Molecular-Level Insight into the Differential Oxidase and Oxygenase Reactivities of de Novo Due Ferri Proteins

Article abstract of DOI:10.1021/jacs.5b03524

Using the single-chain due ferri (DFsc) peptide scaffold, the differential oxidase and oxygenase reactivities of two 4A→4G variants, one with two histidines at the diiron center (G4DFsc) and the other with three histidines (3His-G4DFsc(Mut3)), are explore

Full text of DOI:10.1021/jacs.5b03524

Subscriber access provided by UNIV OF MISSISSIPPI
Article
Molecular level insight into the differential oxidase and
oxygenase reactivities of de novo Due Ferri proteins
Rae Ana Snyder, Susan E. Butch, Amanda J. Reig, William F. DeGrado, and Edward I. Solomon
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.5b03524 • Publication Date (Web): 19 Jun 2015
Downloaded from http://pubs.acs.org on June 24, 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.
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 47
Journal of the American Chemical Society
1
2
3
4
5
Molecular level insight into the differential oxidase and oxygenase reactivities
6
7
of de novo Due Ferri proteins
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
Rae Ana Snyder,† Susan E. Butch, ‡ Amanda J. Reig, ‡,* William F. DeGrado,¥,* and Edward I.
Solomon†,§,*
Department of Chemistry, Stanford University, Stanford, California 94305, Department of
Chemistry, Ursinus College, Collegeville, Pennsylvania 19426, Department of Pharmaceutical
Chemistry, University of California San Francisco, San Francisco, California 94143 and
Stanford Synchrotron Radiation Laboratory, Stanford University, SLAC, Menlo Park, California
94025
Abstract: Using the DFsc (single-chain due ferri peptide) scaffold, the differential oxidase and
oxygenase reactivities of two 4A→ 4G variants, one with 2 histidines at the diiron center
(G4DFsc) and the other with 3 histidines (3His-G4DFsc(Mut3)), are explored. By controlling the
reaction conditions, the active form responsible for 4-aminophenol oxidase activity in both
G4DFsc and 3His-G4DFsc(Mut3) is determined to be the substrate bound biferrous site. Using
CD, MCD, and VTVH MCD spectroscopies, 4-aminophenol is found to bind directly to the
biferrous sites of the DF proteins. In G4DFsc, 4-aminophenol increases the coordination of the
biferrous site, while in 3His-G4DFsc(Mut3), the coordination number remains the same and the
substrate likely replaces the additional bound histidine. This substrate binding enables a 2-
electron process where 4-aminophenol is oxidized to benzoquinone imine and O2 is reduced to
H2O2. In contrast, only the biferrous 3His variant is found to be active in the oxygenation of p-
anisidine to 4-nitroso-methoxybenzene. From CD, MCD and VTVH MCD, p-anisidine addition
is found to minimally perturb the biferrous centers of both G4DFsc and 3His-G4DFsc(Mut3),
indicating that this substrate binds near the biferrous site. In 3His-G4DFsc(Mut3), the
coordinative saturation of one iron leads to the 2-electron reduction of O2 at the second iron to
generate an end-on hydroperoxo-Fe(III) active oxygenating species.
† Department of Chemistry, Stanford University
‡ Department of Chemistry, Ursinus College
¥ Department of Pharmaceutical Chemistry, University of California San Francisco
§ Stanford Synchrotron Radiation Laboratory, Stanford University.
* To whom correspondence should be addressed. A.J.R.: e-mail, areig@ursinus.edu; phone, (610) 409-3383;
W.F.D.: e-mail, Bill.DeGrado@ucsf.edu; phone, (415) 476-9679; E.I.S.: e-mail, edward.solomon@stanford.edu;
phone, (650) 723-9104.
1
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 47
1
2
3
4
Introduction
5
6
7
8
9
Binuclear non-heme iron enzymes catalyze a variety of biologically significant reactions with
relevance to biotechnology, 1,2 environmental remediation, 3 and drug design. 4,5 The catalyzed
reactions are usually O2 dependent and generally include oxidation and oxygenation of
substrates. 5-8 Representative binuclear non-heme iron enzymes that function as oxidases include
alternative oxidase (AOX) and flavin diiron protein (FDP). 9-14 AOX has a diiron center that
participates in the mitochondrial electron transport chain in plants and bacteria, catalyzing
ubiquinol oxidation with O2 (which is reduced to H2O2) and allowing for cyanide resistant
respiration. 14 Flavin diiron proteins are oxidases that protect anaerobes from oxidative stress. 9-13
Among the many binuclear non-heme iron enzymes that function as oxygenases are methane
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
monooxygenase (MMO, methane to methanol), 15-18 toluene-4-monooxygenase (T4MO, toluene
19
to methylphenol),
nitrobenzoate) 20,21
and p-aminobenzoate N-oxygenase (AurF, p-aminobenzoate to p-
.
The diiron active sites of binuclear non-heme iron enzymes are coordinated to the protein
by histidine and carboxylate residues and are often contained within common 4-helix bundle
motifs (Figure 1). 5-8,15,16,22-25 While a two-histidine and four-carboxylate coordination of the
diiron center (two of the carboxylates bridge the ferrous centers) is observed in many cases, such
15-17
5,25
23,24
as in MMO,
ribonucleotide reductase
and delta-9 desaturase,
the number of
coordinating histidine and carboxylate residues can vary. 6,20,26 An additional histidine binds to
one of the iron sites (3His/4-carboxylate ligand set) in AurF and Myo-inositol oxygenase
(MIOX) has a diiron center with a four-histidine and two-carboxylate coordination. 20,27-31 Many
binuclear non-heme iron enzymes react with O2 in their biferrous states to form activated
2
ACS Paragon Plus Environment

