Structure-based enzyme tailoring of 5-hydroxymethylfurfural oxidase

Article abstract of DOI:10.1021/acscatal.5b00031

5-Hydroxymethylfurfural oxidase (HMFO) is a flavin-dependent enzyme that catalyzes the oxidation of many aldehydes, primary alcohols, and thiols.The three-step conversion of 5-hydroxymethylfurfural to 2,5-furandicarboxylic acid is relevant for the industrial production of biobased polymers.The remarkable wide substrate scope of HMFO contrasts with the enzymes precision in positioning the substrate to perform catalysis.We have solved the crystal structure of HMFO at 1.6 ? resolution, which guided mutagenesis experiments to probe the role of the active-site residues in catalysis.Mutations targeting two active-site residues generated engineered forms of HMFO with promising catalytic features, namely enantioselective activities on secondary alcohols and improved 2,5-furandicarboxylic acid yields.

Full text of DOI:10.1021/acscatal.5b00031

Subscriber access provided by COLORADO COLLEGE
Article
Structure-based enzyme tailoring of 5-hydroxymethylfurfural oxidase
Willem Pieter Dijkman, Claudia Binda, Marco W. Fraaije, and Andrea Mattevi
ACS Catal., Just Accepted Manuscript • DOI: 10.1021/acscatal.5b00031 • Publication Date (Web): 09 Feb 2015
Downloaded from http://pubs.acs.org on February 18, 2015
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.
ACS Catalysis 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 25
ACS Catalysis
1
2
3
4
5
6
7
8
9
10
11
1
Structureꢀbased enzyme tailoring of
5ꢀhydroxymethylfurfural oxidase
2
12
13
14
15
16
3
Willem P. Dijkman†,§, Claudia Binda‡,§, Marco W. Fraaije†,*, Andrea Mattevi‡,*
17
18
19
20
21
22
4
†Molecular Enzymology group, Groningen Biomolecular Sciences and Biotechnology Institute,
University of Groningen, Nijenborgh 4, 9747 AG, Groningen, The Netherlands
5
23
24
25
26
6
‡Dept. Biology and Biotechnology, University of Pavia, via Ferrata 1, 27100 Pavia, Italy
27
28
29
7
30
31
32
8
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
ACS Paragon Plus Environment
1

ACS Catalysis
Page 2 of 25
1
2
3
4
5
6
1
2
KEYWORDS:
5ꢀhydroxymethylfurfural oxidase
7
8
9
3
4
5
6
7
8
enzyme
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
2,5ꢀfurandicarboxylic acid
biocatalysis
protein engineering
crystal structure
9
ABSTRACT
10
11
12
13
14
15
16
17
18
5ꢀhydroxymethylfurfural oxidase (HMFO) is a flavinꢀdependent enzyme that catalyzes the
oxidation of many aldehydes, primary alcohols, and thiols. The threeꢀstep conversion of 5ꢀ
hydroxymethylfurfural to 2,5ꢀfurandicarboxylic acid is relevant for the industrial production of
biobased polymers. The remarkable wide substrate scope of HMFO contrasts with the enzyme’s
precision in positioning the substrate to perform catalysis. We have solved the crystal structure
of HMFO at 1.6 Å resolution, which guided mutagenesis experiments to probe the role of the
activeꢀsite residues in catalysis. Mutations targeting two activeꢀsite residues generated
engineered forms of HMFO with promising catalytic features, namely enantioselective activities
on secondary alcohols and improved 2,5ꢀfurandicarboxylic acid yields.
19
ACS Paragon Plus Environment
2

Page 3 of 25
ACS Catalysis
1
2
3
1
4
INTRODUCTION
5
6
7
8
9
2
The enzyme 5ꢀhydroxymethylfurfural oxidase from Methylovorus sp. strain MP688 (HMFO,
EC 1.1.3.47) catalyzes the oxidation of 5ꢀhydroxymethylfurfural (1) to 2,5ꢀfurandicarboxylic
acid (4),1 the latter can be esterified to form polymers with a wide variety of applications.2,3
Because 5ꢀhydroxymethylfurfural (1) can be derived from sugars, the enzyme is of particular
interest for the production of bioꢀbased plastic materials in an environmentallyꢀfriendly industrial
process. The overall reaction consists of three oxidation steps and HMFO can performs all these
steps as depicted in Scheme 1.4 The alcohol group of 5ꢀhydroxymethylfurfural (1) is first
oxidized to the corresponding aldehyde to generate furanꢀ2,5ꢀdicarbaldehyde (2).4 This
compound undergoes spontaneous hydration to the gemꢀdiol which is oxidized by the enzyme to
5ꢀformylꢀ2ꢀfurancarboxylic acid (3). Finally, also this monocarboxylic intermediate product is
released, nonꢀenzymatically hydrated, and oxidized by HMFO to the dicarboxylic reaction
product.4 Importantly, the poor activity of HMFO towards 5ꢀformylꢀ2ꢀfurancarboxylic acid (3) is
limiting the efficiency of the overall process. Increasing the activity towards 5ꢀformylꢀ2ꢀ
furancarboxylic acid (3) is addressed in this study.
3
4
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
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
HMFO is a member of the glucoseꢀmethanolꢀcholine (GMC) family of oxidoreductases.1
HMFO, like other members of this enzyme family, relies on the presence of the flavin adenine
dinucleotide (FAD) cofactor as prosthetic group and an active site histidine (H467 in HMFO)
serving as base. This residue, which is conserved in all GMC enzymes,5–10 initiates the reaction
by removing the proton from the alcohol group of the substrate. The second step, taking place
simultaneously with or after proton abstraction,11–13 is the hydride transfer from the αꢀcarbon of
the substrate to the FAD N5 and thereby completing the oxidation reaction. This results in the
reduced flavin which is reoxidized by molecular oxygen, yielding hydrogen peroxide as byꢀ
ACS Paragon Plus Environment
3

