Light-Activated Electron Transfer and Turnover in Ru-Modified Aldehyde Deformylating Oxygenases

Article abstract of DOI:10.1021/acs.inorgchem.8b00673

Conversion of biological molecules into fuels or other useful chemicals is an ongoing chemical challenge. One class of enzymes that has received attention for such applications is aldehyde deformylating oxygenase (ADO) enzymes. These enzymes convert aliphatic aldehydes to the alkanes and formate. In this work, we prepared and investigated ADO enzymes modified with Ru^II(tris-diimine) photosensitizers as a starting point for probing intramolecular electron transfer events. Three variants were prepared, with Ru^II-modification at the wild type (WT) residue C70, at the R62C site in one mutant ADO, and at both C62 and C70 in a second mutant ADO protein. The single-site modification of WT ADO at C70 using a cysteine-reactive label is an important observation and opens a way forward for new studies of electron flow, mechanism, and redox catalysis in ADO. These Ru-ADO constructs can perform the ADO catalytic cycle in the presence of light and a sacrificial reductant. In this work, the Ru photosensitizer serves as a tethered, artificial reductase that promotes turnover of aldehyde substrates with different carbon chain lengths. Peroxide side products were detected for shorter chain aldehydes, concomitant with less productive turnover. Analysis using semiclassical electron transfer theory supports proposals for hopping pathway for electron flow in WT ADO and in our new Ru-ADO proteins.

Full text of DOI:10.1021/acs.inorgchem.8b00673

Article
pubs.acs.org/IC
Cite This: Inorg. Chem. XXXX, XXX, XXX−XXX
Light-Activated Electron Transfer and Turnover in Ru-Modified
Aldehyde Deformylating Oxygenases
Rajneesh K. Bains, Jessica J. Miller, Hannah K. van der Roest, Sheng Qu, Brad Lute,
and Jeffrey J. Warren*
Department of Chemistry, Simon Fraser University, 8888 University Drive, Burnaby, British Columbia V5A 1S6, Canada
S
* Supporting Information
ABSTRACT: Conversion of biological molecules into fuels or other useful
chemicals is an ongoing chemical challenge. One class of enzymes that has received
attention for such applications is aldehyde deformylating oxygenase (ADO)
enzymes. These enzymes convert aliphatic aldehydes to the alkanes and formate. In
this work, we prepared and investigated ADO enzymes modified with RuII(tris-
diimine) photosensitizers as a starting point for probing intramolecular electron
transfer events. Three variants were prepared, with RuII-modification at the wild
type (WT) residue C70, at the R62C site in one mutant ADO, and at both C62
and C70 in a second mutant ADO protein. The single-site modification of WT
ADO at C70 using a cysteine-reactive label is an important observation and opens
a way forward for new studies of electron flow, mechanism, and redox catalysis in
ADO. These Ru-ADO constructs can perform the ADO catalytic cycle in the
presence of light and a sacrificial reductant. In this work, the Ru photosensitizer serves as a tethered, artificial reductase that
promotes turnover of aldehyde substrates with different carbon chain lengths. Peroxide side products were detected for shorter
chain aldehydes, concomitant with less productive turnover. Analysis using semiclassical electron transfer theory supports
proposals for hopping pathway for electron flow in WT ADO and in our new Ru-ADO proteins.
rate-limiting step in ADO catalysis. In related mechanistic
work, radicals were shown to form at Tyr (Y) and Cys (C)
residues and mutations to those residues affect aldehyde
yields.10 Those Y and C radicals are on a pathway that extends
from surface residues C70 to the iron(FeB)-ligating His63
(Figure 1).10 Crystallographic data11 and computational12
analyses of the active site suggests that His attached to FeB is
dynamically ligated and depends on the redox state of the
INTRODUCTION
■
Diiron cyanobacterial aldehyde deformylating oxygenase
(ADO) enzymes have attracted attention for their unique
chemistry and potential uses in the production of biofuels.1 In
natural systems, ADO enzymes convert fatty aldehydes to the
corresponding Cn−1 alkane and formate (eq 1).2 In vivo, the
aldehyde substrates derive from acyl-carrier proteins.3,4 From
the perspective of those aldehyde substrates, the overall
reaction is redox neutral. However, turnover in ADO is
coupled to the 4H+/4e− reduction of O2 to two H2O.5
Remarkably, enzymatic cascades that include ADO can be
integrated into synthetic metabolic pathways in other microbes
to produce alkanes of various chain lengths, albeit in modest
yields.6,7 In vitro enzymatic assays display turnover rates near 1
min−1 using either chemical or protein-based reducing
systems.5,8,9
RCH2CHO + O2 + 4H+ + 4e−
→ RCH3 + HCO2H + H2O
(1)
The reaction catalyzed by ADO (eq 1) requires four single
electron transfer (ET) steps. The natural reductase for ADO is
not yet known, but a recent report showed that a
cyanobacterial 2Fe-2S ferredoxin (petF) is an effective
reductant of the diferric active site with a rate constant of
∼400 s−1.10 This ferredoxin is superior to other chemical
reduction systems5,8,9 and shows that enzymatic ET is faster
than observed turnover frequencies. In this context, ET is not a
Figure 1. Structure of ADO from S. elongatus PCC7942 (PDB ID
4RC5). The diiron site is shown as orange spheres, H63 is in green,
the site of mutation at R62 is in cyan, and the α-helix that connects
C70 and the iron site is in magenta.
