhler, V.; Lewis, J. C.; Ward, T. R.
Chem. Rev. 2017, 10.1021/acs.chemrev.7b00014.
(11)
T.; Kitagawa, S.; Ueno, T. Chem. Asian J. 2014, 9, 1373.
(12) Cherrier, M. V.; Girgenti, E.; Amara, P.; Iannello, M.; MarchiꢀDelapierre, C.; Fontecillaꢀ
a) Abe, S.; Maity, B.; Ueno, T. Chem. Commun. 2016, 52, 6496; b) Tabe, H.; Abe, S.; Hikage,
Camps, J. C.; Menage, S.; Cavazza, C. J. Biol. Inorg. Chem. 2012, 17, 817.
18
ACS Paragon Plus Environment

Page 19 of 20
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
(13)
(14)
Bäckvall, J. E. Modern Oxidation Methods; WileyꢀVCH, Weinheim, 2004.
Cavazza, C.; Bochot, C.; RousselotꢀPailley, P.; Carpentier, P.; Cherrier, M. V.; Martin, L.;
MarchiꢀDelapierre, C.; FontecillaꢀCamps, J. C.; Menage, S. Nat. Chem. 2010, 2, 1069.
(15)
1881.
(16)
(17)
Clifton, I. J.; Ge, W.; Adlington, R. M.; Baldwin, J. E.; Rutledge, P. J. Chembiochem 2011, 12,
Fish, W. W. Methods Enzymol. 1988, 158, 357.
Addy, C.; Ohara, M.; Kawai, F.; Kidera, A.; Ikeguchi, M.; Fuchigami, S.; Osawa, M.;
Shimada, I.; Park, S.ꢀY.; Tame, J. R. H.; Heddle, J. G. Acta Crystallogr., Sect. D 2007, 63, 221.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(18)
(19)
1994, 33, 5872.
(20)
(21)
Hussain, H.; Green, I. R.; Ahmed, I. Chem. Rev. 2013, 113, 3329.
Wang, T. X.; Kelley, M. D.; Cooper, J. N.; Beckwith, R. C.; Margerum, D. W. Inorg. Chem.
Barnes, R. J.; Sinha, A.; Michelsen, H. A. J. Phys. Chem. A 1998, 102, 8855.
a) Meunier, B.; de Visser, S. P.; Shaik, S. Chem. Rev. 2004, 104, 3947; b) Handbook of
porphyrin science; Kadish, K.; Smith, K.; Guilard, R., Eds.; World Scientific, 2010; Vol. 10; c)
Urlacher, V. B.; Girhard, M. Trends Biotechnol. 2011, 30, 26; d) Fasan, R. ACS Catal. 2012, 2, 647.
(22)
George, A.; Sale, K. L.; Fox, B. G.; Simmons, B. A.; Adams, P. D. PNAS 2016, 113, 14324; b)
Chauhan, N.; Thackray, S. J.; Rafice, S. A.; Eaton, G.; Lee, M.; Efimov, I.; Basran, J.; Jenkins, P. R.;
Mowat, C. G.; Chapman, S. K.; Raven, E. L. J. Am. Chem. Soc. 2009, 131, 4186.
a) McAndrew, R. P.; Sathitsuksanoh, N.; Mbughuni, M. M.; Heins, R. A.; Pereira, J. H.;
(23)
a) RousselotꢀPailley, P.; Bochot, C.; MarchiꢀDelapierre, C.; JorgeꢀRobin, A.; Martin, L.;
FontecillaꢀCamps, J. C.; Cavazza, C.; Menage, S. Chembiochem 2009, 10, 545; b) Esmieu, C.;
Cherrier, M. V.; Amara, P.; Girgenti, E.; MarchiꢀDelapierre, C.; Oddon, F.; Iannello, M.; JorgeꢀRobin,
A.; Cavazza, C.; Ménage, S. Angew. Chem. Int. Ed. 2013, 52, 3922; c) Rondot, L.; Girgenti, E.;
Oddon, F.; MarchiꢀDelapierre, C.; JorgeꢀRobin, A.; Ménage, S. J. Mol. Catal. A: Chem. 2016, 416, 20;
d) Lopez, S.; Rondot, L.; Cavazza, C.; Iannello, M.; BoeriꢀErba, E.; Burzlaff, N.; Strinitz, F.; Jorgeꢀ
Robin, A.; MarchiꢀDelapierre, C.; Menage, S. Chem. Commun. 2017, 53, 3579.
(24)
a) Spannring, P.; Bruijnincx, P. C. A.; Weckhuysen, B. M.; Klein Gebbink, R. J. M. Catal. Sci.
Technol. 2014, 4, 2182; b) Chen, H.; Ji, H.; Zhou, X.; Xu, J.; Wang, L. Catal. Commun. 2009, 10, 828.
(25)
(26)
A. Arch. Biochem. Biophys. 2014, 544, 18; c) Harrison, P. J.; Bugg, T. D. H. Arch. Biochem. Biophys.
2014, 544, 105.
(27)
O. J. Org. Chem. 2010, 75, 3705.
(28) a) Ray, K.; Pfaff, F. F.; Wang, B.; Nam, W. J. Am. Chem. Soc. 2014, 136, 13942; b) Oloo, W.
N.; Que, L. Acc. Chem. Res. 2015, 48, 2612.
(29) a) Seo, M. S.; In, J. H.; Kim, S. O.; Oh, N. Y.; Hong, J.; Kim, J.; Que, L.; Nam, W. Angew.
Chem. Int. Ed. 2004, 43, 2417; b) Jensen, M. P.; Lange, S. J.; Mehn, M. P.; Que, E. L.; Que, L. J. Am.
Chem. Soc. 2003, 125, 2113.
(30)
Rev. 2006, 106, 3364; b) Butler, A.; Sandy, M. Nature 2009, 460, 848.
(31) Swamy, P.; Kumar, M. A.; Reddy, M. M.; Naresh, M.; Srujana, K.; Narender, N. RSC
Advances 2014, 4, 26288.
Marasco, E. K.; SchmidtꢀDannert, C. Chembiochem 2008, 9, 1450.
a) Rajagopalan, A.; Lara, M.; Kroutil, W. Adv. Synth. Cat. 2013, 355, 3321; b) Geng, J.; Liu,
a) Bach, R. D.; Dmitrenko, O. J. Am. Chem. Soc. 2006, 128, 1474; b) Bach, R. D.; Dmitrenko,
a) Vaillancourt, F. H.; Yeh, E.; Vosburg, D. A.; GarneauꢀTsodikova, S.; Walsh, C. T. Chem.
(32)
2008, 47, 9937.
Cherrier, M. V.; Cavazza, C.; Bochot, C.; Lemaire, D.; FontecillaꢀCamps, J. C. Biochemistry
§§ The oxygen insertion was proposed on the basis of a previous study (reference 14) and on mass
spectrometry study of O2 activation of Fe-L1 alone in aqueous solution (Figure S18).
19
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 20 of 20
1
2
3
4
Graphic for TOC.
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
20
ACS Paragon Plus Environment