ACS Catalysis
Page 4 of 25
1
2
3
4
5
6
7
8
9
1
2
product (Scheme 1). The active site histidine is crucial for catalysis, functioning as the main
anchoring element for the alcohol group of the substrate.1 The nature of this alcohol groups is
very important as only primary alcohols, primary thiols, and aldehydes are oxidized by HMFO.14
Secondary alcohols are not converted, indicating that the enzyme is very specific concerning the
substrate type.1 On the other hand, HMFO is strikingly promiscuous concerning the side chain
next to the alcohol group. Most (but not all) of the identified substrates are aromatic, including
furanic, phenylic and cinnamylic alcohols. Ring substituents with different size, polarity and
charge are allowed in the active site of HMFO, both on the para and meta positions relative to
the alcohol.1
3
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
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
To investigate how this biocatalyst can be so precise in accepting aldehydes, primary
alcohols and thiols and yet so promiscuous concerning the rest of the substrate, we solved the
HMFO crystal structure. The threeꢀdimensional model provides the framework for protein
engineering studies aimed at the generation of variants with interesting and improved
biocatalytic properties. In particular, several HMFO mutants were designed and produced which
display higher activities towards 5ꢀformylꢀ2ꢀfurancarboxylic acid (3), thereby improving the
enzymatic formation of 2,5ꢀfurandicarboxylic acid (4) from 5ꢀhydroxymethylfurfural (1). In
addition, mutant enzymes were designed which can enantioselectively convert secondary
alcohols, an activity not found in the wild type enzyme.
20
21
MATERIALS AND METHODS
Purification of HMFO
ACS Paragon Plus Environment
4

Page 5 of 25
ACS Catalysis
1
2
3
1
4
For kinetic and spectral analysis, HMFO and its mutants were expressed as a Nꢀterminal
SUMO fusion protein in Escherichia coli BL21 (DE3) and purified using affinity
chromatography as described before.1 For structural characterization similar procedures were
used in the first steps of HMFO purification, including affinity purification that was performed
on a 5 mL HisTrap HP column using an Äkta Purifier system (GE Healthcare). In addition, tag
cleavage was performed by incubating the eluted protein with 0.1 mg of SUMO protease (homeꢀ
made) and dialyzing overnight at 4 °C against 50 mM Tris/HCl pH 7.5, 150 mM NaCl. The
dialyzed sample was loaded again on a 5 mL HisTrap HP column to remove the tag and the
SUMO protease. HMFO was eluted in the flowꢀthrough and then loaded on a Superdex200 16/60
(GE Healthcare) equilibrated with 50 mM potassium phosphate buffer pH 7.6. A single and
sharp peak was obtained. Based on the retention time of HMFO and protein standards of known
size, this is consistent with a monomeric homogeneous form of the protein. Fractions were
pooled and concentrated with Amicon 30K (Millipore), yielding about 30 mg of pure protein
(from 10 g of E. coli cells, wet weight).
5
6
7
8
2
3
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
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
Site directed mutagenesis
To introduce mutations into HMFO, a wholeꢀplasmid PCR was performed. The primers used
to introduce the mutations are listed in Table S1 of the supporting information. The template
DNA was cleaved with DpnI (New England BioLabs) and the PCRꢀproduct was purified
afterwards using a PCR purification kit (Qiagen). E. coli TOP10 cells were transformed with the
plasmids and the introduction of the mutation was confirmed by sequencing.
Product formation by mutant enzymes V367 and W466 and V367R-W466F
ACS Paragon Plus Environment
5

ACS Catalysis
Page 6 of 25
1
2
3
4
5
6
7
8
9
1
2
2,5ꢀFurandicarboxylic acid (4) generation by wild type HMFO and the mutant enzymes
V367K, V367R, W466F and W466A or the double mutant enzyme V367RꢀW466F was assayed
by using 2.0 ꢁM or 20 ꢁM enzyme with 5.0 mM 5ꢀhydroxymethylfurfural (1) or 5ꢀformylꢀ2ꢀ
furancarboxylic acid (3) in a 100 mM potassium phosphate buffer of pH 7.0 (25 °C, 1000 RPM).
The reaction was stopped by heating the mixture at 70 °C. Subsequent centrifugation (5 minutes
at 13000 × g) was used to remove the denatured enzyme. The products formed were analyzed by
HPLC, as described elsewhere.1
3
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
4
5
6
7
8
9
Kinetic analysis
10
11
12
13
14
15
16
17
18
19
Steady state parameters for HMFO wild type and HMFO mutants were obtained by monitoring
the conversion of 4ꢀhydroxyꢀ3ꢀmethoxybenzyl alcohol (vanillyl alcohol). Formation of the
product, 4ꢀhydroxyꢀ3ꢀmethoxybenzaldehyde (vanillin), was measured at 340 nm (ε340 = 14 mMꢀ1
cmꢀ1) using between 0.10 and 4.2 ꢀM enzyme, depending on the HMFO variant, while varying
the substrate concentration between 0.01 and 40 mM. All kinetic experiments were performed at
atmospheric oxygen conditions. The assay was performed in a 50 mM Tris/HCl buffer pH 7.5 at
25 °C. To determine the steady state parameters of HMFO V367R, W466F and V367RꢀW466F
on 5ꢀformylꢀ2ꢀfurancarboxylic acid (3) (TCI Europe), oxygen depletion was measured as
described elsewhere.1 The observed rates were fitted with MichaelisꢀMenten kinetics equation, v
= (kcat (app) ⋅ [S])/((KM (app) + [S]) or with an equation taking substrate inhibition into account, v =
20
21
22
23
(kcat (app) ⋅ [S]/(KM (app) + [S] ⋅ (1 + [S]/Ki (app))). To determine the kcat (app)/KM (app) values for wildꢀ
type and mutant HMFO on 5ꢀformylꢀ2ꢀfurancarboxylic acid (3), the oxidation of 0.056 up to 4.0
mM substrate was followed at 310 nm (ε310 = 2.5 mMꢀ1 cmꢀ1) for 60 minutes in a 50 mM
potassium phosphate buffer of pH 8.0. The slope of the linear relation between the observed rate
ACS Paragon Plus Environment
6