Received: March 17, 2018
© XXXX American Chemical Society
A
DOI: 10.1021/acs.inorgchem.8b00673
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 2. (A) Optical spectra of Ru (red dashed) and wild-type ADO (blue solid). (B) Optical (solid) and fluorescence (dashed) spectra of Ru-
ADO proteins: Ru-C70 (blue), Ru-C62 (red), and Ru-C62/Ru-C70 (green). The arrows indicate axes corresponding to each set of spectra.
Additional optical spectra are given in the Supporting Information.
The codon-optimized gene for Synechococcus elongatus ADO was a
gift from D. Baker (University of Washington). The ADO mutants
(R62C and R62C/C70S/C116S) were prepared using the Agilent
Quikchange protocol. Introduction of the desired mutations was
confirmed from DNA sequencing performed by Eurofins Operon.
Protein expression and purification was performed using standard
procedures (see Supporting Information).
active site. Ligands on FeA appear to be static. The His
coordination behavior highlights the importance of FeB and is
consistent with a mechanism that could involve ET (likely via
hopping) from the ADO surface near C70 to the iron active
site.13,14 Further support for a hopping mechanism comes from
the observation of Y and C radicals during turnover,10 where
these are likely derive from chemistry carried out in the Fe
active site.15 However, to learn more about intramolecular ET
in ADO enzymes, new models are needed that allow for rapid
initiation of redox reactions. To address this, we integrated Ru-
tris(diimine) photosensitizers at two different sites on the
surface of ADO. Herein we describe the preparation,
characterization, and turnover characteristics of Ru-ADO
proteins.
For Ru-labeling, 1,4-dithiothreitol (DTT) was added to ADO
solutions to a final concentration of 20 mM and incubated at room
temperature for at least 45 min. DTT was removed using a PD-10
desalting column. Ru was dissolved in dimethyl sulfoxide (DMSO)
and added (1.2 equiv for single labeling and 2.4 equiv for double
labeling) to 60 μM protein solutions, and the resulting mixtures were
stirred in the dark at 4 °C overnight. Final DMSO concentrations
were 1%. No precipitation was observed during incubation.
Optimization of the protocol showed that 60 μM protein with a
20% excess of Ru-label and a 12 h labeling time (12 h vs 4 or 8 h)
gave the highest and most reproducible yields (70+%). Labeling at
room temperature was more rapid but also resulted in some protein
precipitation. Crude Ru-modified protein solutions were exchanged
into 5 mM 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid
(HEPES; pH 7.6) using a PD-10 column and loaded onto an
FPLC DEAE Hi-Prep column eluted with an increasing NaCl
gradient. Fractions containing Ru-labeled protein were pooled and
exchanged into 50 mM sodium phosphate, 100 mM NaCl, pH 7.5
using a PD-10 column. Protein samples were stored in the dark at 4
°C. Typical isolated yields of Ru-modified proteins ranged from 30 to
50%. Labeling was confirmed using UV−visible spectroscopy and
trypsin digestions (see Supporting Information).
Photochemical reaction mixtures contained 10 μM Ru-ADO + 10
mM sodium diethyldithiocarbamate (DTC) in 50 mM sodium
phosphate and 100 mM NaCl (pH 7.5) buffer. Ru-ADO was mixed
with 2 mol equiv of ferrous ammonium sulfate prior to reactions. For
all reactions, a single reaction mixture was split into two aliquots,
where one was shielded from light. The remaining mixture was
exposed to visible light from a Kodak extra bright lamp projector
source. Mixtures were stirred continuously in a temperature-
controlled water bath. Prior to NMR analysis, reaction solutions
were centrifuged briefly to remove insoluble oxidized DTC, and 10%
(v/v) D2O was added. The solution was then transferred to NMR
tubes (5 mm diameter), and NMR spectra were collected. Ethyl
acetate-extracted aqueous reaction mixtures were analyzed using GC-
MS. Full descriptions of protocols are given in the Supporting
Information.
EXPERIMENTAL SECTION
■
Materials and Instrumentation. Solvents and buffer salts were
from J. T. Baker unless otherwise noted. RuCl3 was from Pressure
Chemical. cis-Dichlororuthenium(2,2′-bipyridyl)2 (Cl2Ru(bpy)2)16
and (4-bromomethyl-4′-methyl-2,2′-bipyridyl)(2,2′-
bipyridyl)2RuII(PF6)2 (denoted Ru)17 were prepared according to
the literature. Full details are given in the Supporting Information.
Nickel nitrilotriacetic acid resin was from Thermo Fisher. All other
chromatography media was from GE Healthcare. Polymerase chain
reaction (PCR) primers were purchased from Eurofins Operon.
Restriction enzymes and Q5 DNA polymerase were from New
England Biolabs. Plasmid DNA was isolated using a Qiagen Spin
Miniprep kit.
UV−Visible spectra were obtained using either a Photon Control
SPM-002-EH CCD with an SPLC 1DH deuterium/tungsten light
source or a Cary 100-Bio UV−visible spectrometer. A Bruker
microFLEX MALDI-TOF instrument with a 48-well ground steel
target was used for all mass spectrometry (MS) experiments. The
dried droplet method was used with sinapinic acid as the matrix.
