This is an open access article published under an ACS AuthorChoice License, which permits
copying and redistribution of the article or any adaptations for non-commercial purposes.

Research Article
Cite This: ACS Catal. 2019, 9, 5837−5846

pubs.acs.org/acscatalysis

Improved Electro- and Photocatalytic Water Reduction by Conﬁned
Cobalt Catalysts in Streptavidin
Arnau Call,†,‡ Carla Casadevall,†,‡ Adrian Romero-Rivera,§ Vlad Martin-Diaconescu,†
Dayn J. Sommer,∥ Sílvia Osuna,§,⊥ Giovanna Ghirlanda,*,∥ and Julio Lloret-Fillol*,†,⊥

Downloaded via 84.88.84.34 on March 9, 2020 at 12:37:01 (UTC).
See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

†

Institute of Chemical Research of Catalonia (ICIQ), The Barcelona Institute of Science and Technology, Avinguda Països Catalans
16, 43007 Tarragona, Spain
§
Institut de Química Computacional i Catàlisi (IQCC) and Departament de Química, Universitat de Girona, Carrer Maria Aurèlia
Capmany 69, 17003 Girona, Spain
∥
School of Molecular Sciences, Arizona State University, Tempe, Arizona 85287-1604, United States
⊥
Catalan Institution for Research and Advanced Studies (ICREA), Passeig Lluïs Companys, 23, 08010, Barcelona, Spain
S
* Supporting Information

ABSTRACT: Incorporation of biotinylated aminopyridine cobalt complexes
derived from the triazacyclononane scaﬀold into the streptavidin protein leads to
formation of artiﬁcial metalloenzymes for water reduction to hydrogen. The
synthesized artiﬁcial metalloenzymes have lower overpotential (at the half-peak
up to 100 mV) and higher photocatalytic hydrogen evolution activity (up to 14and 10-fold increase in TOF and TON, respectively, at pH 12.5) than the free
biotinylated cobalt complexes. 1H-NMR, EPR and XAS highlight the presence of
the metal complexes upon supramolecular attachment to the streptavidin. pHdependent catalytic studies and molecular dynamics (MD) simulations suggest
that the increase in the catalytic activity could be induced by the protein residues
positioned close to the metal centers. These ﬁndings illustrate the ability of the
biotin−streptavidin technology to produce artiﬁcial metalloproteins for photo- and electrocatalytic hydrogen evolution reaction.
KEYWORDS: water reduction to hydrogen, photocatalysis, electrocatalysis, cobalt complexes, streptavidin−biotin

1. INTRODUCTION

coordination sphere suﬀers from signiﬁcant synthetic challenges, although it is crucial to improve the TOF.6,7
In this regard, hybrid natural−artiﬁcial systems could serve
as testing grounds to study the eﬀect of protein scaﬀold
encapsulation and second-sphere interactions in the catalytic
performance of coordination complexes.3b,8 Several artiﬁcial
protein−catalysts hybrids have been developed either by
assembling synthetic catalysts into the protein pocket9 or by
creating biomimetic systems made by metal substitution and/
or mutation.10 One of the most convenient synthetic
approaches builds on supramolecular host−guest interactions,
for example, using streptavidin (SA) as a host protein and
exploiting its high aﬃnity for biotin (also known as vitamin B7
or vitamin H). SA binds biotin with one of the strongest
noncovalent interactions known in nature (dissociation
constant (Kd) on the order of ∼10−14), ensuring quantitative
incorporation of biotinylated molecules.11 In addition, the
biotin−SA complex is resistant to organic solvent, temperature,
and pH. On the basis of a concept ﬁrst reported by
Whitesides12 and further developed by the Ward group, the
biotin−SA technology has been applied to create artiﬁcial

Among the possible alternative fuels, molecular hydrogen
obtained from light-driven water splitting is ideal from an
environmental point of view.1 However, sustainable generation
of hydrogen is extraordinarily challenging and requires a
profound understanding of the diﬀerent reactions involved. In
this context, rational design of catalytic systems serves as a tool
to interrogate strategies to maximize eﬃciency and robustness
as well as to unravel the mechanistic details.2 Likewise, nature
has served as inspiration by studying and mimicking hydrogenase enzymes, which reversibly catalyze the reduction of
protons to molecular hydrogen with high eﬃciency (turnover
frequencies of up to 21 000 s−1).3
The high activity of hydrogenases is ascribed in part to the
control of the geometry of the active site, the second
coordination sphere, and the protection of the catalytic
center.4 However, their characterization and further development for practical applications is a demanding task.
Conversely, catalysts active in H2 evolution based on metal
complexes are tunable5 for improving the catalytic activity and
overpotentials and increasing the oxygen tolerance. However,
most organometallic catalysts are only eﬃcient with strong
acids in organic solvents or they are not suﬃciently robust in
water to be catalytically active. Moreover, control of the second
© 2019 American Chemical Society