Page 7 of 25
ACS Catalysis
1
2
3
1
4
and the concentration of 5ꢀformylꢀ2ꢀfurancarboxylic acid (3) was taken as the kcat (app)/KM (app)
5
6
7
8
2
3
value.
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
4
Activity towards secondary alcohols
5
The activity of HMFO wild type and mutant enzymes towards the secondary alcohols Sꢀ1ꢀ
phenylethanol and Rꢀ1ꢀphenylethanol (both Sigma Aldrich) was assayed by incubating 5.0 ꢁM
enzyme with 5.0 mM substrate, the reaction was monitored for 50 minutes. Control reactions
without enzyme or without substrate were also performed. Steady state kinetics of the HMFO
W466F and W466A mutants (5 ꢀM) were obtained using between 0.05 to 50 mM Sꢀ1ꢀ
phenylethanol. 1ꢀPhenylethanol conversion was measured by the formation of the product,
acetophenone, at 247 nm (ε247 = 10.8 mMꢀ1 cmꢀ1). Experiments were performed in a 50 mM
potassium phosphate buffer of pH 8.0 at 25 °C at atmospheric oxygen concentrations.
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
Crystallization and structure determination
Pure HMFO (both wild type and H467A mutant, 10 mg/mL in 50 mM potassium phosphate
buffer pH 7.6) was crystallized using the hangingꢀdrop vapor diffusion method by mixing 1.0 µl
of protein with 1.5 µl of a reservoir solution containing 16ꢀ22% w/v PEG3350 and 200 mM
magnesium formate. Yellow flat crystals grew in 2ꢀ5 days at 20 °C. For Xꢀray data collection
crystals were soaked in a cryosolution consisting of 26% w/v PEG3350, 200 mM magnesium
formate, 15% v/v glycerol and flashꢀcooled in liquid nitrogen. For the structural characterization
of enzymeꢀsubstrate complexes, crystals were soaked in cryosolutions containing different
substrates (5ꢀhydroxymethylfurfural, vanillyl alcohol, (2E)ꢀ3ꢀphenylpropꢀ2ꢀenꢀ1ꢀol (cinnamyl
alcohol), furanꢀ2,5ꢀdicarbaldehyde, phenylmethanethiol, (4ꢀnitrophenyl)methanethiol or
ACS Paragon Plus Environment
7

ACS Catalysis
Page 8 of 25
1
2
3
4
5
6
7
8
9
1
2
terephthalaldehyde) using different concentrations depending on compound solubility. Bleaching
of the yellow color of the crystal, indicative for flavin reduction, was monitored under the
microscope followed by flashꢀcooling in liquid nitrogen. In the case of the catalyticallyꢀinactive
H467A mutant enzyme, coꢀcrystallization with 10 mM 5ꢀhydroxymethylfurfural was also carried
out. Xꢀray data collection was performed at the beamlines ID23ꢀEH1 (ESRF, Grenoble, France),
X06SA (SLS, Villigen, Switzerland) and P13 (DESYꢀPETRAIII, Hamburg, Germany). Data
processing and scaling were performed using MOSFLM15 and programs of the CCP4 package.16
The HMFO structure was initially solved by molecular replacement using the program
BALBES17 which selected choline oxidase as the best starting model (PDB code = 4MJW; 31%
sequence identity). Model building and structure analysis were performed by the program
COOT,18 whereas refinement was carried out by REFMAC5.19 Data collection and refinement
statistics are reported in Table S2. Figures were created by CCP4mg.20 Atomic coordinates and
structure factors were deposited with the Protein Data Bank (codes 4UDP, 4UDQ, 4UDR).
3
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
4
5
6
7
8
9
10
11
12
13
14
15
RESULTS AND DISCUSSION
16
17
18
19
20
21
22
Three structures of the enzyme, the wild type in its oxidized and reduced forms and a
mutant enzyme (HMFO H467A), were solved at 1.6ꢀ1.9 Å resolution (Figures 1A and S1; Table
S2). Out of the 531 residues of HMFO, only the four Nꢀterminal and two Cꢀterminal amino acids
lack clear electron density whereas loop 326ꢀ330 is poorly ordered in the oxidized enzyme.
Though being monomeric in solution, the two HMFO molecules present in the asymmetric unit
form a dimer that involves a limited interface area with no catalytically relevant regions (less
than 4% of the total protein surface). Comparison with other proteins of the GMC family
ACS Paragon Plus Environment
8