Theoretical molecular weights are obtained from the protein sequence
using ExPASy: Compete Pi/MW tool.18 Circular dichroism spectra
were collected using a Chirascan qCD spectrometer and converted
using the following formula [θ] = θMr/(nCl), where Mr, n, C, and l
represent molecular weight (Da), number of residues, concentration
(in mg/mL), and the cuvette path length, respectively. Fluorescence
spectra were collected on Jobin-Yvon FluoroLog3 fluorometer. All
samples were deoxygenated using repeated pump-purge cycles with
argon backfill. Samples were excited at 455 nm. NMR experiments
were performed using a Bruker AVANCE III 500 MHz running
IconNMR under TopSpin 2.1. A Simpson water suppression pulse
program was used for all aqueous NMR experiments. Gas
chromatography-mass spectrometry (GC-MS) experiments were
performed using an Agilent 6890 series gas chromatograph with
5973 Network Mass Selective Detector equipped with a capillary DB-
5 column (0.25 mm × 30m × 0.25 μm).
RESULTS
■
Analysis of the X-ray structure of ADO from S. elongatus strain
PCC794219 reveals the presence of three C (residues 70, 106,
116). Of those, C70 and C116 are the most exposed to solvent.
Cysteine-reactive (4-bromomethyl-4′-methyl-2,2′-bipyridyl)-
(2,2′-bipyridyl)2RuII(PF6)2 (denoted as Ru) was used to
B
DOI: 10.1021/acs.inorgchem.8b00673
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
modify the proteins described in this work; such Ru-complexes
have been used in several studies of ET in metallopro-
teins.20−22 Surprisingly, treatment of WT ADO with Ru23
results in a single cysteine modification. Trypsin digestion and
analysis of the fragments using mass spectrometry reveals that
C70 is the modified site (see Supporting Information). Site-
directed mutagenesis was also used to produce two other
mutant ADO proteins: R62C and R62C/C70S/C116S.
Residue 62 was chosen for modification because it is surface-
exposed and connected directly to the Fe active site via His63.
As shown in Ru-modified cytochromes P450, the site of
labeling is an important factor in photochemical enzyme
activity.24 Treatment of R62C ADO with Ru yields a protein
modified at both C62 and C70, as confirmed by trypsin
digestion experiments, while the R62C/C70S/C116S variant is
only labeled once (Ru-C62). Circular dichroism spectra of
unlabeled mutants and Ru-modified analogues are nearly
identical, suggesting that that Ru-labeling does not dramati-
cally alter ADO’s helical secondary structures. Analogous
mutations to WT Cys residues are known in Nostoc punctiforme
ADO, and these mutations thermally destabilize the protein by
ca. 6 °C, but the secondary structures are not strongly
affected.25 Finally, Ru modification of WT ADO lowers the
melting temperature by ∼3 °C (see Supporting Information).
The optical and fluorescence spectra (Figure 2) of Ru-ADO
proteins show characteristic metal-to-ligand charge transfer
(MLCT) absorption bands, and the maximum emission
intensity is at 600 nm (with 455 nm excitation). There is a
slight red shift in absorbance between the Ru-modified
proteins (λmax = 458 nm) and the label (λmax = 455 nm).
The intensity of the MLCT band is proportional to the
number of Ru labels. The extinction coefficients at 455 nm are
ε (Ru-C70) = 14 500 M−1 cm−1, ε (Ru-C62) = 14 500 M−1
cm−1, and ε (Ru-C62/Ru-C70) = 29 000 M−1 cm−1, based on
spectra of related ruthenium polypyridyl complexes.26 Addi-
tional UV−visible spectra are given in the Supporting
Information. The normalized fluorescence maxima show the
same trend. The observed luminescence decay time constants
(τ) in oxygenated buffer for *Ru are 102 ns (Ru-C70), 78 ns
(Ru-C62), and 51 ns (Ru-C62/Ru-C70) (see Supporting
product expected from DTC quenching of *RuII. Analysis
using 1H NMR (see Supporting Information) also shows
formation of a downfield singlet at 8.35 ppm (Figure 3). This
1
Figure 3. H NMR spectra of 10 μM Ru-C70 ADO samples under
photocatalytic conditions using undecanal as substrate. Spectrum 1
(bottom) is a dark control, and spectra 2 (middle) and 3 (top) were
collected after 20 and 40 min of illumination, respectively. The *
indicates the chemical shift of HCOO−.
peak was assigned to formate based on spiking with authentic
sample and comparison to external standards. The control
(dark) samples routinely showed formation of trace formate
over the course of 40 min, which we attribute to the weak
reducing power of DTC, which is present in large excess.
Integrated peak areas suggest formation of 40 10 μM (5
2 mg/L) over the course of 60 min of irradiation. This places
an upper limit on formate production of Ru-ADO near 0.7 μM
min−1, consistent with other chemical reducing systems.29
Identical samples kept in the dark show no disulfiram
precipitate or almost no production of formate (see above).
Production of formate also was detected with aldehydes of
varying chain lengths ranging from C6 to C11 (with the
exception of C9, which was not tested). The substrate
specificity and turnover properties of ADO with these other
chain lengths has been investigated elsewhere in detail.30
Control reactions where any one of the reaction components
(Ru-ADO, DTC, or aldehyde) were omitted showed no
production of formate. Octanal was used as a test substrate due
to its higher water solubility. Following literature examples of
photoreduction of proteins using noncovalently bound Ru-
(bpy)3Cl2,31 addition of Ru(bpy)3Cl2 in a 1:1 molar ratio with
unmodified WT ADO, and all other reaction components,
shows production of trace formate (<1 μM). In this control,
consumption of DTC was observed. These experiments
demonstrate that covalent attachment of Ru to ADO is
important for function.