Page 3 of 47
Journal of the American Chemical Society
1
2
3
4
intermediates. 5-8,32-39 MMO reacts with O2 to form a high valent intermediate (Q) that acts as the
5
6
18,40-42
oxygenating species.
Alternatively in AurF, a peroxy-intermediate species has been
7
8
9
proposed as the N-oxygenating species. 29,38
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: Crystal structure of ribonucleotide reductase (left) (PDB 1RIB) 43 with the 4-helix bundle motif
bolded and the NMR structure of Zn substituted DFsc (right) (PDB 2HZ8) 44. A close up of the active site
with 2-histidines+4-carboxylate coordination is shown in the bottom right. The 4A shown in red of the
NMR structure of DFsc are mutated to 4G (A10G, A14G, A43G, A47G) in G4DFsc.
Due ferri (DF) de novo proteins, peptides that self-assemble as 4-helix bundle units
(Figure 1, right) with binuclear non-heme iron centers, are used to model native binuclear non-
heme iron enzymes. They offer the simplicity of small molecules while still maintaining the
native protein environment. 45-47 The single chain DF peptides (DFsc), a subset of DF, form
scaffolds that mimic the folding properties and asymmetry of native diiron proteins. 48,49 The
original DFsc variant, which was constructed to contain the common 2-His/4-carboxylate ligand
set (found in ribonucleotide reductase, MMO, delta-9 desaturase, etc), displayed O2 reaction
rates similar to natural diiron proteins. 50 In order to increase solvent and substrate accessibility, a
4A→4G variant (G4DFsc, DFsc with mutations = A10G, A14G, A43G, A47G) was made. 51
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 47
1
2
3
4
5
6
7
8
9
While still retaining the 2-His/4-carboxylate ligand set, the new variant displayed a decrease in
the O2 reaction rate. 52 Additionally, a version of G4DFsc where three (rather than two) histidine
residues are near the diiron center (3His-G4DFsc(Mut3)) was constructed to model the ligation
of AurF. 51 In order to incorporate the 3rd His residue at the diiron site (I100H), three additional
mutations (Mut3 = Y18L, I37N, and L81H) were required for protein stability.
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
The perturbation of G4DFsc to incorporate an additional histidine at the diiron active site
was found to significantly alter its reactivity. 51 Our previous studies evaluated arylamine oxidase
and oxygenase activities of G4DFsc and 3His-G4DFsc(Mut3) using 4-aminophenol and p-
anisidine as substrates. While G4DFsc and 3His-G4DFsc(Mut3) were shown to have comparable
O2 reaction rates, 52 the activity studies revealed that G4DFsc catalyzed the oxidation of 4-
51
aminophenol, while the 3His form only had minimal activity.
The 3His form did, however,
show oxygenase reactivity towards the amine group of p-anisidine, while G4DFsc appeared
inactive in this reaction. Thus, 3His-G4DFsc(Mut3) was demonstrated to be a functional model
of AurF.
The present study elucidates how incorporation of an additional histidine at the diiron
center modifies the reactivities of G4DFsc. These investigations define the active species
involved and give insight into the differential oxidase and oxygenase activities of 2His (G4DFsc)
and 3His (3His-G4DFsc(Mut3)) forms of diiron proteins. The structures and reactivities of the
substrate bound biferrous species are probed using near IR (NIR) circular dichroism (CD),
magnetic circular dichroism (MCD) and variable temperature variable field (VTVH) MCD
spectroscopies. NIR CD and MCD are used to probe the ferrous d → d transitions, which
correlate to the geometry of each Fe(II) center. 6 VTVH MCD uses these excited state transitions
to study the ground state and extract information on the magnetic coupling between the two iron
4
ACS Paragon Plus Environment

Page 5 of 47
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
sites, which correlates with the bridging ligands. This also provides the zero-field splitting (ZFS)
of the individual iron centers, which compliments the geometric information obtained from CD
and MCD spectroscopies. Based on these studies, we elucidate how the properties of substrate
binding to the biferrous sites define the differences in reactivities observed between the 2His
(G4DFsc) and the 3His (3His-G4DFsc(Mut3)) forms of de novo designed binuclear non-heme
iron proteins.
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
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 47
1
2
3
4
5
6
7
8
9
Results and Analysis.
Oxidase Activity.
Reactivity.
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
H2N
NH
NH2
+ 2
NH2
NH2
3
N
+ 1
NH
HO
O
HO
4
2
2H+, 2e-
1
Figure 2: Two-electron oxidation of 4-aminophenol followed by coupling with m-phenylenediamine to
form an aminoindoaniline dye.
Based on our prior studies, the biferrous G4DFsc reacts with O2 to first form a paramagnetic
species (attributed to a biferric site with only µ-1,3 carboxylate bridges, Species B) that decays
into a diamagnetic species (attributed to a hydroxo-bridged antiferromagnetically coupled
biferric site, Species C). 52 However, 3His-G4DFsc(Mut3) only formed the biferric paramagnetic
species B due to the additional His coordinatively saturating one iron center. In our prior studies
that showed oxidase reactivity for G4DFsc, but not for the 3His form, Fe(II) was added under
aerobic conditions to a solution of apo-protein and 4-aminophenol (1) in the presence of m-
phenylenediamine (3), which couples to the oxidized product (quinone imine) (2) to yield an
aminoindoaniline dye (4) with λmax = 486 nm (ε = 13,200 cm-1 M-1) at pH 7 (Figure 2). 51 The
conditions of the prior studies, which allowed for the simultaneous presence of the biferric
52
(Species B and C)
and biferrous forms and may have included contributions from Fe(II)
binding to the protein, complicated our ability to identify the catalytically active species. Using
51
more controlled reaction conditions, the prior studies were thus extended to identify the
reactive species of G4DFsc responsible for catalyzing the oxidation of 4-aminophenol, as well as
address the apparent lack of activity in the 3His form.
6
ACS Paragon Plus Environment