Received: December 13, 2018
Revised: April 25, 2019
Published: May 13, 2019
5837

DOI: 10.1021/acscatal.8b04981
ACS Catal. 2019, 9, 5837−5846

Research Article

ACS Catalysis
metalloenzymes for a broad variety of catalytic applications.13
Very recently, this strategy has also been explored in hydrogen
evolution catalysis. It has been observed that the resulting
artiﬁcial hydrogenases are able to drive H2 evolution when
using [Ru(bpy)3]Cl2 as a photosensitizer and ascorbate as a
sacriﬁcial electron donor, albeit with no signiﬁcant improvement over the “naked” catalyst.14
We recently described a new family of robust, well-deﬁned,
and highly tunable aminopyridine cobalt complexes derived
from the triazacyclononane (tacn) scaﬀold for the electro- and
photocatalytic H2 evolution in water.15 In view of these unique
properties and inspired by the previous studies, here we study
the encapsulation of two new biotinylated aminopyridine
cobalt complexes in SA for photo- and electrocatalytic H2
production. We found that encapsulation increases the activity
signiﬁcantly under both sets of conditions. Detailed electroand photochemical pH-dependent hydrogen evolution studies
and molecular dynamics (MD) simulations have been
performed to elucidate the enhanced catalytic activity
observed.

ligands AmidePy2tacn and Amide‑SPPy2tacn was carried out by
acylating the free nitrogen atoms of H Py 2 tacn and
NH2‑SP
Py2tacn, respectively (section S.1.3). 1H NMR spectroscopy, high-resolution electrospray ionization mass spectrometry (HR-MS-ESI), magnetic measurements, and elemental
analysis conﬁrm formation of d7 high-spin (S = 3/2) CoII
complexes (Figures S1−S7). The cyclic voltammogram (CV)
of 1Bio and 1Amide in CH3CN shows a reversible peak ascribable
to the CoII/I couple at −1.09 V (all redox processes are
reported vs SCE unless otherwise notiﬁed) (Figures S10 and
S12). Moreover, the reduction potential for the CoII/I
transition in 2Bio (−1.06 V) is close to that of 2Amide (−1.15
V) (Figures S11 and S13). We note that complex 2Bio exhibits
an additional irreversible wave at −0.71 V, which can be
attributed to reduction of the biotin moiety since it is also
observed in the CV of the free Bio‑SPPy2tacn ligand and in the
biotin (Figures S14−S17).
Incorporation of 1Bio and 2Bio into SA is quantitative, as
indicated by the HABA (2-(4′-hydroxyazobenzene)benzoic
acid) displacement assay.16 The linear proﬁle of the HABA
displacement by the biotinylated complexes indicates that their
aﬃnity for SA is similar to previously reported systems and in
the range of that obtained with D-biotin (i.e., >1010 M−1,
section S.1.5).16 The stoichiometry of the complex incorporated in the synthesized and isolated metalloenzymes 1Bio C SA
and 2Bio C SA was evaluated by measuring the Co
concentration by ICP-MS and the protein concentration by
UV−vis. These results are in concordance with previous
reports, indicating saturation of the binding sites of the SA
employed (SA is a tetramer containing a binding site per unit
of monomer). Moreover, the paramagnetic 1H NMR studies
performed with free SA, 2Bio, and 2Bio C SA indicate that the
main proton resonance features of the cobalt complex are
present with slight modiﬁcations after encapsulation into the
SA and that some of the 1H NMR features of the SA protein
are also altered in the 2Bio C SA sample, consistent with the
supramolecular assembly (Figures S22−S25).
We also studied cobalt K-edge X-ray absorption spectroscopy (XAS) and EPR to provide information about the
electronic and geometric structure of the metal coordination
environment. The Co K-edge XAS spectra of 2Bio and 2Bio C
SA have a comparable proﬁle, suggesting a similar coordination
environment and oxidation state. The rising edges centered
around 7719.3 eV are consistent with a CoII center (Figure
S26).18 Furthermore, the intensity of the pre-edges at 7710.4
eV, of 0.06 normalized units, suggests a centrosymmetric
pseudo-octahedral environment.18a This is further supported
by EPR experiments of 2Bio and 2Bio C SA, which at liquid
nitrogen temperatures show low-spin CoII signals with gx and gy
values centered around a g of 2.2 and gz values approaching 2.0,
similar to previously reported values for elongated distorted
pseudo-octahedral CoII complexes (Figure S27.B).6,19 On the
other hand, the EPR spectra of 1Bio and 1Bio C SA suggest an
axially compressed pseudo-octahedral environment at the CoII
center with gz values centered around a g of 2.3 and gx and gy
values close to 2.0 (Figure S27.A).6,19 At 15 K the EPR spectra
are more informative. The basic EPR features observed in the
[Co(OTf)(TsPy2tacn)](OTf) (2), [Co(Cl)(TsPy2tacn)](Cl)
2Cl, 1Bio and 2Bio series guide one to understand the EPR of
1Bio C SA and 2Bio C SA (Table S2). Interestingly, while the
EPR spectrum of 2Bio C SA is similar to that of 2Bio, in the case
of 1Bio C SA it is essentially diﬀerent than 1Bio. The EPR
spectrum of 2Bio C SA recorded at 15 K presents three