2+
Information). For comparison, the lifetime for *Ru(bpy)3 is
2+
140 ns under the same conditions (Ru(bpy)3 = tris(2,2′-
bipyridyl)ruthenium dication).
Electronically excited *RuII complexes can be quenched by
sacrificial electron donors to yield the corresponding reduced
Ru complex.27 Reduced Ru is a strong reductant, with E°′ near
−1.2 V versus the normal hydrogen electrode (NHE).28 Here,
DTC was used as the sacrificial quencher. The low water
solubility and limited reactivity of the DTC oxidation product
(disufiram) make it useful in investigation of protein
photocatalysis.22,23 Addition of 10 mM DTC to 10 μM
solutions of Ru-ADO in 50 mM sodium phosphate and 100
mM sodium chloride, pH 7, quenches the steady-state
luminescence of all proteins by ca. 60% (spectra are shown
in the Supporting Information).
When using aldehydes shorter than C11, other products were
detected using 1H NMR and GC-MS at reaction times greater
than 20 min. For example, GC-MS analyses showed trace
production of alcohols, as is known for ADO.32 At those longer
reaction times, and with shorter aldehydes (C6−C8), a
1
resonance at 9.1 ppm also was observed in H NMR spectra.
This chemical shift suggests production of hydroperoxide-
containing compounds. This hypothesis is consistent with
detection of alkyl peroxyl radicals during ADO turnover.10 To
confirm peroxide production, postillumination reaction
mixtures were extracted with PPh3 in CDCl3 and analyzed
using 31P NMR. A 31P NMR resonance at 29 ppm was
observed (see Supporting Information), which was confirmed
as OPPh3 by spiking with authentic sample. The modest
turnover on Ru-ADO can be partially attributed to production
Next, the ability of Ru-ADO to turn over using light and
sacrificial reductants was investigated. Aqueous solutions
containing 10 μM Ru-ADO, 10 mM DTC, and saturating
undecanal (∼160 μM) in 50 mM sodium phosphate, 100 mM
sodium chloride, pH 7.5, were irradiated for up to 60 min at
room temperature using a halogen light source with a UV
cutoff filter (λ > 380 nm). During irradiation, a white
precipitate formed, consistent with formation of the disulfiram
C
DOI: 10.1021/acs.inorgchem.8b00673
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
of these side products. Any production of hydrogen peroxide
also could inhibit turnover.8
GC-MS was used to identify the decane produced from
decarbonylation of undecanal by Ru-ADO. The decane
product was extracted using ethyl acetate. The yields assayed
using GC-MS were: 2.1 0.1 (Ru-C62/Ru-C70), 2.0 0.1
(Ru-C70), and 0.6 0.1 (Ru-C62) mg/L, respectively. The
apparent yields of alkanes are somewhat lower than for
formate, which we attribute to inefficiencies in the extraction
protocol. The NMR experiments do not require extraction and
are a better measure of true yields. Notably, the production of
formate and decane for Ru-C62 ADO (with a C70S mutation)
is significantly (ca. 70%) lower than the other two Ru-ADO
models with Ru at C70. In recent work, the C71A variant of N.
punctiforme ADO (analogous to residue C70 in this work)
showed a 27% reduction in activity with respect to the WT
protein, demonstrating the importance of that cysteine residue
for ADO function.10 Overall, the two Ru-ADO proteins with
Ru at C70 show greater activity than the variant labeled only at
C62.
Figure 4. Illustration of the Y chain that extends from the S. elongatus
ADO Fe site to the protein surface (C70 or Y17). The peptide
connecting Ru to the Fe active site in Ru-ADO is shown at the top.
Distances are to/from aromatic ring centers. PDB ID 4RC5.
protective roles in other metalloenzymes.36 This could be an
interesting example of a case where such a chain plays a direct
functional role in enzyme catalysis. In ADO, this series of
residues may play a role in turnover that is reminiscent of
electron transport in some ribonucleotide reductases
(RNRs).37 Some of these Y residues were identified as sources
of radicals for ADO under turnover conditions, and mutations
to those Y reduced, but did not destroy, ADO catalytic
activity.10 Residue C70 is situated near the protein surfaces and
is about equidistant from Y125 and Y21. In the following
paragraphs, we analyze potential ET pathways in the context of
natural and Ru-ADO.
Recent elucidation of the energetics of ET via tyrosine
residues in RNR38,39 inspired us to analyze multistep ET in
ADO using semiclassical ET theory (eq 2), which is a typical
starting point for understanding ET reactions in proteins.14
Much work is left to be done in ADO, but the following
analysis is a starting point. The following analysis uses typical
DISCUSSION
■
WT ADO can be readily modified at the surface-exposed
residue C70 using a thiol-reactive ruthenium(II) complex. This
was the only detected labeled product, although there is
another partially exposed cysteine (C116) in the WT protein.
It is surprising that covalent modification at ADO-C70 with a
redox-active molecule does not affect the ability of the enzyme
to turn over under photocatalytic conditions. In contrast,
elimination of this residue and attachment of Ru elsewhere (in
Ru-C62 ADO) impairs the ability of ADO to function.
Generation of a strong Ru reductant near the active site (as in
Ru-C62) is insufficient to give an active enzyme. We note that
ET from both the Ru-C62 site and the Ru-C70 site to the Fe
active site likely involves the peptide backbone that connects
the Ru-label to the Fe-ligated His63. In this context, it is
notable that single-Ru modification at C62 and C70 result in
markedly different abilities of the resulting ADO to produce
formate. This underscores the importance of the ET pathway
involving C70 and suggests that future investigations of
function ET in these Ru-modified proteins must account for
this feature.