Page 7 of 47
Journal of the American Chemical Society
1
2
3
4
(A)
(B)
5
G4DFsc
3His-G4DFsc(Mut3)
6
4-AP anaerobic addition
4-AP anaerobic addition
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
(C)
(D)
G4DFsc
4-AP and O2 sat. buffer
added simultaneously
3His-G4DFsc(Mut3)
4-AP and O2 sat. buffer
added simultaneously
(E)
(F)
(G)
G4DFsc
Species B
G4DFsc
Species C
4-AP aerobic addition
3His-G4DFsc(Mut3)
Species B
4-AP aerobic addition
4-AP aerobic addition
Figure 3: The oxidation of 4-aminophenol (4-AP) for various reaction conditions over 10 min. The
reaction conditions include: (1) the addition of O2 saturated buffer to biferrous G4DFsc (A) and 3His-
G4DFsc(Mut3) (B) with 4-AP (in the presence of m-phenylenediamine) already loaded; (2) the
simultaneous addition of O2 saturated buffer and 4-AP (in the presence of m-phenylenediamine) to
biferrous G4DFsc (C) and 3His-G4DFsc(Mut3) (D); (3) the addition of 4-AP (in the presence of m-
phenylenediamine) to the biferric paramagnetic Species B (E) and diamagnetic Species C (F) for G4DFsc
and the biferric diamagnetic Species B for 3His-G4DFsc(Mut3) (G). Aminoindoaniline dye formation
was used to identify the oxidation of 4-AP (see Figure 2). Final concentrations (after O2 saturated buffer
addition) were 10 µM for protein, 3.9 mM 4-aminophenol, and 10 mM m-phenylenediamine.
7
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 47
1
2
3
4
5
6
7
8
9
To test the biferrous forms of G4DFsc and 3His-G4DFsc(Mut3), Fe(II) was added to the
apo-protein in an anaerobic environment (glove box) to exclude possible iron binding effects that
may complicate the reactivity. 4-aminophenol was also added to the Fe(II) protein under
anaerobic conditions and incubated for 15 min. This was followed by the anaerobic addition of
m-phenylenediame, and then, O2 saturated buffer at 4 ºC. Under these conditions, 4-aminophenol
oxidation was observed in both G4DFsc and 3His-G4DFsc(Mut3) (Figure 3 (A) and (B)). Since
these results were not consistent with our prior study, 4-aminophenol (with m-phenylenediame)
was also added simultaneously with O2 saturated buffer to biferrous protein. These conditions
reproduced the observations of prior studies, where the 3His form showed much reduced activity
compared to G4DFsc (Figure 3 (C) and (D)). Biferric species B and C in G4DFsc, as well as the
biferric form of 3His-G4DFsc(Mut3), were also tested and showed no activity (Figure 3 (E), (F)
and (G)). The 4-aminophenol oxidation with both substrate loaded G4DFsc and 3His-
G4DFsc(Mut3) biferrous forms produced multiple equivalents of the indoaniline dye per diiron
site over time (5 eq. after 20 min, Figure S9). These data indicate that the biferrous forms of
G4DFsc and 3His-G4DFsc(Mut3) are the catalytically active species and that in the case of the
3His form, substrate addition prior to oxidation is required for reactivity.
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

Page 9 of 47
Journal of the American Chemical Society
1
2
3
4
Spectroscopic definition of substrate binding.
5
6
7
(A)
(B)
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
G4DFsc
+4-AP
3His
+4-AP
Figure 4: 4-aminophenol (4-AP) perturbation of the MCD spectra of (A) G4DFsc and (B) 3His-
G4DFsc(Mut3) in the absence (red) and presence (green) of 4-aminophenol. All spectra taken at 2 K and
7 T (baseline subtracted). Protein concentrations were ~1 mM with 15-fold excess substrate.
To understand the effects of substrate loading on the biferrous sites, MCD and VTVH MCD
spectroscopies were employed to determine the structures of the Fe(II)Fe(II) forms of these DF
proteins in the presence of 4-AP. While addition of 4-aminophenol to the Fe(II)Fe(II) states of
G4DFsc and 3His-G4DFsc(Mut3) only slightly perturbs their CD spectra (Figure S3), the
substrate addition significantly alters the MCD spectra (Figure 4) for both DF forms. In the
presence of excess 4-aminophenol (15 fold), an additional higher energy feature appears in the
low temperature MCD spectrum of biferrous G4DFsc (Figure 4 (A)) and the MCD intensity
significantly increases for biferrous 3His-G4DFsc(Mut3) (Figure 4 (B)). These changes indicate
that 4-aminophenol is appreciably perturbing the biferrous site in each protein and likely binding
to an Fe(II) center (Figure S18). The CD and MCD features for G4DFsc in the presence of 4-AP
were Gaussian resolved into 3 bands at 7,540(+), 9,000(-), and 10,540(+) cm-1 in the CD
spectrum, and at 7,800, 9,100, and 10,500 cm-1 in the MCD spectrum (Figure 5 (A)). The CD
9
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 47
1
2
3
4
5
6
7
8
9
and MCD spectra for 3His-G4DFsc(Mut3) with excess 4-aminophenol were also resolved into 3
bands (Figure 5 (B)) that have the same signs and energies as those of the 3His form without 4-
aminophenol but with ~7x greater intensity with substrate bound. 52 Note that the small shifts in
energy and narrowing of the band shapes in the MCD spectra are due to the lower temperature of
data collection. The 3 bands for the 3His form (7,400(+), 9,300(-), and 10,500(+) cm-1 for CD
and 7,700, 9,400, and 10,300 cm-1 for MCD) also have the same sign and are at very similar
energies as those for G4DFsc with 4-aminophenol bound (G4DFsc+4-AP). Biferrous 3His-
G4DFsc(Mut3) was previously described as having a 5-coordinate+6-coordinate site based on its
CD and MCD spectra, while that of G4DFsc was described as a 5-coordinate+5coordinate site. 52
The similarities in energy and sign of these CD and MCD features for G4DFsc+4-AP, 3His-
G4DFsc(Mut3)+4-AP, and 3His-G4DFsc(Mut3) indicate that all three forms have equivalent
coordination environments with one 6-coordinate and one 5-coordinate trigonal bipyramidal
Fe(II) center. This would suggest that 4-aminophenol addition increases the coordination of
G4DFsc (5-coordinate+5-coordinate to 5-coordinate+6-coordinate), while maintaining 5-
coordinate+6-coordinate for 4-aminophenol bound 3His-G4DFsc(Mut3) but with a significant
perturbation of the diiron site (i.e. the increased MCD intensity).
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
10
ACS Paragon Plus Environment