2. RESULTS AND DISCUSSION
We synthesized and studied two new biotinylated cobalt
complexes (CatBio) based on the 1,4-di(picolyl)-1,4,7triazacyclononane (HPy2tacn) moiety by employing two
diﬀerent rigid spacer groups for the covalent bond between
the biotin and the catalytic center (Figure 1)15a in order to
investigate the catalytic activity of the cobalt center when
positioned inside the protein pocket.13f,16 The biotin moiety
was covalently attached to the tacn backbone either via the
secondary nitrogen atom on the tacn (BioPy2tacn) or via a free
amine on the spacer group directly linked to the tacn
(Bio‑SPPy2tacn) through reaction with biotin N-hydroxysuccinimide active ester (NHS-biotin) (Scheme S1).

Figure 1. Biotinylated cobalt complexes and artiﬁcial metalloenzymes
employed in this work.

The biotin-functionalized aminopyridine ligands and complexes [Co(BioPy2tacn)(MeCN)](OTf)2 (1Bio) and [Co(Bio‑SPPy2tacn)(MeCN)](OTf)2 (2Bio) (Figure 1) were
synthetized adapting a general procedure described in the
literature (sections S.1.3 and 1.4).15a,17 The homologous
nonbiotinylated [Co(AmidePy2tacn)(MeCN)](OTf)2 (1Amide)
and [Co(Amide‑SPPy2tacn)(MeCN)](OTf)2 (2Amide) complexes
were also synthesized with the aim to study how the absence of
biotin aﬀects their behavior (Scheme S1). The synthesis of the
5838

DOI: 10.1021/acscatal.8b04981
ACS Catal. 2019, 9, 5837−5846

Research Article

ACS Catalysis
diﬀerent Co center environments as for 2Bio (Figure S33). The
predominant one corresponds to a CoII high-spin center (S =
3/2), and two minor signals can be assigned to low-spin CoII
centers coordinated to a diﬀerent sixth ligand (CH3CN and
H2O). In contrast, the main feature in the EPR spectrum of
1Bio C SA correspond to the rhombic signal with g values
centered at 2.37, 2.07, and 2.03, which is in agreement with
coordination of a water molecule. This correlates with the fact
that the cobalt complex in 1Bio C SA is more internalized into
the protein cavity as judged by the molecular dynamics (MD)
simulations (dicussed below). Therefore, the presence of
mainly a low-spin CoII rhombic signal may suggest that the
protein pocket environment may regulate the binding ligand at
the metal center. In the case of the 2Bio C SA, MD simulations
show that the cobalt complex is exposed to the protein surface,
but it is much more exposed to the reaction media than 1Bio C
SA. For both series of complexes the XAS and EPR
spectroscopic data indicate only minor electronic and geometric environment modiﬁcation of the CatBio after inclusion
into the SA protein. Lastly, incorporation of the cobalt catalyst
proceeds, conserving the SA secondary structure, as assessed
by CD spectroscopy (Figure S21).
We compared the catalytic activity of these artiﬁcial
metalloenzymes with the naked biotinylated cobalt complexes
1Bio and 2Bio in the electrochemical reduction of water to H2.
The electrochemical measurements were performed at pH
ranging from 3.5 to 6 (citrate/NaCl buﬀer, 100 mM). At pH
3.5 the CV of 1Bio showed a high current intensity irreversible
peak with an onset potential at −1.21 V and a maximum at
−1.36 V (overpotential (η) = 0.78 V)20 (Figure 2A, black
trace), indicative of a catalytic process. Under the same
conditions, complex 2Bio showed similar electrochemical
features but shifted to a more negative potential (onset
potential about −1.23 V and a maximum at −1.47 V, η = 0.79
V, Figure 2C). The electrocatalytic wave in both complexes
(1Bio and 2Bio) displays a pH dependence. The currents appear
to follow a nonstandard behavior in the case of 2Bio with a
change in slope as the potential is increased. The electrocatalytic peak current (i c) increases with the proton
concentration (Figures S35 and S36), suggesting that the
rate-determining step of the catalytic process relies on a proton
transfer from water.21 The onset, the peak potential for proton
reduction (Ep), and the half peak (E1/2) of 1Bio shift toward less
negative potentials when increasing the proton concentration,
following a PCET (proton-coupled electron transfer) reaction
(Figure 2 and Figure S46).22 The Nernstian response was
found to be −37 mV per pH unit (Figure 2B), close to
expected for a 2-electron and 1-proton reduction process (−30
mV per pH). This value is qualitatively similar to the −26 mV
per pH unit reported for a Co(Py5) system21b or to the −24
mV per pH unit observed in cobalt diamine systems.21a
Complex 2Bio reveals only a minor dependence of the Ep and
onset potential with the pH (Figure 2D). The molecular
behavior of 1Bio and 2Bio was conﬁrmed by rinsing tests (Figure
S36), which indicates the absence of active species attached to
the electrode surface.23 Furthermore, the catalytic current
shows a ﬁrst-order dependence on the catalyst concentration24
(Figure S37) similarly to what was observed for the parent
complex 2.15a
The electrochemical properties of CatBio C SA were
interrogated as a thin ﬁlm,3a,9c,10a,25 prepared by drop casting
over a glassy carbon electrode and carbon cloth (see sections
S.1.7 and S.1.8), which yields an intense electrocatalytic wave