ET parameters for proteins: λ = 0.8 eV and HAB
=
HAB0exp(−βr), where HAB = 186 cm−1, β = 1.1 Å−1, and r
is the distance between cofactors.40 HAB is the electronic
coupling strength of the reactants and products at the
transition state, and β is the distance decay constant.40
Additional details about these calculations are given in the
Supporting Information. Given the complexity of the ET chain
in ADO, some limiting cases are presented here to
demonstrate the importance of electron flow through Y and/
or C.
0
The Ru-modified ADO described here is active, but the
activity is low with respect to other assays that employ weaker
reductants than Ru (E° = −1.2 V). Side reactions of Ru are
problematic and account, in part, for this observation. More
DTC is consumed than can be accounted for by the 4e−
stoichiometry required by ADO, which could be due to
adventitious reactions of electronically excited Ru33 or reduced
Ru with O2. In addition, atmospheric concentrations of O2
have been shown to inhibit catalysis by ADO,7,8,34 consistent
with our observed production of peroxides in Ru-ADO under
photocatalytic conditions (see above).
The above observations motivated us to analyze ET
pathways and energetics in ADO and Ru-ADO proteins.
Inspection of the X-ray structures of ADO enzymes shows a
series of tyrosine residues that extend from the iron site to the
surface of the protein, near C70 (e.g., Figure 4). This “chain”
of Y residues is structurally conserved across ADO enzymes
from different organisms35 and is widely conserved in aligned
protein sequences from ADO enzymes from different species
(see Supporting Information). In addition, the chain of redox-
active Tyr residues resembles those that are proposed to play
2
4π3
i
j
y
z
−(ΔG° + λ)
4λkBT
2
j
z
kET
=
HAB expj
z
z
z
j
j
h2λkBT
(2)
k
{
An upper limit for the potentials of activated intermediates
can be approximated as 1.4 V based on the potentials for alkyl
radicals41 or for related Fe model compounds.42 The proton-
coupled potential for C−S·/C−SH at pH 7 is 0.82 V, and that
for Y−O·/Y−OH is 0.83 V. Likewise, E°(C−S·/C−S−) = 0.73
V, and E°(Y−O·/Y−O−) = 0.65 V.41 Thus, the driving force
(−ΔG°) for ET from a surface residue (C or Y) of WT ADO
(unmodified with Ru) to the active site is estimated as less
than or equal to 0.75 V.43 Given the nature of the iron or
organic intermediates, it is unlikely that the net reaction is
thermodynamically uphill, but any individual step could be.
Given a limiting value of −ΔG° = −200 meV,14,38 ET could
proceed at 2 × 105 s−1 with tunneling distances under 8 Å
(using eq 2).
D
DOI: 10.1021/acs.inorgchem.8b00673
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Direct reduction of intermediate alkyl radicals, either by Ru
(E° = −1.2 V) or ferredoxin,10 is unlikely, as there is evidence
that the H· atom required to form the alkane product comes
from the ADO active site.10,29 Ru is a strong reductant, so the
driving force for ET is deeply in the Marcus inverted region,
with kcalc ≪ 1 s−1 (r ≈ 20 Å, as an estimate of the Ru−Fe
distance). In contrast, milder reductants, such as petF
ferredoxin (E°′ = −0.412 V44) or DTC (E°′ ≈ −0.3 V45),
can be orders of magnitude faster simply due to their weaker
reducing power with respect to Ru. In contrast, Ru is predicted
to be a good reductant for the ferric ADO resting state. The
reduction potential for the FeIIIFeIII resting state is not yet
known, but our estimate is between −0.1 and 0.3 V based on
potentials for related diiron enzymes.46−48 Reduction of the
FeIIIFeIII site can have a single-step Ru → Fe ET rate that is at
least as fast as reduction by ferredoxin (400 s−1). A proposed
cycle for Ru-ADO accounting for this ET analysis is set out in
Figure 5.
oxidize ADO active sites or nearby redox-active amino acids.
Our analysis of reductive ET suggests that photochemical ET
to the diferric active site, generating the O2-reactive diferrous
enzyme, can occur readily with Ru as a reductant. Other
intramolecular ET reactions are a possible pathway for ET
from the surface of ADO to activated intermediates in the
native ADO enzymes. Side reactions involving O2 were
identified as challenges in developing an ADO system with
better photochemical activity.
ASSOCIATED CONTENT
* Supporting Information
■
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.inorg-
chem.8b00673.
Detailed ADO expression and purification procedures,
mass spectrometry and circular dichroism data, UV−
visible and luminescence spectra, additional electron
transfer analyses (PDF)
AUTHOR INFORMATION
Corresponding Author
*E-mail: j.warren@sfu.ca
■
ORCID
Jeffrey J. Warren: 0000-0002-1747-3029
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
Simon Fraser Univ., the National Sciences and Engineering
Research Council of Canada (NSERC, RGPIN05559 to
J.J.W.), the Canada Summer Jobs program (Project No.
014504120 to J.J.W.), and the Canadian Institute for Advanced
Research Azrieli Global Scholars and Bio-Inspired Solar Energy
programs (to J.J.W.) supported this work. J.J.M. acknowledges
support from NSERC and SFU for postgraduate fellowships.
H.K.v.d.R. acknowledges support from the NSERC USRA
program. N. Merbouh, S. Grunnert, and P. Mulyk assisted with
GC-MS experiments. The plasmid encoding ADO was a gift
from D. Baker.