Page 11 of 47
Journal of the American Chemical Society
1
2
3
4
5
6
(A)
(B)
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
Figure 5: Gaussian resolution of the CD and MCD spectra of G4DFsc (A) and 3His-G4DFsc(Mut3) (B)
with the addition of 4-aminophenol. Black arrows denote where VTVH MCD isotherms were taken. CD
was taken at 4 °C. MCD was taken at 2 K and 7 T (baseline subtracted). Protein concentrations are ~1
mM with 15-fold excess substrate.
11
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 12 of 47
1
2
3
4
5
6
(A)
(B)
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
Figure 6: The effect of 4-aminophenol (4-AP) addition on VTVH MCD data. Overlay with (purple) and
without (grey) 4-AP for G4DFsc at 9,100 cm-1 (A) and 3His-G4DFsc(Mut3) at 10,400 cm-1 (B). 2 K, 3 K,
5 K, 7.5 K, 10 K, 15 K, 20 K, and 25 K isotherms are shown for all data sets.
The VTVH MCD data for both G4DFsc and 3His-G4DFsc(Mut3) are perturbed with the
addition of 4-aminophenol (Figure 6), which further substantiates the above results that suggest
4-aminophenol binds to these biferrous sites, in this case perturbing their ground states.
Interestingly, the VTVH MCD data for G4DFsc+4-AP and 3His-G4DFsc(Mut3)+4-AP are
similar to each other (Figure 7 (A)).
The spin-manifold for a biferrous system is described by the spin-Hamiltonian (eq. 1): 6
!
!
ꢀ = −2ꢀꢀ! ∙ ꢀ! + ꢀ! ꢀ!! − !! ꢀ! ꢀ! + 1 + ꢀ! ꢀ!!! − ꢀ!!
+
!
!
ꢀ! ꢀ!! − !! ꢀ! ꢀ! + 1 +ꢀ! ꢀ!!! − ꢀ!!
(eq. 1)
where J, D1, D2, E1, and E2 quantify the magnetic coupling between Fe1 and Fe2, and the axial
6
and rhombic ZFS of Fe1 and Fe2, respectively. As previously described, this equation is
applied to the uncoupled basis set |ꢀ!, ꢀ!, ꢀ!!, ꢀ!! of Fe1 and Fe2 to obtain eigenvectors and
eigenvalues from a 25x25 matrix. VTVH MCD data can be fit using a model, also previously
12
ACS Paragon Plus Environment

Page 13 of 47
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
described, 6 that expresses the MCD intensity for a non-Kramers system (i.e. the biferrous site) as
a series of doublets where each doublet has an effective g-value (g||), rhombic ZFS (δ), and C-
term (Atot) and B-term MCD intensities. These doublets correlate to the sublevels given by the
eigenvectors and eigenvalues obtained from eq. 1, and thus, allow determination of the exchange
coupling associated with bridging ligation and the ZFS associated with the ligand field of each
iron center. The VTVH MCD data for both 3His-G4DFsc(Mut3)+4-AP (Figure 7 (B)) and
G4DFsc+4-AP (Figure S5) were fit to the doublet model with similar parameters as described in
the Supporting Information. This gives a system that is antiferromagnetically coupled with
oppositely signed ZFS values for each Fe(II) center (-J < 1 cm-1, D1 = 5-15 cm-1 with (E/D)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
0.33 and -D2 = 5-15 cm-1 with (E/D)2 = 0.33, see Table 1). These parameter ranges are similar to
52
those for 3His-G4DFsc(Mut3) found previously
(Table 1) that describe a weakly
antiferromagnetic exchange coupled site indicating µ-1,3 carboxylate bridges.
13
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 14 of 47
1
2
3
4
5
(A)
(B)
3His+4-AP
G4DFsc+4-AP
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
Figure 7: VTVH MCD data for biferrous G4DFsc and 3His-G4DFsc(Mut3) following 4-aminophenol
addition. (A) Overlay of the VTVH MCD for G4DFsc (purple) (at 9,100 cm-1) and 3His-G4DFsc(Mut3)
(orange) (at 10,400 cm-1) with addition of 4-aminophenol, and (B) Doublet fit to the VTVH MCD of
3His-G4DFsc(Mut3)+4-AP. 2 K, 3 K, 5 K, 7.5 K, 10 K, 15 K, and 20 K isotherms are shown.
Table 1: Spin-Hamiltonian parameters for G4DFsc and 3His-G4DFsc(Mut3)
G4DFsc
3-4
3His
1-3
G4DFsc+4-AP 3His+4-AP
-J (cm-1)
< 1
< 1
Spin
D1 (cm-1)
5-10
0.33
10-15
0.33
5-15
0.33
5-15
0.33
Hamiltonian
Parameters
(E/D)1
D2 (cm-1)
(E/D)2
-7 to -14
0.15
-10 to -15
0.33
-5 to -15
0.33
-5 to -15
0.33
14
ACS Paragon Plus Environment