Figure 2. CV of (A) 1Bio (1 mM, black trace) and 1Bio C SA (thin
ﬁlm, red trace) and (C) 2Bio (1 mM, black trace) and 2Bio C SA (thin
ﬁlm, red trace) (see section S.1.7 for detailed experimental
conditions). Plot of the half intensity of the catalytic peak potential
(Ep) as a function of pH of (B) 1Bio (1 mM) and 1Bio C SA (thin ﬁlm)
and (D) 2Bio (1 mM) and 2Bio C SA (thin ﬁlm).

(Figure S38). For comparison, we also analyzed and compared
the CV of 2Bio C SA (73 μM) in solution, which gives similar
results under the same experimental conditions (Figure S39).
When the cathodic scan was performed in the presence of
naked SA the observed current was negligible at this redox
potential (Figure S38), indicating that the catalytic wave
observed for CatBio C SA is due to the hosted CatBio. The
reduction potential of the reduction peak did not change with
the amount of protein deposited on a carbon cloth electrode,
suggesting that protein−protein interactions possible in the
ﬁlm do not interfere in the redox process (Figure S40). The
catalytic activity of the CatBio C SA adducts display a pH
dependence (Figures S42 and S43) comparable to that
obtained with the CatBio and CatAmide catalysts (Figures S35,
S36, S44−S48, S55, and S56). A clear shift in the redox
potential of the catalytic value (ΔEp, ΔEp/2, and onset
potential) was observed when the CatBio were incorporated
into the host SA. The magnitude of the redox potential shift
when the current intensity is one-half of the value regarding
the current intensity at the maximum of the catalytic peak
(ΔEp/2) varies with the pH value and ranges between 20 and
30 mV for 1Bio and 1Bio C SA (Figure 2B) and between 60 and
100 mV for 2Bio and 2Bio C SA (Figure 2D) (Figures S44−
S46). This result is in sharp contrast with those previously
reported for some artiﬁcial metalloenzymes based on cobalt
complexes.9c,10a,e For instance, incorporating cobaloxime
derivatives into the heme pocket of apo sperm-whale
myoglobin results in a negative shift of about 100 mV,
increasing the overpotential.9c Conversely, the positive
direction of the shift observed for our systems suggests
beneﬁcial interactions of the SA pocket with the catalyst
(presumably the secondary coordination sphere) that reduces
the water reduction overpotential. However, a direct
comparison of current density is not straightforward since
the protein concentration on the electrode is unknown.
To verify the formation of H2 we performed controlled
potential electrolysis (CPE) at −1.4 V during 3 h under static
conditions for 2Bio and 2Bio C SA (Figures S49−S51). In both
5839

DOI: 10.1021/acscatal.8b04981
ACS Catal. 2019, 9, 5837−5846

Research Article

ACS Catalysis

Figure 3. Photoinduced H2 production at diﬀerent pH values for reaction of CatBio and CatBio C SA (5 μM in Co), PSIr (250 μM), TEOA (0.4 M)
in 0.4 mL of CH3CN:H2O (3:7 v/v) containing phosphate buﬀer (100 mM) and NaCl (100 mM) after 4 h irradiation (1100 W·m−2, λ > 400 nm).
Time traces of H2 evolved using (A) 2Bio and (B) 2Bio C SA. (C) TON for 2Bio (gray) and 2Bio C SA (black) and TON2Bio C SA/TON2Bio (red). (D)
Initial TOF (h−1) for 2Bio (gray), 2Bio C SA (black), and TOF2Bio C SA/TOF2Bio (red). (E) TON for 1Bio (gray) and 1Bio C SA (black) and
TON1Bio C SA/TON1Bio (red). (F) Initial TOF (h−1) for 1Bio (gray), 1Bio C SA (black), and TOF1Bio C SA/TOF1Bio (red). Error bars are average of
two duplicates.