Page 9 of 25
ACS Catalysis
1
2
revealed that choline oxidase,21 pyridoxineꢀ4ꢀoxidase22 and arylꢀalcohol oxidase23 have the most
similar structural architecture to HMFO (rmsd calculated on the Cα positions is 2.0, 2.1 and 2.2
Å, respectively), followed by glucose oxidase24 (rmsd of 2.3 Å) and pyranoseꢀ2ꢀoxidase25 (rmsd
of 3.0 Å). Choline oxidase and glucose oxidase have dimeric structures, which, unlike HMFO,
are essential for their catalytic activity and involve about 10% of their monomer surface.
The 531 HMFO residues fold into a globular and compact structure organized in two domains:
the FADꢀbinding domain (residues 5ꢀ158, 208ꢀ307, 372ꢀ402, 466ꢀ529), characterized by the
typical Rossmann fold topology that embeds the nonꢀcovalently bound FAD cofactor, and a
smaller cap domain (residues 159ꢀ207, 308ꢀ371, 403ꢀ465) that covers the flavin site. Some
oxidases of the GMC family, like choline oxidase and pyranoseꢀ2ꢀoxidase, contain a covalently
bound FAD cofactor, linked via a 8αꢀN3ꢀhistidyl bond. In HMFO, V101 replaces the histidine
present in these enzymes. Mutagenesis experiments show that introducing a histidine in HMFO
does not result in a covalently bound cofactor and renders the enzyme inactive, possibly because
the imidazole ring interferes with proper positioning of the FAD cofactor (Table S3). Only with
the double mutant enzyme HMFO F67VꢀV101H, designed to leave space for the histidine side
chain, some activity was observed but again, the covalent link was not formed.
3
1
4
5
6
7
8
2
3
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
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
At the interface between the FAD and the cap domains of HMFO, a deep and narrow cleft is
formed which represents the enzyme active site (Figure 1B). The flavin ring and the side chain of
H467 are positioned at the bottom of this cleft. The latter residue is strictly conserved in the
oxidases of the GMC family and was shown to be essential for catalysis by functioning as both
active site base and Hꢀbond acceptor for the substrate OH group.1 As anticipated, the H467A
mutant enzyme is catalytically almost inactive (activity decrease by >3 orders of magnitude, see
Table 1) and the elucidation of its crystal structure showed that the mutation does not result in
ACS Paragon Plus Environment
9

ACS Catalysis
Page 10 of 25
1
2
3
4
5
6
7
8
9
1
2
any conformational change (the overall rmsd for the Cα atoms with respect to the wild type
protein is 0.39 Å) apart from the binding of three additional water molecules that replace the
H467 side chain (Figure S1). Together with H467, N511 forms a hydrogenꢀbond dyad for the
substrate. N511 is a conserved residue in several GMC type oxidases, and removal of the side
chain in the N511A mutant enzyme reduces catalysis, although the effect is much smaller
compared to the H467A mutant enzyme (Table 1). Apart from these two residues, the HMFO
active site is quite hydrophobic (Figure 2). Many attempts were made to obtain the structure of
HMFO in complex with either substrate or product by soaking crystals in a mother liquor
containing a ligand. By monitoring the bleaching of the yellow color of the crystal (indicating the
reduction of the FAD cofactor), we were able to obtain crystals of HMFO in the reduced state.
The elucidated structure of the reduced enzyme was essentially indistinguishable from that of the
oxidized protein. Moreover, in the substrate reduced enzyme crystals, we could not observe
electron density indicating the presence of any ligand. This might be related to the architecture of
the HMFO active site cleft, characterized by its free entrance (Figure 1B). Other GMC oxidases
also feature a narrow channel for substrate binding, but in these enzymes the active site is more
buried and the access to it is often gated by conformational changes of specific loops.22,23,26,27
A combination of modeling and mutagenesis experiments were performed to gain insights into
the role of several residues in substrate binding. On the basis of the constraints imposed by the
position of the CαꢀH oxidation site towards the FAD cofactor and the hydroxyl group interaction
with H467,28 we modeled the 5ꢀhydroxymethylfurfural (1) molecule in the enzyme active center
(Figure 3A). The shape of the cleft constrains the furan ring (as well as the aromatic moiety of
vanillyl alcohol, not shown) in a unique conformation, squeezed by the hydrophobic residues
surrounding the active site (Figure 2). To investigate this aspect, residues M103, W369, F434,
3
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
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
ACS Paragon Plus Environment
10

Page 11 of 25
ACS Catalysis
1
2
3
1
4
V465 and W466 were mutated to alanine and the steady state kinetics of the mutant enzymes
were studied using vanillyl alcohol as a model substrate (Table 1). The most pronounced effects
are observed in the case of the HMFO M103A, W369A, V465A and W466A mutants. These
residues are close to the active site base H467 and their removal creates a large space which
likely affects the proper orientation of the substrate with respect to the flavin. In addition, W466
is in close proximity of the flavin ring (Figure 2) and, consistently, removal of this side chain
turns out to affect protein stability (Tables S4). F434 is further away from the oxidation site and
the effect on catalysis of F434A is smaller compared to the mutant enzymes described above.
The overall picture emerging from these experiments is that the active site funnel is designed to
create a sterically restrained site that positions the reactive primary alcohol of the substrate in the
proper Hꢀbonding environment and geometric relation with the flavin to promote catalysis.
Along this line, we reasoned that some of the created mutant enzymes, though less efficient on
primary alcohols, could instead become active on bulkier secondary alcohols as consequence of
an enlarged the active site cleft. Consistently, the W466F and W466A were found to be active on
Sꢀ1ꢀphenylethanol, with W466A performing best (Table 2). Remarkably, the W466 mutant
enzymes are strictly enantioselective as they have no activity on R-1ꢀphenylethanol (Table S5),
similar to the results for the analogues mutation in the fungal arylꢀalcohol oxidase.29 These data
are in full agreement with the structural modelling of 1ꢀphenylethanol in HMFO because the Rꢀ
enantiomer would not bind with the Cα hydrogen pointing towards the flavin N5 as requested for
oxidation. The Sꢀenantiomer, on the other hand, would have its methyl group colliding with the
W466 side chain, a problem which is overcome by the W466F or W466A mutations (Figure 3B).
The rate limiting step in the HMFO catalyzed oxidation of 5ꢀhydroxymethylfurfural (1) to 2,5ꢀ
furandicarboxylic acid (4) is the oxidation of 5ꢀformylꢀ2ꢀfurancarboxylic acid (3) (Scheme 1).4
5
6
7
8
2
3
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
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
ACS Paragon Plus Environment
11