Page 15 of 47
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
The 4-aminophenol likely binds to the Fe(II)Fe(II) site as a phenolate with the oxygen as
the donor atom (pKa of hydroxo group ~ 10.8). While this phenolate is likely binding to an open
coordinate position of the 2His DF form (G4DFsc), it appears to be replacing a ligand in the 3His
form (based on the increase in MCD intensity). Given the spectroscopic similarities between
G4DFsc+4-AP and 3His-G4DFsc(Mut3)+4-AP, as described above, we propose that the 4-
aminophenolate has replaced the additional active site His at position 100 in 3His-
G4DFsc(Mut3).
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.
The change in the MCD spectrum for both G4DFsc and 3His-G4DFsc(Mut3) upon the
addition of 4-aminophenol provides a spectroscopic approach to determine the rate of substrate
binding. A solution of 4-aminophenol (in 54% d-glycerol) was anaerobically mixed into a
sample of biferrous protein (also in 54% d-glycerol) in an MCD cell and frozen at different time
intervals. The intermediate spectrum of G4DFsc taken from a sample frozen (in liquid nitrogen)
at 0.5 min after addition of 4-AP (Figure 8 (A), teal) is a combination of the final spectrum (after
20 min, dark green) and the initial spectrum (red), and overlays with a composite spectrum
15
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 16 of 47
1
2
3
4
generated using 55-70% of the G4DFsc+4-AP final MCD spectrum and 45-30% of the G4DFsc
5
6
7
MCD spectrum (Figure S6). Assuming a simple A → B binding scheme and using this 0.5 min
8
9
(A)
(B)
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
min.
0
G4DFsc
3His
min.
0
0.5
0.5
20
5
20
Figure 8: MCD spectra monitoring 4-aminophenol binding at different time intervals for (A) G4DFsc and
(B) 3His-G4DFsc(Mut3). All time points taken at 2 K and 7 T. The 0.5 min. time interval in (A) is a
composite of the initial (red) and end (dark green) spectra (30/70). The 5 min time interval in (B) has
intensity consistent with a composition of the initial (red) and end (dark green) spectra (25/75).
interval spectrum, the rate of 4-aminophenol binding is estimated to be kB = 0.027-0.042 s-1.
While the MCD spectrum for 3His-G4DFsc(Mut3) after 4-aminophenol addition at the same
intermediate time interval (0.5 min) (Figure 8 (B), teal) shows little perturbation from its initial
spectrum (red), the spectrum at 5 min after 4-aminophenol addition shows an increase in
intensity (dark yellow) that results in a final spectrum after 20 min (dark green). The
intermediate spectrum (5 min) overlays with a composite spectrum generated using 60-75 % of
the final and 40-25 % of the initial spectrum (Figure S8). As calculated above for G4DFsc, the
rate for 4-aminophenol binding to 3His-G4DFsc(Mut3) is kB = 0.003-0.0046 s-1. These data
demonstrate that 4-aminophenol binds an order of magnitude more rapidly to the biferrous site of
the 2His form than to the biferrous site of the 3His DF variant. This result would be consistent
16
ACS Paragon Plus Environment

Page 17 of 47
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
with a binding mechanism where the 4-aminophenol displaces the 3rd His at the diiron site in
3His-G4DFsc(Mut3), while simply increasing the coordination number of one Fe(II) center of
the 2His form.
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
Based on previous studies, the rates for O2 reactivity of the biferrous G4DFsc and 3His-
G4DFsc(Mut3) forms without substrate are 0.01-0.03 s-1 and 0.01-0.06 s-1. 52 Thus, the rate of O2
oxidation is greater than the rate of substrate binding for the biferrous 3His form, but comparable
for biferrous G4DFsc. These substrate binding and O2 reaction rates thus provide an explanation
for the difference in observed reactivity where the 3His form is active only when the reduced
protein has been incubated with substrate leading to replacement/substitution of the 3rd active site
His ligand, while incubation is not required for activity in the 2His form.
(A)
(B)
0 min.
CD
20 min.
Normalized Intensity
of CD at 10,600 cm-1
Figure 9: The biferrous CD features during the 4-aminophenol oxidation reaction for 3His-
G4DFsc(Mut3). (A) CD of 3His-G4DFsc(Mut3)+4-AP before (red) and after (green) addition of O2
saturated buffer. (B) The time course over 160 min of the normalized intensity of the CD feature at
10,600 cm-1 (red line is the kinetic fit). CD data collected at 4 °C in the presence of ~15-fold excess
substrate (~0.25 mM protein).
17
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 18 of 47
1
2
3
4
5
6
7
8
9
CD spectroscopy was used to detect the amount of the biferrous form of the 3His variant
present during 4-aminophenol oxidation under turnover conditions. After maintaining 90-100%
intensity for ~10 min. (i.e. close to steady state) (Figure 9 (A) and Figure S10), there is a very
slow decay of these NIR CD (the decay can be fit with 2 rates, kd1 = 3 x 10-4 s-1 and kd2 =1.5 x
10-5 s-1, Figure 9 (B)). The decay may be correlated with the consumption of substrate and the
oxidation of 3His-G4DFsc(Mut3) under substrate limited conditions. Given that the resting
biferric form of the 3His does not show oxidase activity over 10 min (Figure 3 (G)) and the
sustained presence of the biferrous form during turnover of 4-aminophenol oxidation, a 2-
electron process may be invoked where O2 is reduced by 2 electrons to H2O2 and 4-aminophenol
is oxidized to quinone imine at the biferrous site. Any 2Fe(III) peroxy-intermediate during
turnover should be rapidly reduced to the biferrous form. These data also suggest that 4-
aminophenol inhibits the rebinding of the 3rd His residue to the biferrous site during steady state
conditions. As substrate is depleted, the slow decay of the biferrous site (over ~ 2.7 hours, Figure
9 (B)) thus reflects rebinding of the 3rd His and oxidation of the 3His site to the minimally
reactive 3His biferric species. Thus, after substrate binding, the 3His and 2His forms have
similar active sites that are competent for 4-aminophenol oxidase reactivity where the bound
substrate is oxidized and O2 is reduced in a 2-electron process.
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
ACS Paragon Plus Environment