biotin interactions appear negligible as there is a linear
correlation between the current intensity and the CatBio
concentration (vide supra). All of these observations together
with the XAS, EPR and catalysis show no coordination of
biotin to the cobalt center and indicate that the ΔEp observed
between CatBio and CatBio C SA arises form second
coordination sphere interactions between the catalytic center
and the protein pocket.
We also investigated whether the catalytic activity of 1Bio and
Bio
2 could be inﬂuenced by protein encapsulation in lightdriven H2 evolution. The catalytic activity of 2Bio and 2Bio C SA
was studied by irradiation (1100 W·m−2 of visible light, λ >
400 nm) of a solution containing the catalysts (5 μM),
[Ir(ppy)2(bpy)]PF6 (PSIr) (250 μM) and TEOA (0.4
M).26,15a Due to the low solubility of PSIr in water, 30% in
volume of MeCN was used as a cosolvent. We veriﬁed the
stability of 2Bio C SA by monitoring its secondary structure by
CD at increasing concentrations of CH3CN. No changes in the
CD were observed up to 50% of cosolvent (Figures S57 and
S58) nor in all of the pH range studied (Figures S59−S62).
Hydrogen evolution was monitored as a function of time by
analysis of aliquots from the reaction cuvette headspace by gas
chromatography at diﬀerent pH values adjusted with
phosphate buﬀer (100 mM). The pH of the buﬀer solutions
was measured and adjusted (when needed) after addition of
TEOA using NaOH or HCl. The catalytic activity of 2Bio raises
from pH 7 to a maximum at about pH 10.5 (TON, 89; TOF,
49·h−1; Figure 3).27 For 2Bio C SA the catalytic activity was
higher than for 2Bio at all pH values, and the maximum in
catalytic activity changes from pH 10.5 to 11.5 (Table S4). A
pH bell-shaped proﬁle is observed for the catalytic activity of
both 2Bio and 2Bio C SA (Figure 3). This bell shape has been
previously observed in photocatalytic systems such as Ru/
ascorbate28 and ﬂuorescein/Et3N.29 The stablished reason for
this catalytic activity pH behavior is that at lower pH the

cases H2 was detected by sampling the headspace of the
electrochemical cell after the CPE and subjecting the sample to
chromatographic analysis (gas chromatography-thermal conductivity detector, GC-TCD). In the case of 2Bio C SA, after an
initial current decrease it remains about constant for the rest of
the experiment with good Faraday yield. In contrast, under the
same CPE conditions, 2Bio yields about 560 TON of H2 with
good Faraday yield, but suﬀers a fast and intense activation
process during the CPE, leading to a higher current density
(Figure S51). This is indicative of the formation of a more
active species. A CV after the CPE with a clean GC electrode
did not show any wave in the CV at the CoII/I redox couple
2Bio. In contrast, after 3 h of CPE the CV of the 2Bio C SA
solution presents a wave, although shifted and signiﬁcantly
reduced.
We hypothesized that the observed diﬀerence in activity
between CatBio and CatBio C SA is due to secondary sphere
interactions with the protein scaﬀold. To rule out other
possibilities, such as coordination of biotin to the cobalt center,
we also studied the electrochemical behavior of model
complexes 1Amide and 2Amide. We found that 1Amide and
2Amide exhibit similar electrochemical behaviors (CoII/I redox
couples and pH behavior) to the CatBio homologous
complexes. In addition, the cyclic voltammograms of 1Amide
and 2Amide in CH3CN shows negligible changes upon addition
of large amounts of esteriﬁed biotin (up to 20 equiv, biotin was
esteriﬁed for solubility reasons, Figures S53 and S54). We also
studied computationally (B3LYP-3D/6−31+G** level of
theory) the possible coordination modes for the N, O, and S
atoms of biotin to the cobalt center in oxidation states I and II.
All optimized structures showed that intramolecular coordination of biotin to the cobalt center is largely endergonic. In
addition, an exchange between a coordinated pyridine (or tacn
backbone) and a N, O, or S atom of biotin is also highly
endergonic (Figures S79−S81 and S84−S86). The bimolecular
5840

DOI: 10.1021/acscatal.8b04981
ACS Catal. 2019, 9, 5837−5846

Research Article

ACS Catalysis
eﬀective electron donor concentration (TEOA) is reduced,
while at higher pH values the catalytic activity decreases due to
the lower concentration of protons available, resulting in the
catalytic activity vs pH in a bell-shaped proﬁle.
The diﬀerence in the catalytic activity between 2Bio and 2Bio
C SA increases at higher pH values. At pH 12.5 the catalytic
activity of 2Bio C SA is 10- and 14-fold higher in terms of TON
and TOF, respectively, than the values obtained for 2Bio.
Moreover, at pH 12.5 the catalytic activity of 2Bio is negligible.
This behavior may be rationalized if it is considered that the
protein protects the cobalt center and alters the eﬀective
available protons for the metal protonation regarding the pH.
Nevertheless, this is a very complex system, and other possible
contributions to the observed behavior cannot be discarded. It
is important to note that protonation of the CoI species to give
CoIII−H is thought to be the rate-limiting step in cobalt-based
catalysts at basic pH values.15a
Although 1Bio shows lower catalytic activity than 2Bio, the
catalytic enhancement by the protein environment is also
remarkable with an enhancement factor in TON at pH 11.5 of
7.3 (1Bio C SA) and 3.6 (2Bio C SA), suggesting a better
protecting eﬀect by the protein pocket due to the absence of
the spacer (Figure 3E and 3F and Table 1). Moreover, control