ACS Catalysis
Page 12 of 25
1
2
3
4
5
6
7
8
9
1
2
Therefore, increasing the rate of catalysis for this last oxidation step would lead to higher product
yields. These thoughts led us to analyze the activity of some of the mutant enzymes towards
aldehyde substrates, which have to be in their gemꢀdiol form in order to be converted by HMFO.
Indeed, both secondary alcohols and gemꢀdiols have bulky substituents on the αꢀcarbon
compared to primary alcohols. Therefore, although less active on primary alcohols (Table S6),
the W466F and W466A mutants were tested against the aldehyde 5ꢀformylꢀ2ꢀfurancarboxylic
acid (3), both showing increased activity with a noticeable conversion of 49% of 5ꢀformylꢀ2ꢀ
furancarboxylic acid (3) into 2,5ꢀfurandicarboxylic acid (4) in 24 hours (three times more than
the wild type enzyme; Table S6). With these studies W466 was identified as a valuable target
site for mutants featuring improved activity on aldehyde substrates by creating a more spacious
binding cleft for a gemꢀdiol substrate.
3
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
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
Both furanic and phenylic substrates with a negatively charged carboxylic acid on the para
position are much less efficiently converted than their nonꢀcharged analogues.1,4 This poor
activity of wild type HMFO towards for instance 5ꢀformylꢀ2ꢀfurancarboxylic acid (3) is likely
due to a high KM (app) value (Table 2). Similar results were described recently for arylꢀalcohol
oxidase. This enzyme cannot perform the three oxidations required to form 2,5ꢀfurandicarboxylic
acid (4) from 5ꢀhydroxymethylfurfural (1), as it is not active towards 5ꢀformylꢀ2ꢀfurancarboxylic
acid (3).30 To improve activity of HMFO for 5ꢀformylꢀ2ꢀfurancarboxylic acid (3), V367
represents a candidate for mutagenesis (see Figure 3A). Two HMFO mutants, V367K and
V367R, were produced and characterized. It was gratifying to observe that both enzyme variants
showed significantly improved activity towards 5ꢀformylꢀ2ꢀfurancarboxylic acid (3) (Table S6).
In particular, V367R is able to form 97% of 2,5ꢀfurandicarboxylic acid (4) in 6 hours, whereas
the wild type enzyme needs 24 hours to reach 92% conversion. This can also be seen from the
ACS Paragon Plus Environment
12

Page 13 of 25
ACS Catalysis
1
2
3
1
4
amounts of catalyst needed to obtain 50% conversion: the V367R mutant enzyme outperforms
the wild type protein by reaching similar product quantities in 12 hours with 10 times less
biocatalyst (Table S6). Similar results are obtained when the entire reaction is tested, i.e. using 5ꢀ
hydroxymethylfurfural (1) as initial substrate (Table S6). Clearly, HMFO V367R results in a
better 5ꢀhydroxymethylfurfural (1) to 2,5ꢀfurandicarboxylic acid (4) conversion (~3 fold higher).
These data demonstrate that the introduction of a positive charge at this side of the active site
increases the activity towards 5ꢀformylꢀ2ꢀfurancarboxylic acid (3) by helping to position the
substrate.
5
6
7
8
2
3
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
4
5
6
7
8
9
These mutagenesis experiments identified W466 as site for mutations increasing activities on
the aldehyde 5ꢀformylꢀ2ꢀfurancarboxylic acid (3) and V367 as a site for improving the activity
on carboxylic acidꢀcontaining substrates. A logical continuation was to evaluate the effect of the
combined mutations. The V367RꢀW466F double mutant was constructed and tested for its
activity towards 5ꢀformylꢀ2ꢀfurancarboxylic acid (3). The double mutant enzymes has an almost
tenfold higher kcat (app) when compared to the individual single mutant enzymes and also the
lowest KM (app) value (Table 2), resulting in a catalytic efficiency (kcat/KM) for 5ꢀformylꢀ2ꢀ
furancarboxylic acid (3) which is over 1000 fold higher than wild type HMFO. HPLC analysis
confirms the double mutant is superior to both the single mutants and wild type HMFO for the
formation of 2,5ꢀfurandicarboxylic acid (4) (Table S6). Moreover, the double mutant displayed
enhanced efficiency in the overall conversion 5ꢀhydroxymethylfurfural (1) to 2,5ꢀ
furandicarboxylic acid (4), a feature of selfꢀevident relevance for the biotechnological
applications of the enzyme.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
CONCLUSIONS
ACS Paragon Plus Environment
13