Page 19 of 47
Journal of the American Chemical Society
1
2
3
4
Oxygenase Activity.
5
6
Reactivity.
7
8
9
Our prior studies found that 3His-G4DFsc(Mut3), but not G4DFsc, was active in the
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
51
oxygenation of p-anisidine.
This involved the oxidation of p-anisidine (P-AN, 5) to the
corresponding hydroxylamine (6). The hydroxylamine was directly observed when p-amino-
benzonitrile was used as the substrate. However, in the case of the p-anisidine, the
hydroxylamine product was instead oxidized to 4-nitroso-methoxybenzene (N-MB, 7), which
coupled with a second molecule of P-AN to form 4-methoxy-N-(4-nitrosophenyl)aniline (M-
NPA, 8) (Figure 10). The experimental conditions for these studies involved adding Fe(II) to a
solution of apo-protein and substrate under aerobic conditions. While Fe(II) should be bound,
these conditions allowed for either the biferric or biferrous forms of 3His-G4DFsc(Mut3) as the
active species. This work extends on prior results to identify which form of 3His-G4DFsc(Mut3)
is active.
O
O
N
O
O
O
+ O2, 2H+, 2e-
NH2
H2O
+ 5
O
OH
NH
N
N
H
N-MB
P-AN
2H+, 2e-
M-NPA
CH3OH
6
7
5
8
Figure 10: The reaction scheme of the oxidation of p-anisidine (P-AN) (5) to 4-hydroxylamine-
methoxybenzene (6), with addition oxidation to 4-nitroso-methoxybenzene (N-MB) (7), and the coupling
of N-MB and P-AN to form 4-methoxy-N-(4-nitrosophenyl)aniline (M-NPA) (8).
When p-anisidine substrate was incubated with biferrous 3His-G4DFsc(Mut3)
anaerobically and then reacted with O2, the UV-Vis absorption spectra showed the production of
19
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 20 of 47
1
2
3
4
5
6
7
8
9
N-MB at 342 nm and the coupled product M-NPA at 431 nm (Figure 11 (A)). The UV-Vis
absorption spectra for a sample in which O2 and p-anisidine were added simultaneously to
biferrous 3His-G4DFsc(Mut3) also showed reactivity but only ~1/2 that of the biferrous protein
with substrate preloaded (Figure 11 (B)). The biferric form of 3His-G4DFsc(Mut3) showed no
activity (Figure 11 (C)) provided the buffer was exchanged after the biferric state was formed
and prior to p-anisidine addition. [Note that without exchanging the buffer of the biferric protein,
p-anisidine oxidation was observed (Figure S11 (A)). However, this oxidation can be attributed
to the presence of hydrogen peroxide (where the protein has generated a stoichiometric amount),
which independently did test positively for p-anisidine oxidation (Figure S11 (B)) and showed
similar absorption changes (Figure S11 (C)) as exhibited by the biferric form if the buffer was
not exchanged.] These UV-Vis absorption results were confirmed by LCMS, where anaerobic
addition of p-anisidine to biferrous 3His-G4DFsc(Mut3) followed by addition of O2 saturated
buffer (Figure S12 (A)) led to greater amounts of N-MB (m/z=138, retention time 12.6 min) and
product-NPA (m/z=229, retention time 24.21 min.) than the H2O2 present in the biferric reaction
mixture (Figure S12 (B)) (2 x the area of the peak associated with nitroso product and ~18 x the
area of the peak associated with the coupled product). These results show that substrate binding
to the biferrous 3His protein initiates the amino-oxygenation reaction.
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

Page 21 of 47
Journal of the American Chemical Society
1
2
3
4
5
6
(A)
(B)
(C)
7
8
9
O2 reaction with
biferrous
Simult. P-AN and O2
addition to biferrous
3His
P-AN addition to
biferric 3His
3His+P-AN
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
0 min
60 min
Figure 11: The oxidation of p-anisidine as followed by UV-Vis absorption spectroscopy. (A) p-anisidine
was added anaerobically to biferrous 3His-G4DFsc(Mut3) followed by addition of O2 saturated buffer.
(B) p-anisidine and O2 saturated buffer were added simultaneously to biferrous 3His-G4DFsc(Mut3). (C)
p-anisidine was added to biferric 3His-G4DFsc(Mut3) (buffer exchanged prior to p-anisidine addition to
eliminate any presence of H2O2 in solution). (A) and (B) are data collected over 180 min. Data taken at 20
°C. Final samples contained 65 µM of protein and 1 mM of substrate.
Alternatively, for biferrous G4DFsc, the anaerobic addition of the p-anisidine substrate
followed by O2 saturated buffer led to no discernable peaks in the LCMS associated with product
formation (Figure S12 (C)). The competition between substrate binding and O2 reactivity can
thus be excluded as a possible explanation for the lack of activity of the 2His form that was
previously reported. 51
Spectroscopic definition of substrate binding.
Since the reactivity studies indicate that substrate binding to the reduced 3His protein is
necessary for amino-oxygenation reactivity, the effect of p-anisidine on the biferrous site was
investigated with CD, MCD and VTVH MCD spectroscopy. Substrate addition (+P-AN) to
G4DFsc and 3His-G4DFsc(Mut3) only minimally perturbs the CD (Figure 12 (A) and (B) top)
and MCD (Figure 12 (A) and (B) bottom) spectra with slight shifts in the energies and intensities
21
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 22 of 47
1
2
3
4
5
6
7
8
9
of the bands. The most notable change is that the lower energy feature of the MCD spectrum of
the 3His form shifts from ~7,700 cm-1 (shoulder in red spectrum, Figure 12 (B) bottom) to
~7,400 cm-1 (green spectrum in Figure 12 (B) bottom). These small changes indicate that the
substrate binding affects the biferrous site, but does not change the coordination, and thus, is not
likely bound directly to the diiron center.
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)
(B)
Figure 12: The effect of p-anisidine addition on the CD and MCD. The CD (top) and MCD (bottom) of
(A) G4DFsc and (B) 3His-G4DFsc(Mut3) with unbound (red) and bound (green) forms. Protein
concentration is ~1 mM with 20-30 fold excess substrate. MCD data taken at 2 K and 7 T (baseline
subtracted). CD data taken at 4 °C. Arrows in bottom spectra indicate where VTVH MCD data were
taken.
22
ACS Paragon Plus Environment