Figure 5. Proposed catalytic cycle for electron delivery in Ru-ADO
proteins. The sequential addition of electrons are denoted as e(n)− (n
= 1−4).
There are still many questions about ET in ADO enzymes.
The following suggestions can be made regarding functional
ET in Ru- and WT-ADO based on the above analysis. Ru can
act as a reductant for oxidized ADO, but it is probably not a
suitable reductant for other intermediates. The milder
reduction potential of ferredoxins (such as petF) are more
appropriate to support catalysis. In that context, DTC is a
probable reductant of Ru-ADO intermediates in the present
work. Finally, the comparatively low activity of Ru-C62 ADO
can be attributed to the position of Ru with respect to the
natural Y/C ET chain. During turnover, holes are funneled
away from Ru-C62. ET to reduce any intermediates is in the
inverted region and is calculated to be very slow from position
62 (in comparison to Ru-C70). Thus, to study functional ET
in ADO, it is not sufficient to simply place a photosensitizer/
reductant anywhere near the active site.
In sum, three different Ru-modified ADO proteins were
characterized and evaluated for their ability to decarbonylate
aliphatic aldehyde substrates under photocatalytic conditions.
The order of increasing catalytic reactivity was: Ru-C70 ≈ Ru-
C62/Ru-C70 > Ru-C62. Importantly, the unique reactivity of
wild-type ADO C70 opens the door to new applications or
investigations of electron flow during catalysis, for example,
using time-resolved spectroscopies or electrochemistry. We are
presently investigating the ability of oxidized Ru (Ru(III)) to
REFERENCES
■
(1) Schirmer, A.; Rude, M. A.; Li, X.; Popova, E.; del Cardayre, S. B.
Microbial Biosynthesis of Alkanes. Science 2010, 329, 559−562.
(2) Warui, D. M.; Li, N.; Nørgaard, H.; Krebs, C.; Bollinger, J. M.;
Booker, S. J. Detection of Formate, Rather than Carbon Monoxide, As
the Stoichiometric Coproduct in Conversion of Fatty Aldehydes to
Alkanes by a Cyanobacterial Aldehyde Decarbonylase. J. Am. Chem.
Soc. 2011, 133, 3316−3319.
(3) Lin, F.; Das, D.; Lin, X. N.; Marsh, E. N. G. Aldehyde-forming
fatty acyl-CoA reductase from cyanobacteria: expression, purification
and characterization of the recombinant enzyme. FEBS J. 2013, 280,
4773−4781.
(4) Warui, D. M.; Pandelia, M.-E.; Rajakovich, L. J.; Krebs, C.;
Bollinger, J. M.; Booker, S. J. Efficient Delivery of Long-Chain Fatty
Aldehydes from the Nostoc punctiforme Acyl-Acyl Carrier Protein
Reductase to Its Cognate Aldehyde-Deformylating Oxygenase.
Biochemistry 2015, 54, 1006−1015.
(5) Li, N.; Chang, W.-c.; Warui, D. M.; Booker, S. J.; Krebs, C.;
Bollinger, J. M. Evidence for Only Oxygenative Cleavage of Aldehydes
to Alk(a/e)nes and Formate by Cyanobacterial Aldehyde Decarbon-
ylases. Biochemistry 2012, 51, 7908−7916.
E
DOI: 10.1021/acs.inorgchem.8b00673
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(6) Howard, T. P.; Middelhaufe, S.; Moore, K.; Edner, C.; Kolak, D.
M.; Taylor, G. N.; Parker, D. A.; Lee, R.; Smirnoff, N.; Aves, S. J.;
Love, J. Synthesis of customized petroleum-replica fuel molecules by
targeted modification of free fatty acid pools in Escherichia coli. Proc.
Natl. Acad. Sci. U. S. A. 2013, 110, 7636−7641.
(25) Hayashi, Y.; Yasugi, F.; Arai, M. Role of Cysteine Residues in
the Structure, Stability, and Alkane Producing Activity of
Cyanobacterial Aldehyde Deformylating Oxygenase. PLoS One
2015, 10, e0122217.
(26) McClanahan, S. F.; Dallinger, R. F.; Holler, F. J.; Kincaid, J. R.
Mixed-ligand poly(pyridine) complexes of ruthenium(II). Resonance
Raman spectroscopic evidence for selective population of ligand-
localized 3MLCT excited states. J. Am. Chem. Soc. 1985, 107, 4853−
4860.
́
(7) Kallio, P.; Pasztor, A.; Thiel, K.; Akhtar, M. K.; Jones, P. R. An
engineered pathway for the biosynthesis of renewable propane. Nat.
Commun. 2014, 5, 4731.
(8) Andre, C.; Kim, S. W.; Yu, X.-H.; Shanklin, J. Fusing catalase to
an alkane-producing enzyme maintains enzymatic activity by
converting the inhibitory byproduct H2O2 to the cosubstrate O2.
Proc. Natl. Acad. Sci. U. S. A. 2013, 110, 3191−3196.
(27) Deronzier, A.; Meyer, T. J. Photoinduced reduction of tris(2,2′-
bipyridine)ruthenium(2+) by some dithio anions. Inorg. Chem. 1980,
19, 2912−2917.
(9) Das, D.; Eser, B. E.; Han, J.; Sciore, A.; Marsh, E. N. G. Oxygen-
Independent Decarbonylation of Aldehydes by Cyanobacterial
Aldehyde Decarbonylase: A New Reaction of Diiron Enzymes.