ACS Catalysis
Page 14 of 25
1
2
3
4
5
6
7
8
9
1
2
An intriguing aspect of HMFO is its dual character of being very precise in the oxidation of
primary alcohols (i.e. devoid of any activity on secondary alcohols) and at the same time so
promiscuous in converting a broad range of primary alcohols. At the heart of this feature is an
activeꢀsite cleft which provides an essential Hꢀbonding anchoring point that activates the
substrate and positions the substrate αꢀcarbon in direct contact with the flavin. The crystal
structure shows that a constellation of mostly hydrophobic amino acids shapes the active site clef
to make it selective for aromatic and aliphatic primary alcohols. This active site architecture is
amenable to protein engineering to design biocatalysts with enhanced activities towards
substrates with varying bulkiness and electrostatic properties. In particular, it was possible to
identify mutant enzyms that are active on secondary alcohols, with the added value that their
enantioselectivity can be used for kinetic resolution of chiral secondary alcohols.31–33 Equally
valuable, mutant enzymes were created with increased activity towards the intermediates formed
during the multistep HMFOꢀmediated production of 2,5ꢀfurandicarboxylic acid (4). These
properties culminate in a double mutant that combines the best of these individual mutant
enzymes, having the highest turnover number and affinity for 5ꢀformylꢀ2ꢀfurancarboxylic acid
(3), leading to the highest 2,5ꢀfurandicarboxylic acid (4) yield. All these findings will guide
further investigations on HMFO catalytic activities and will be valuable for industries focused on
green chemistry approaches for bioꢀbased plastics production.
3
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
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
ACS Paragon Plus Environment
14

Page 15 of 25
ACS Catalysis
1
2
3
1
4
FIGURES
5
6
7
8
9
2
3
Figure 1. The threeꢀdimensional structure of HMFO.
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
A
4
5
B
6
7
8
9
10
11
12
13
14
15
(A) Ribbon diagram of the overall structure with its FAD cofactor bound (represented with
carbon, oxygen, nitrogen and phosphor atoms in yellow, red, blue and magenta). (B) HMFO
surface highlighting the active site cleft that is about 20 Å deep and on average 10 Å large (the
representation was frontꢀ and backꢀclipped to make the cleft space visible). In front of the flavin
ring the side chain of the conserved His467 is at the bottom of the active site.
ACS Paragon Plus Environment
15

ACS Catalysis
Page 16 of 25
1
2
3
4
5
1
2
Figure 2. Stereo view of the active site of HMFO (reduced state).
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
3
4
Residues surrounding the space in front of the flavin ring are shown, including three water
molecules (red spheres) that are conserved in all determined HMFO crystal structures (only one
is missing in the oxidized enzyme, which may be related to the lower resolution). Color coding is
as in Figure 1 (sulfur atoms in green). Hydrogen bonds are represented as dashed lines.
5
6
7
8
9
ACS Paragon Plus Environment
16

Page 17 of 25
ACS Catalysis
1
2
3
1
4
Figure 3. Modeling of substrate binding in HMFO active site.
5
6
7
2
A
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
3
24
25
4
B
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
5
6
7
8
9
10
11
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
12
13
14
15
16
17
(A) Proposed model for the binding of 5ꢀhydroxymethylfurfural substrate (1). The furan ring
orientation is constrained by the narrow shape of the cleft and the substrate αꢀcarbon is in the
proper position to be oxidized by the flavin.28 The protein atom closest to the substrate hydroxyl
group is Nε2 of His467, which is perfectly positioned to form a Hꢀbond with the substrate. The
orientation of the His467 imidazole ring was assigned on the basis of a favorable Hꢀbond
interaction between its Nδ1 atom and the neighboring His307 side chain (not shown for clarity).
ACS Paragon Plus Environment
17

ACS Catalysis
Page 18 of 25
1
2
3
4
5
6
7
8
9
1
2
3
4
5
6
7
The arginine side chain of the V367R mutant was modeled to show the possible interaction with
the substrate, which would provide an explanation for the increased activity of this mutant with
5ꢀformylꢀ2ꢀfurancarboxylic acid (3) (Table 2). (B) Similarly to 5ꢀhydroxymethylfurfural (1), the
secondary alcohol Sꢀ1ꢀphenylethanol was modeled in HMFO active site. The model is consistent
with the biochemical data showing that wild type HMFO is not able to oxidize this substrate
because the substrate methyl group would collide with W466 side chain. According to the model,
the smallest distance between the methyl group of Sꢀ1ꢀphenylethanol and W466 is 2.3 Å.
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
8
ACS Paragon Plus Environment
18

Page 19 of 25
ACS Catalysis
1
2
3
1
4
SCHEMES
5
6
7
8
9
2
3
Scheme 1. The oxidation of 5ꢀhydroxymethylfurfural (1) to 2,5ꢀfurandicarboxylic acid (4) by
HMFO.
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
4
5
Conversion of 1 to 4 is initiated by oxidation of the alcohol group of 1, resulting in furanꢀ2,5ꢀ
dicarbaldehyde (2). The hydrated form of 2 is converted to 5ꢀformylꢀ2ꢀfurancarboxylic acid (3).
The last step is the oxidation of the hydrated form of 3, yielding 4.
6
7
8
9
TABLES.
Table 1. Steady state parameters of the different HMFO variants on vanillyl alcohol a
Enzyme
kcat (app) (sꢀ1)
KM (app) (mM)
kcat (app) /KM (app)
(sꢀ1 mMꢀ1)
29
Wild type
M103A
W369A
F434A
21 ± 0.42
0.72 ± 0.066
1.8 ± 0.11
3.3 ± 0.051
4.3 ± 0.0063
15 ± 0.68
1.8
1.1 ± 0.067
1.2 ± 0.23
3.9
12
V465A
V367K
V367R
W466F
W466A
H467A b
N511A
1.5 ± 0.058
5.1 ± 0.11
0.92 ± 0.16
0.29 ± 0.032
0.28 ± 0.027
0.20 ± 0.027
1.0 ± 0.15
1.6
18
6.0 ± 0.12
21
1.7 ± 0.043
0.75 ± 0.025
0.0047 ± 0.00023
1.0 ± 0.014
8.5
0.75
0.0057
0.55
0.82 ± 0.18
2.0 ± 0.10
ACS Paragon Plus Environment
19