Page 23 of 47
Journal of the American Chemical Society
1
2
3
4
5
6
(A)
(B)
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
Figure 13: The effect of p-anisidine addition on the VTVH MCD isotherms. (A) The VTVH MCD of
G4DFsc (grey) and G4DFsc+P-AN (purple) at 9,100 cm-1 is shown. (B) The VTVH MCD of 3His-
G4FDsc(Mut3) (grey) and 3His-G4FDsc(Mut3)+P-AN (purple) at 10,400 cm-1 is shown. 2 K, 3 K, 5 K,
7.5 K, 10 K, 15 K, and 20 K isotherms are shown.
(A)
(B)
Figure 14: Doublet fits for the VTVH MCD data for (A) G4DFsc+P-AN and (B) 3His-G4DFsc(Mut3)+P-
AN. Data for G4DFsc were collected at 9,100 cm-1 and those for the 3His form were collected at 10,500
cm-1.
23
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 24 of 47
1
2
3
4
5
Table 2: Spin Hamiltonian parameters for G4DFsc and 3His-G4DFsc(Mut3) in the presence of p-
anisidine.
6
7
G4DFsc 3His-G4DFsc(Mut3)
8
9
-J (cm-1)
D1 (cm-1)
(E/D)1
< 2
5-15
2-3
5-10
Spin
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
Hamiltonian
Parameters
0.33
0.33
D2 (cm-1)
-5 to -15
-5 to -10
0.33
0.33
(E/D)2
While only slightly perturbing the CD and MCD features of both biferrous G4DFsc and 3His-
G4DFsc(Mut3), p-anisidine addition does markedly change the VTVH MCD behavior for both
forms (VTVH MCD curves change from grey to purple upon substrate binding in Figure 13 (A)
and (B)). For G4DFsc+P-AN, the VTVH MCD data (Figure 14 (A)) were fit using the doublet
model (vide supra) as described in Supporting Information. The VTVH MCD data of
G4DFsc+P-AN indicate that -J < 2 cm-1 (Table 2) for this form, which is less in magnitude than
that found for G4DFsc without substrate (-J ~ 3-4 cm-1). For 3His-G4DFsc(Mut3)+P-AN,
doublet model fitting (see Supporting Information) of the VTVH MCD data (Figure 14 (B))
indicates that there is a slight increase in the magnitude of the antiferromagnetic coupling (-J~2-3
cm-1 3His-G4DFsc(Mut3)+P-AN, and 1-2 cm-1 3His-G4DFsc(Mut3)-P-AN) (Table 2).
The antiferromagnetic exchange coupling values for both G4DFsc+P-AN and 3His-
G4DFsc+P-AN are consistent with µ-1,3 carboxylate bridges between the two Fe(II) sites.
However, the change in their magnitude suggests that p-anisidine is binding near the biferrous
site and perturbing the carboxylate bridges between the ferrous centers.
24
ACS Paragon Plus Environment

Page 25 of 47
Journal of the American Chemical Society
1
2
3
4
Kinetics.
5
6
7
8
9
Our previous studies of the O2 reactivity of biferrous G4DFsc and 3His-G4DFsc(Mut3) suggest
that the exchange coupling and O2 reaction rates are correlated, resulting from the O2 reaction
occurring at only one Fe(II) site but being reduced by 2 electrons. The transfer of the 2nd electron
from the remote Fe(II) site would then occur through the carboxylate bridges. In this model, the
decrease in the carboxylate bridge mediated exchange coupling of G4DFsc upon addition of p-
anisidine would result in decreased O2 reactivity. This is confirmed where the rate of the O2
reaction for G4DFsc+P-AN is 0.002 s-1 (Figure S14) (relative to 0.01-0.03 s-1 without p-
anisidine). 52 In contrast, the binding of p-anisidine to the 3His form results in only a small
change in exchange coupling (-J from 1-2 cm-1 to 2-3 cm-1) and the rate of obtaining steady state
for nitroso production (k ~ 0.045 s-1, Figure S15) is close to the O2 reaction rate of the 3His form
without substrate (0.01-0.06 s-1). 52
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
25
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 26 of 47
1
2
3
4
(A)
(B)
5
6
7
CD at 10,500 cm-1
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
Figure 15: NIR CD spectra of the p-anisidine reaction with 3His-G4DFsc(Mut3). (A) CD spectra of the
3His protein in the presence of p-anisidine before exposure to O2 (green), under aerobic conditions (teal),
and 30 min after first allowing O2 containing buffer (~250 µM of O2) to react for 60 sec and then sealing
the CD cell to create a closed system. Protein concentrations for red and teal are 250 µM. (B) The time
dependence of the 10,500 cm-1 feature after first allowing 3His-G4DFsc(Mut3)+P-AN to react with O2
followed by sealing the CD cell to create a closed system (see red in (A)). Data collected at 20 °C.
NIR CD and MCD spectroscopy were used to observe the biferrous bands of 3His-
G4DFsc(Mut3) during the arylamine oxygenation reaction. Under aerobic conditions (i.e.
turnover conditions), there are no NIR CD features present during production of the nitroso
product (teal, Figure 15 (A)), which indicates that 3His-G4DFsc(Mut3) is mostly present in the
biferric form at steady state, and thus, any reduction of the diiron center during turnover (Figure
S16) must be rate limiting. In oxygen-depleted environments during nitroso production, the NIR
CD and MCD features associated with biferrous 3His-G4DFsc(Mut3) are regenerated (Figure 15
(A, red) and (B) and Figure S17 (A)). Since p-anisidine does not reduce the biferric form of
3His-G4DFsc(Mut3) (Figure S17 (B)), the regeneration of Fe(II)Fe(II) state suggests that the
biferric 3His form is reduced by a reaction intermediate during catalysis.
The oxygenation of p-anisidine to the nitroso product is a 4-electron process where the
first 2-electron step should be the hydroxylation of the amine group followed by oxidation to the
26
ACS Paragon Plus Environment