Figure 4. Traces of the photoinduced H2 production by (from top to
bottom) 2Bio C SA, 1Bio C SA, 2Bio, and 1Bio (5 μM) together with
PSCu (250 μM), Et3N (0.15 M) in 0.4 mL of CH3CN:H2O (3:7 v/v)
after 9 h irradiation (1100 W·m−2, λ > 400 nm).

a high aﬃnity of the protein toward 1Bio and 2Bio. The presence
of a spacer group between the biotin and the catalytic center in
2Bio confers additional ﬂexibility to the ligand, leading to two
main diﬀerent conformations (Figure S70). As in biotin−SA
complex, tryptophan residues (Trp79, Trp92, Trp108) bind
the ligand through CH··· π interactions (the mean distance
between the center of mass of Trp108 and biotin rings is ca.
4.0 Å for both 1Bio and 2Bio, see Figure 5).30
To gain insights into the origin of the improvement in
catalytic activity, we evaluated the conformational dynamics of
the binding pocket in the presence of 1Bio and 2Bio (Figure 5).
Polar residues around the cobalt complex together with the
assistance of water could promote the catalysis. Both 1Bio and
2Bio are exposed to several solvent molecules (ca. 3−5 waters
in 1Bio and 2−4 waters in 2Bio, Figures S69 and S72). In
addition, many polar residues that could act as a proton donor
or activate a water molecule to promote metal protonation are
found close to the ligand (Figure 5). 1Bio shows low ﬂexibility
during the MD simulations, sampling most of the time the
same conformation, which positions the cobalt center near
polar Ser112, Ser88, Glu51, and Asn49 residues (S67 and
S68). In contrast, 2Bio is more ﬂexible and adopts two diﬀerent
conformations during the MD simulations (the bend
conformation highlighted in the square of Figure 5 and
extended conformation of Mon3 in S34A). In the bend
conformation, the metal center is situated close to polar
residues Ser112 and Leu124 (Figure 5). For most of the
simulation time the hydrophobic residue Leu124 in the bend
conformation is near the metal center (Figures 5 and S71),
suggesting that this residue could be mutated to enhance
activity. In contrast, in the extended conformation, the metal
center is close to the polar residues Asn49 and His87 (Figures
S70E and S71). We investigated whether the ﬂexibility of the
metal complexes is restricted upon encapsulation by performing MD simulations for 1Bio and 2Bio in solution. In the
absence of protein, both complexes have a certain ﬂexibility.
However, complex 2Bio, with the longer link, is the most
ﬂexible one, as indicated by the computed root-mean square
deviations (RMSD, see Figure S74). The higher ﬂexibility of
2Bio is also retained upon protein encapsulation (compare
RMSD of 1Bio and 2Bio inside the protein, Figures S75 and
S76), which allows a higher water accessibility, consistent with
the water shell and g(r) analysis (Figure S73).

Table 1. Photocatalytic Activities of 2Bio and 2Bio C SA at pH
11.5a
1Bio
TON
TONCatBio C SA/TONCatBio
TOF
TOFCatBio C SA/TOFCatBio

7
7.3
1.8
5.7

1Bio C SA
51
10.3

2Bio

2Bio C SA

46
3.6
16
5.5

166
88

Cobalt catalyst (5 μM in Co) PSIr (250 μM), TEOA (0.4 M) in 0.4
mL of CH3CN:H2O (3:7 v/v) containing phosphate buﬀer (100
mM) and NaCl (100 mM) (pH 11.5) after 4 h irradiation (1100 W·
m−2, λ > 400 nm).
a