ACS Catalysis
Page 20 of 25
1
2
3
4
5
1
2
a Experiments performed at 25 °C, in 50 mM Tris/HCl buffer of pH 7.5 at atmospheric oxygen
concentration. Product formation was measured at 340 nm.
6
7
8
9
3
4
5
b Data published elsewhere.1
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 2. Steady state parameters of the engineered HMFO variantsa
Enzyme
kcat (app) (sꢀ1)
KM (app) (mM)
kcat (app) /KM (app)
(sꢀ1 mMꢀ1)
0.0005
wild type
V367Rc
n.s.b
> 4.0
1.4
0.056
0.079
0.46
0.041
W466F
3.0
0.026
V367RꢀW466Fc
wild type
W466A
0.21
n.d.
20
2.2
n.d.d
n.d.
0.55 ⋅ 10ꢀ3
0.10 ⋅ 10ꢀ3
0.011
W466F
0.010
98
a
6
7
8
9
The activity of HMFO variants towards 5ꢀformylꢀ2ꢀfurancarboxylic acid (3) was measured
using O2 depletion whereas the activity towards Sꢀ1ꢀphenylethanol was measured by the
formation of product at 247 nm (ε247 = 10.8 mMꢀ1 cmꢀ1) in a 50 mM potassium phosphate buffer
of pH 8.0 at 25 °C at atmospheric oxygen concentration.
b
10
11
12
13
14
15
n.s. = no saturation. Activity of HMFO at saturated substrate conditions could not be
measured, preveting estimation of the kcat (app). The kcat (app) /KM (app) value for wild type HMFO
was determined by measuring the decrease of 5ꢀformylꢀ2ꢀfurancarboxylic acid (3) (ε310 = 3.7
mMꢀ1 cmꢀ1) at low concentration of 5ꢀformylꢀ2ꢀfurancarboxylic acid (3) (≤ 4.0 mM), where the
linear relation between the rates (sꢀ1) and the substrate concentration (mM) represents the kcat
(app)/KM (app)
.
c
16
17
Both HMFO V367R and V367RꢀW466F show substrate inhibition, with a Ki (app) of 76 mM
and 7.3 mM respectively.
d
18
19
n.d. = not detectable. No activity can be detected towards Sꢀ1ꢀphenylethanol using wild type
HMFO.
20
ACS Paragon Plus Environment
20

Page 21 of 25
ACS Catalysis
1
2
3
1
4
ASSOCIATED CONTENT
5
6
7
8
9
2
3
4
5
6
7
8
Supporting Information. Primers used for the generation of the mutant enzymes (Table S1),
electron density showing the H467A mutation in the crystal structure (Figure S1), Data
collection and refinement statistics for HMFO, oxidized and reduced, and HMFO H467A (Table
S2), steady state kinetics of HMFO V101H mutants (Table S3), spectral properties and apparent
melting temperatures of all HMFO mutants (Table S4), activity towards secondary alcohols
(Table S5), data on 5ꢀhydroxymethylfurfural (1) and 5ꢀformylꢀ2ꢀfurancarboxylic acid (3)
conversion using different experimental conditions and HMFO mutant enzymes (Table S6).
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
9
AUTHOR INFORMATION
10
11
Corresponding Author
*Eꢀmail: m.w.fraaije@rug.nl; andrea.mattevi@unipv.it
12
Author Contributions
13
14
The manuscript was written through contributions of all authors. All authors have given approval
to the final version of the manuscript. §WPD and CB contributed equally.
15
Funding Sources
16
17
This work was carried out within the BEꢀBasic R&D Program, for which an FES subsidy was
granted from the Dutch Ministry of Economic Affairs, Agriculture, and Innovation (EL&I).
18
19
20
ACKNOWLEDGMENT
We thank the European Synchrotron Radiation Facility (ESRF), the Swiss Light Source (SLS)
and the Deutsches ElektronenꢀSynchrotron (DESYꢀPETRAIII) for providing beamtime and
ACS Paragon Plus Environment
21

ACS Catalysis
Page 22 of 25
1
2
3
4
5
6
1
2
assistance. We thank also the BioStructꢀX program (project n. 5275) for funding synchrotron
trips.
7
8
9
3
4
ABBREVIATIONS
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
HMFO, 5ꢀhydroxymethylfurfural oxidase; GMC, glucoseꢀmethanolꢀcholine; FAD, flavin
5
6
adenine dinucleotide.
REFERENCES
7
(1)
(2)
Dijkman, W. P.; Fraaije, M. W. Appl. Environ. Microbiol. 2014, 80, 1082–1090.
8
9
Jong, E. de; Dam, M. A.; Sipos G J; Gruter, M. In Biobased Monomers, Polymers, and
Materials; American Chemical Society, 2012; 1–13.
10
(3)
(4)
Kamm, B. Angew. Chem. Int. Ed. Engl. 2007, 46, 5056–5058.
11
12
Dijkman, W. P.; Groothuis, D. E.; Fraaije, M. W. Angew. Chem. Int. Ed. Engl. 2014, 53,
6515–6518.
13
(5)
(6)
Romero, E.; Gadda, G. Biomol. Concepts 2014, 5, 299–318.
14
15
Smitherman, C. L.; Rungsrisuriyachai, K.; Germann, M. W.; Gadda, G. Biochemistry
2014.
16
17
(7)
Mugo, A. N.; Kobayashi, J.; Yamasaki, T.; Mikami, B.; Ohnishi, K.; Yoshikane, Y.; Yagi,
T. Biochim. Biophys. Acta 2013, 1834, 953–963.
18
19
(8)
(9)
Yue, Q. K.; Kass, I. J.; Sampson, N. S.; Vrielink, A. Biochemistry 1999, 38, 4277–4286.
Wongnate, T.; Sucharitakul, J.; Chaiyen, P. Chembiochem 2011, 12, 2577–2586.
20
21
(10) HernándezꢀOrtega, A.; Lucas, F.; Ferreira, P.; Medina, M.; Guallar, V.; Martínez, A. T.
Biochemistry 2012, 51, 6595–6608.
22
(11) Fan, F.; Gadda, G. J. Am. Chem. Soc. 2005, 127, 2067–2074.
23
24
(12) HernándezꢀOrtega, A.; Borrelli, K.; Ferreira, P.; Medina, M.; Martínez, A. T.; Guallar, V.
Biochem. J. 2011, 436, 341–350.
25
(13) Wongnate, T.; Chaiyen, P. FEBS J. 2013, 280, 3009–3027.
ACS Paragon Plus Environment
22