Page 27 of 47
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
nitroso product. The above results suggest that the biferrous 3His-G4DFsc(Mut3) variant
catalyzes the oxygenation of the amine group to a hydroxylamine and results in a biferric
species. This hydroxylamine may then be oxidized to the nitroso product by the Fe(III)Fe(III)
state of the 3His protein, regenerating the Fe(II)Fe(II) state in the rate limiting step.
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
Discussion
Our previous studies identified markedly different reactivities in the 2His and 3His 4A→4G
variants of DFsc, where 4-aminophenol oxidase activity was observed in the 2His variant
(G4DFsc), but not in the 3His variant (3His-G4DFsc(Mut3)), and arylamine oxygenase activity
was detected for the 3His variant, but not the 2His form. 51 This study has extended these results
to understand the molecular basis for these reactivity differences.
Oxidase. By systematically controlling the reaction conditions, we have identified the active
species for the 4-aminophenol oxidation. Both the 2His and 3His variants display oxidase
activity when substrate binding to the biferrous site preceded the reaction with O2. The CD,
MCD and VTVH MCD studies show substantial perturbation of the biferrous sites of both
G4DFsc and 3His-G4DFsc(Mut3) by 4-aminophenol. While 4-aminophenol addition to G4DFsc
resulted in a new higher energy feature (~10,500 cm-1) in the MCD spectrum, its addition to
3His-G4DFsc(Mut3) causes a marked increase in MCD intensity (Δε ~1 to ~7 cm-1M-1). These
perturbations suggest that this substrate binds directly to the biferrous sites of both forms.
Gaussian resolution of the CD and MCD features for both G4DFsc and 3His-
G4DFsc(Mut3) in the presence of 4-AP reveal three bands that are similar in sign and energy to
3His-G4DFsc(Mut3) without 4-AP and suggest that all three forms have a 5-coordinate+6-
coordinate trigonal bipyramidal biferrous site. Given the lack of increased coordination in the
27
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 28 of 47
1
2
3
4
5
6
7
8
9
3His form upon substrate addition, it may be inferred that structural limitations prevent the
substrate from simply binding to the open site on the 5-coordinate iron center of the 3His variant.
The VTVH MCD data of biferrous G4DFsc and 3His-G4DFsc(Mut3) were also
perturbed by the binding of 4-aminophenol. The associated spin-Hamiltonian parameters show
that both forms are weakly antiferromagnetic coupled, indicating they still have µ-1,3
carboxylates bridging the biferrous site. These spectral data show that G4DFsc+4-AP and 3His-
G4DFsc(Mut3)+4-AP have similar site structures where 4-aminophenol binds to and increases
the coordination of one Fe(II) in G4DFsc while replacing a ligand (likely the additional His) in
3His-G4DFsc(Mut3).
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
Substrate binding to the biferrous site initiates the oxidase activity. Therefore, the kinetics
of 4-aminophenol binding to the biferrous sites of G4DFsc and 3His-G4DFsc(Mut3) were
explored with "freeze quench" MCD spectroscopy. The binding rate of 4-aminophenol for
G4DFsc (kB = 0.027-0.042 s-1) was found to be an order of magnitude greater than that for the
3His variant (kB = 0.003-0.0046 s-1). However, the O2 reaction rates of both proteins are
comparable at around ~10-2 s-1. 52 Based on the kinetics, 4-aminophenol binding to the 3His
variant is an order of magnitude less than the O2 reaction rate. Thus, when substrate and O2 are
simultaneously added, the biferrous 3His form first reacts with O2 to generate an inactive biferric
species prior to substrate binding, which renders the protein ineffective for substrate oxidation.
Since the 2His form exhibits faster 4-aminophenol binding (likely due to the decreased
coordination of the biferrous site), the slow substrate binding in the 3His variant may be
attributed to its competition with the additional His ligand. The 3His-G4DFsc(Mut3) variant
therefore becomes active only upon anaerobic incubation of substrate with the biferrous form,
presumably following replacement of the additional His ligand at the active site by 4-AP.
28
ACS Paragon Plus Environment

Page 29 of 47
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
NIR CD spectroscopy was used to monitor the ferrous d → d transitions during 4-
aminophenol oxidation and showed the dominant presence of the biferrous 3His-G4DFsc(Mut3)
species during steady state. Given that the resting biferric forms were inactive for substrate
oxidation, a 2-electron process may be invoked where O2 is reduced to H2O2 (which is lost) and
4-aminophenol is oxidized to the benzoquinone imine at the biferrous site (Figure 16). The
presence of 4-aminophenol likely inhibits rebinding of the 3rd His ligand to the biferrous site
during turnover. There are several possible mechanisms for inhibition of the rebinding of the 3rd
His: 1) one 4-aminophenol molecule stays bound during turnover, or 2) the 4-aminophenol out-
competes the 3rd His protein residue in binding to the diiron center. The former possibility
would involve a ping-pong mechanism since only one open coordination site is available for O2
reduction and 4-aminophenol oxidation, while the latter would allow for direct O2 reduction at
one iron site and 4-aminophenol oxidation at the other.
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
The observed lack of oxidase activity of the biferric species in G4DFsc and 3His-
53
G4DFsc(Mut3) is in contrast to other DF constructs. Since the NIR CD and MCD data
demonstrate that substrate is accessible to the diiron sites, the lack of activity may be related to
either unfavorable thermodynamics (redox potentials of G4DFsc and 3His-G4DFsc(Mut3) are
unknown, but the 2-electron potential for 4-aminophenol is ~300 mV (converted to SHE, pH 7-
9)) 54,55 or the coordinative saturation that is proposed for the ferric centers in G4DFsc and 3His-
G4DFsc(Mut3). 52.
29
ACS Paragon Plus Environment