experiments performed generating the metalloenzyme in situ
before catalysis (by mixing 2Bio with SA ∼30 min before the
catalysis) show the same catalytic activity as the presynthesized
and puriﬁed 2Bio C SA (Figure S65). In contrast, addition of
SA to complex 2Amide does not aﬀect its photocatalytic activity
(Figure S66), pointing toward an enhancement in the H2
production activity due to supramolecular interactions in 2Bio
C SA.
We also performed photochemical studies using [Cu(xantphos)(bathocuproine)]PF6 (PSCu) as photoredox catalyst
and triethylamine (Et3N) as electron donor in nonbuﬀered
solutions (Figure 4). Under these catalytic conditions we
found that the H2 evolution catalytic performance for the
metalloproteins (CatBio C SA) is higher than for the CatBio in
terms of both TOF and TON. Also, a remarkably higher
quantity of hydrogen is formed than in the case of using the Ir
photosensitizer (PSIr) in combination with TEOA.
We performed molecular dynamics (MD) simulations of
streptavidin in complex with 1Bio and 2Bio to evaluate their
binding modes and identify which residues play a crucial role
on catalytic enhancement (see SI for computational details).
These calculations do not capture protein−protein interactions
and therefore are only relevant for isolated proteins such as the
conditions employed in this study. Representative conformations sampled along the accumulated 750 ns of MD
simulations (i.e., three replicas of 250 ns for each case) reveal
5841

DOI: 10.1021/acscatal.8b04981
ACS Catal. 2019, 9, 5837−5846

Research Article

ACS Catalysis

Figure 5. Overlay of representative snapshots visited along the MD simulations for 1Bio and 2Bio. Conformational dynamics of the binding pockets
are also shown, where the most important residues surrounding the biotin complexes are represented together with some selected distances (in
Angstroms). Distances have been computed from the metal center and either the side chain of Ser112, Ser88 (oxygen atom, in purple and green,
respectively), and Leu25 (terminal carbon of isobutyl side chain, orange). Trp108−biotin distance is computed between the center of mass of the
indole ring of Trp108 and the tetrahydrothiophene ring of biotin (in blue). Mean distances and standard deviations are also included (in
Angstroms).

Gas Chromatography. Gas chromatography was carried
out on a SRI instruments, model 310C, GC using a 5 Å
molecular sieve column with a thermal conductivity detector
and argon carrier gas.
Nuclear Magnetic Resonance. All NMR spectra were
recorded in CDCl3 on a Varian 400 or 500 MHz instrument, as
noted.
Circular Dichroism. Spectra measurements were carried out
on an Applied Photophysics Chirascan Circular Dichroism
spectrometer equipped with a photomultiplier detector, dualpolarizing prism design monochromator, photoelastic modulator (PEM), and 150 W xenon light source. An average of 3
scans per sample were measured.
X-ray Diﬀraction. Crystals of 1Amide were grown by slow
diﬀusion of Et2O to a CH2Cl2 solution of the complex. The
crystals for these samples were selected using a Zeiss
stereomicroscope using polarized light and prepared under
inert conditions immersed in perﬂuoropolyether as protecting
oil for manipulation. Crystal structure determinations for
1Amide were carried out using a Apex DUO Kappa 4-axis
goniometer equipped with an APPEX 2 4K CCD area
detector, a Microfocus Source E025 IuS using Mo Kα
radiation, Quazar MX multilayer optics as monochromator,
and an Oxford Cryosystems low-temperature device Cryostream 700 plus (T = −173 °C). On the Bruker device, the
following programs were used: Data collection APEX-2,32 data
reduction Bruker Saint33 V/.60A, and absorption correction
SADABS34 or TWINABS.35 The crystal structure solution was
achieved using the computer program SHELXT.36 Visualization was performed with the program SHELXle.37 Missing
atoms were subsequently located from diﬀerence Fourier
synthesis and added to the atom list. Least-squares reﬁnement
on F2 using all measured intensities was carried out using the

3. CONCLUSIONS
In conclusion, we prepared new cobalt artiﬁcial metalloenzymes based on the biotin−streptavidin technology and
able to reduce water to hydrogen under electro- and
photochemical conditions. We found that incorporation of
the cobalt catalysts into the streptavidin scaﬀold results in
several beneﬁcial eﬀects, especially in the 2Bio catalyst. It
reduces the overpotential for electrocatalytic hydrogen
production and increases the light-driven hydrogen production
rate and turnover numbers. This work demonstrates a
straightforward and successful strategy to enhance the catalytic
activity for water reduction to H2. Future studies will be
performed to extend such cobalt artiﬁcial metalloenzymes to
various other photocatalytic chemical transformations.

4. EXPERIMENTAL SECTION
Materials and Reagents. Reagents and solvents were
purchased from commercial sources and used as received
unless otherwise stated. Triethylamine (Et3N, ≥99% purity)
and triethanolamine (TEOA, ≥99% purity) were purchased
from Sigma-Aldrich and used without further puriﬁcation.
Photosensitizer [Ir(bpy)(ppy)2]PF6 (PSIr) was synthesized
according to literature procedures.26a Ethyl [(3aS,4S,6aR)-2oxo-hexahydrothieno[3,4-d]imidazole-4-yl]pentanoate
(COOEtBiotin) was synthesized following the reported method
from the literature.31 Anhydrous acetonitrile was purchased
from Scharlab. Water (18.2 MΩ·cm) was puriﬁed with a MilliQ Millipore Gradient AIS system. All solvents were thoroughly
degassed and stored under anaerobic conditions.
Physical Methods. UV−vis Spectroscopy. UV−vis spectra
were acquired on a Varian Cary 50 Bio Spectrophotometer.
5842