Angew. Chem., Int. Ed. 2011, 50, 7148−7152.
(10) Rajakovich, L. J.; Nørgaard, H.; Warui, D. M.; Chang, W.-c.; Li,
N.; Booker, S. J.; Krebs, C.; Bollinger, J. M.; Pandelia, M.-E. Rapid
Reduction of the Diferric-Peroxyhemiacetal Intermediate in Alde-
hyde-Deformylating Oxygenase by a Cyanobacterial Ferredoxin:
Evidence for a Free-Radical Mechanism. J. Am. Chem. Soc. 2015,
137, 11695−11709.
(11) Buer, B. C.; Paul, B.; Das, D.; Stuckey, J. A.; Marsh, E. N. G.
Insights into Substrate and Metal Binding from the Crystal Structure
of Cyanobacterial Aldehyde Deformylating Oxygenase with Substrate
Bound. ACS Chem. Biol. 2014, 9, 2584−2593.
(12) Wang, C.; Zhao, C.; Hu, L.; Chen, H. Calculated Mechanism of
Cyanobacterial Aldehyde-Deformylating Oxygenase: Asymmetric
Aldehyde Activation by a Symmetric Diiron Cofactor. J. Phys.
Chem. Lett. 2016, 7, 4427−4432.
(28) Balzani, V.; Bolletta, F.; Gandolfi, M. T.; Maestri, M.
Bimolecular electron transfer reactions of the excited states of
transition metal complexes. Top. Curr. Chem. 1978, 75, 1−64.
(29) Waugh, M. W.; Marsh, E. N. G. Solvent Isotope Effects on
Alkane Formation by Cyanobacterial Aldehyde Deformylating
Oxygenase and Their Mechanistic Implications. Biochemistry 2014,
53, 5537−5543.
(30) Khara, B.; Menon, N.; Levy, C.; Mansell, D.; Das, D.; Marsh, E.
N. G.; Leys, D.; Scrutton, N. S. Production of Propane and Other
Short-Chain Alkanes by Structure-Based Engineering of Ligand
Specificity in Aldehyde-Deformylating Oxygenase. ChemBioChem
2013, 14, 1204−1208.
(31) Nilsson, T. Photoinduced electron transfer from tris(2,2′-
bipyridyl)ruthenium to cytochrome c oxidase. Proc. Natl. Acad. Sci. U.
S. A. 1992, 89, 6497−6501.
(32) Aukema, K. G.; Makris, T. M.; Stoian, S. A.; Richman, J. E.;
Munck, E.; Lipscomb, J. D.; Wackett, L. P. Cyanobacterial Aldehyde
̈
Deformylase Oxygenation of Aldehydes Yields n − 1 Aldehydes and
Alcohols in Addition to Alkanes. ACS Catal. 2013, 3, 2228−2238.
(33) Zhang, X.; Rodgers, M. A. J. Energy and Electron Transfer
Reactions of the MLCT State of Ruthenium Tris(bipyridyl) with
Molecular Oxygen: A Laser Flash Photolysis Study. J. Phys. Chem.
1995, 99, 12797−12803.
(13) Giese, B.; Wang, M.; Gao, J.; Stoltz, M.; Muller, P.; Graber, M.
Electron relay race in peptides. J. Org. Chem. 2009, 74, 3621−3625.
(14) Warren, J. J.; Ener, M. E.; Vlcek, A., Jr.; Winkler, J. R.; Gray, H.
B. Electron hopping through proteins. Coord. Chem. Rev. 2012, 256,
2478−2487.
(15) Paul, B.; Das, D.; Ellington, B.; Marsh, E. N. G. Probing the
Mechanism of Cyanobacterial Aldehyde Decarbonylase Using a
Cyclopropyl Aldehyde. J. Am. Chem. Soc. 2013, 135, 5234−5237.
(16) Sullivan, B. P.; Salmon, D. J.; Meyer, T. J. Mixed phosphine
2,2′-bipyridine complexes of ruthenium. Inorg. Chem. 1978, 17,
3334−3341.
(34) Das, D.; Ellington, B.; Paul, B.; Marsh, E. N. G. Mechanistic
Insights from Reaction of α-Oxiranyl-Aldehydes with Cyanobacterial
Aldehyde Deformylating Oxygenase. ACS Chem. Biol. 2014, 9, 570−
577.
(35) Park, A. K.; Kim, I.-S.; Jeon, B. W.; Roh, S. J.; Ryu, M.-Y.; Baek,
H.-R.; Jo, S.-W.; Kim, Y.-S.; Park, H.; Lee, J. H.; Yoon, H.-S.; Kim, H.-
W. Crystal structures of aldehyde deformylating oxygenase from
Limnothrix sp. KNUA012 and Oscillatoria sp. KNUA011. Biochem.
Biophys. Res. Commun. 2016, 477, 395−400.
(36) Gray, H. B.; Winkler, J. R. Hole hopping through tyrosine/
tryptophan chains protects proteins from oxidative damage. Proc. Natl.
Acad. Sci. U. S. A. 2015, 112, 10920−10925.
(37) Minnihan, E. C.; Nocera, D. G.; Stubbe, J. Reversible, Long-
Range Radical Transfer in E. coli Class Ia Ribonucleotide Reductase.
Acc. Chem. Res. 2013, 46, 2524−2535.