Page 23 of 25
ACS Catalysis
1
2
3
1
4
5
(14) Ewing, T. A.; Dijkman, W. P.; Vervoort, J. M.; Fraaije, M. W.; van Berkel, W. J. H.
2
Angew. Chem. Int. Ed. Engl. 2014, 53, 13206–13209.
6
7
8
9
3
4
(15) Battye, T. G. G.; Kontogiannis, L.; Johnson, O.; Powell, H. R.; Leslie, A. G. W. Acta
Crystallogr. D. Biol. Crystallogr. 2011, 67, 271–281.
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
5
6
7
8
(16) Winn, M. D.; Ballard, C. C.; Cowtan, K. D.; Dodson, E. J.; Emsley, P.; Evans, P. R.;
Keegan, R. M.; Krissinel, E. B.; Leslie, A. G. W.; McCoy, A.; McNicholas, S. J.;
Murshudov, G. N.; Pannu, N. S.; Potterton, E. a; Powell, H. R.; Read, R. J.; Vagin, A.;
Wilson, K. S. Acta Crystallogr. D. Biol. Crystallogr. 2011, 67, 235–242.
9
10
(17) Long, F.; Vagin, A. a; Young, P.; Murshudov, G. N. Acta Crystallogr. D. Biol.
Crystallogr. 2008, 64, 125–132.
11
(18) Emsley, P.; Cowtan, K. D. Acta Crystallogr. D. Biol. Crystallogr. 2004, 60, 2126–2132.
12
13
(19) Murshudov, G. N.; Vagin, a a; Dodson, E. J. Acta Crystallogr. D. Biol. Crystallogr. 1997,
53, 240–255.
14
15
(20) McNicholas, S.; Potterton, E.; Wilson, K. S.; Noble, M. E. M. Acta Crystallogr. D. Biol.
Crystallogr. 2011, 67, 386–394.
16
(21) Quaye, O.; Cowins, S.; Gadda, G. J. Biol. Chem. 2009, 284, 16990–16997.
17
18
(22) Mugo, A. N.; Kobayashi, J.; Mikami, B.; Ohnishi, K.; Yagi, T. Acta Crystallogr. Sect. F.
Struct. Biol. Cryst. Commun. 2012, 68, 66–68.
19
20
(23) Fernández, I. S.; RuízꢀDueñas, F. J.; Santillana, E.; Ferreira, P.; Martínez, M. J.; Martínez,
A. T.; Romero, A. Acta Crystallogr. D. Biol. Crystallogr. 2009, 65, 1196–1205.
21
22
(24) Hecht, H.ꢀJ.; Kalisz, H. M.; Hendle, J.; Schmid, R. D.; Schomburg, D. J. Mol. Biol. 1993,
229, 153–172.
23
(25) Hallberg, B. M.; Leitner, C.; Haltrich, D.; Divne, C. J. Mol. Biol. 2004, 341, 781–796.
24
25
(26) Spadiut, O.; Tan, T.ꢀC.; Pisanelli, I.; Haltrich, D.; Divne, C. FEBS J. 2010, 277, 2892–
2909.
26
27
(27) Quaye, O.; Lountos, G. T.; Fan, F.; Orville, A. M.; Gadda, G. Biochemistry 2008, 47,
243–256.
28
(28) Fraaije, M. W.; Mattevi, a. Trends Biochem. Sci. 2000, 25, 126–132.
29
30
(29) HernándezꢀOrtega, A.; Ferreira, P.; Merino, P.; Medina, M.; Guallar, V.; Martínez, A. T.
Chembiochem 2012, 13, 427–435.
ACS Paragon Plus Environment
23

ACS Catalysis
Page 24 of 25
1
2
3
4
5
6
1
2
3
(30) Carro, J.; Ferreira, P.; Rodríguez, L.; Prieto, A.; Serrano, A.; Balcells, B.; Ardá, A.;
JiménezꢀBarbero, J.; Gutiérrez, A.; Ullrich, R.; Hofrichter, M.; Martínez, A. T. FEBS J.
2014, 10.1111/febs.13177.
7
8
9
4
(31) Turner, N. J. Chem. Rev. 2011, 111, 4073–4087.
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
5
6
(32) O’Reilly, E.; Iglesias, C.; Ghislieri, D.; Hopwood, J.; Galman, J. L.; Lloyd, R. C.; Turner,
N. J. Angew. Chem. Int. Ed. Engl. 2014, 53, 2447–2450.
7
8
9
(33) Escalettes, F.; Turner, N. J. Chembiochem 2008, 9, 857–860.
ACS Paragon Plus Environment
24

Page 25 of 25
ACS Catalysis
1
2
3
1
4
Table of contents and abstract graphics
5
6
2
7
8
9
10
11
12
13
14
15
16
17
3
18
19
20
21
4
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
5
ACS Paragon Plus Environment
25