(38) Ravichandran, K. R.; Taguchi, A. T.; Wei, Y.; Tommos, C.;
Nocera, D. G.; Stubbe, J. A > 200 meV Uphill Thermodynamic
Landscape for Radical Transport in Escherichia coli Ribonucleotide
Reductase Determined Using Fluorotyrosine-Substituted Enzymes. J.
Am. Chem. Soc. 2016, 138, 13706−13716.
(17) Berg, K.; Tran, A.; Raymond, M.; Abrahamsson, M.; Wolny, J.;
̈
Redon, S.; Andersson, M.; Sun, L.; Styring, S.; Hammarstrom, L.;
Toftlund, H.; Åkermark, B. Covalently Linked Ruthenium(II)-
Manganese(II) Complexes: Distance Dependence of Quenching
and Electron Transfer. Eur. J. Inorg. Chem. 2001, 2001, 1019−1029.
(18) http://web.expasy.org/compute_pi/ (Accessed April 4, 2016).
(19) Jia, C.; Li, M.; Li, J.; Zhang, J.; Zhang, H.; Cao, P.; Pan, X.; Lu,
X.; Chang, W. Structural insights into the catalytic mechanism of
aldehyde-deformylating oxygenases. Protein Cell 2015, 6, 55−67.
(20) Berglund, J.; Pascher, T.; Winkler, J. R.; Gray, H. B.
Photoinduced Oxidation of Horseradish Peroxidase. J. Am. Chem.
Soc. 1997, 119, 2464−2469.
(21) Bjerrum, M. J.; Casimiro, D. R.; Chang, I. J.; Di Bilio, A. J.;
Gray, H. B.; Hill, M. G.; Langen, R.; Mines, G. A.; Skov, L. K.;
Winkler, J. R.; Wuttke, D. S. Electron transfer in ruthenium-modified
proteins. J. Bioenerg. Biomembr. 1995, 27, 295−302.
(22) Tran, N.-H.; Nguyen, D.; Dwaraknath, S.; Mahadevan, S.;
Chavez, G.; Nguyen, A.; Dao, T.; Mullen, S.; Nguyen, T.-A.;
Cheruzel, L. E. An Efficient Light-Driven P450 BM3 Biocatalyst. J.
Am. Chem. Soc. 2013, 135, 14484−14487.
(23) Tran, N.-H.; Huynh, N.; Bui, T.; Nguyen, Y.; Huynh, P.;
Cooper, M. E.; Cheruzel, L. E. Light-initiated hydroxylation of lauric
acid using hybrid P450 BM3 enzymes. Chem. Commun. 2011, 47,
11936−11938.
(24) Tran, N.-H.; Huynh, N.; Chavez, G.; Nguyen, A.; Dwaraknath,
S.; Nguyen, T.-A.; Nguyen, M.; Cheruzel, L. A series of hybrid P450
BM3 enzymes with different catalytic activity in the light-initiated
hydroxylation of lauric acid. J. Inorg. Biochem. 2012, 115, 50−56.
(39) Ravichandran, K. R.; Zong, A. B.; Taguchi, A. T.; Nocera, D.
G.; Stubbe, J.; Tommos, C. Formal Reduction Potentials of
Difluorotyrosine and Trifluorotyrosine Protein Residues: Defining
the Thermodynamics of Multistep Radical Transfer. J. Am. Chem. Soc.
2017, 139, 2994−3004.
(40) Gray, H. B.; Winkler, J. R. Electron tunneling through proteins.
Q. Rev. Biophys. 1999, 36, 341−372.
(41) Warren, J. J.; Tronic, T. A.; Mayer, J. M. Thermochemistry of
Proton-Coupled Electron Transfer Reagents and its Implications.
Chem. Rev. 2010, 110, 6961−7001.
(42) Xue, G.; Wang, D.; De Hont, R.; Fiedler, A. T.; Shan, X.;
Mu
̈
nck, E.; Que, L. A synthetic precedent for the [FeIV2(μ-O)2]
diamond core proposed for methane monooxygenase intermediate Q.
Proc. Natl. Acad. Sci. U. S. A. 2007, 104, 20713−20718.
F
DOI: 10.1021/acs.inorgchem.8b00673
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(43) Note that this is a different driving force than for direct
reduction of the site by an external electron donor (e.g., ferredoxin).
Additional analysis is presented in the Supporting Information.
(44) Bottin, H.; Lagoutte, B. Ferrodoxin and flavodoxin from the
cyanobacterium Synechocystis sp PCC 6803. Biochim. Biophys. Acta,
Bioenerg. 1992, 1101, 48−56.
(45) Gregg, E. C.; Tyler, W. P. Polarography of the Bis-
(diethylthiocarbamyl) DisulfideDiethyldithiocarbamate Ion Oxida-
tionReduction System. J. Am. Chem. Soc. 1950, 72, 4561−4565.
(46) Wang, D. L.; Holz, R. C.; David, S. S.; Que, L.; Stankovich, M.
T. Electrochemical properties of the diiron core of uteroferrin and its
anion complexes. Biochemistry 1991, 30, 8187−8194.
(47) Silva, K. E.; Elgren, T. E.; Que, L.; Stankovich, M. T. Electron
Transfer Properties of the R2 Protein of Ribonucleotide Reductase
from Escherichia coli. Biochemistry 1995, 34, 14093−14103.
(48) With equation 2 and a distance of 18−25 Å for the ADO/petF
precursor complex, these E° estimates are consistent with the known
ET rate of reduction of ADO by petF.10
G
DOI: 10.1021/acs.inorgchem.8b00673
Inorg. Chem. XXXX, XXX, XXX−XXX