DOI: 10.1021/acscatal.8b04981
ACS Catal. 2019, 9, 5837−5846

Research Article

ACS Catalysis
program SHELXL 2015.38 All non-hydrogen atoms were
reﬁned including anisotropic displacement parameters.
Cyclic Voltammetry. Electrochemical experiments were
carried out on a CH instruments 1242B potentiostat and
with a VSP potentiostat from Bio-Logic, equipped with the
EC-Lab software. The experiments were performed under inert
(Ar) atmosphere in a custom double-wall jacketed singlecompartment cell. For all electrochemical measurements, a
three-electrode system was used: a 3 mm diameter glassy
carbon working electrode with a surface area of 0.28 cm2, a
platinum mesh counter electrode, and a saturated calomel
reference electrode, unless otherwise speciﬁed. Working
electrodes were polished with 1 μM alumina for 5 min,
followed by 10 min of sonication, prior to use. Cyclic
voltammetry was carried out on samples prepared in 100
mM citrate and 25 mM NaCl buﬀer of desired pH (3.5, 4.0,
4.5, 5.0, 5.5, and 6.0); scans were performed at a 100 mV·s−1
scan rate. All potentials are quoted versus SCE. All electrolyte
solutions were degassed by freeze and pump techniques.
Samples were prepared by diluting the catalyst from a stock
solution (10 mM) in water containing 5% DMSO to solubilize
the catalyst. The ﬁnal percentage of DMSO in the electrochemical cell was 0.5%. Electrochemical experiments CatBio C
SA were done by preparing a thin ﬁlm of CatBio C SA over the
glassy carbon electrodes (0.28 cm2 surface area) by drop
casting of 10 μL of a stock solution of metalloprotein of 10 μM
and drying, otherwise indicated. After drying, electrochemical
measurements were performed by directly submerging the
glassy carbon electrode into the solution.
Electron Paramagnetic Resonance. An EMX Micro Xband EPR spectrometer from Bruker was used to collect data
on 0.35 mM solution samples in a mixture of aqueous buﬀer
and MeCN (7:3, buﬀer = Tris·HCl 100 mM, pH 7.5) using a
ﬁnger Dewar at 77 K and He cryostat from Oxford Instruments
at 50 and 15 K. Data was acquired in perpendicular mode with
a modulation frequency of 100 KHz, a modulation amplitude
of 10 G, a 10.24 ms time constant, and 19.02 ms conversion
time with a microwave power was 0.1851 mW. Spectra were
simulated using the EasySpin software package.39
X-ray Absorption Spectroscopy. All NM Samples were
recorded as 0.35 mM solutions in a mixture of aqueous buﬀer
and MeCN (7:3, buﬀer = Tris·HCl 100 mM, pH 7.5) at the
SOLEIL synchrotron (2.75 GeV, 400mA storage ring) SAMBA
Beamline equipped with a helium cryostat (20 K) and using a
Si(220) double-crystal monochromator together with a 36
channel Ge detector. Data calibration and normalization was
carried out using the Athena software package.40 Energies were
calibrated to the ﬁrst inﬂection point of Co foil spectra set at
7709.5 eV. A linear pre-edge function and a quadratic
polynomial for the postedge were used for background
subtraction and normalization of the edge jump.
Photoinduced H2 Evolution. Photoinduced H2 production
irradiation was performed using a 450W xenon lamp with a
400 nm cutoﬀ ﬁlter, irradiating at a constant 1100 W/m2
throughout the experiment. For each experiment, 400 μL of a
reaction mixture containing all of the reagents was added to an
airtight 1 mm cuvette and degassed extensively with argon
prior to illumination. During irradiation time course 100 μL
samples of the headspace were removed with a gastight syringe
and injected directly for analysis by GC. A GC calibration
curve was obtained by injecting various volumes of a 1% H2,
99% N2 gas mixture. The pH of the buﬀer solutions was

measured and adjusted (if needed) after addition of TEOA by
addition NaOH or HCl.
Theoretical Calculations. DFT Calculations. We studied
the coordination ability of the biotinated ligands BioPy2tacn
and Bio‑SPPy2tacn to the metal center and the possibility of
binding the biotin fragment to the metal center in complexes
1Bio and 2Bio computationally by density functional theory
(DFT) using the Gaussian09 program.41 Geometry optimizations were performed in the unrestricted spin formalism, with
the B3LYP hybrid exchange-correlation functional42 and the
standard 6-31+G** 6d basis set for all atoms. An extra
quadratic convergent SCF step was added when the ﬁrst-order
SCF did not converge (“scf = xqc” keyword). The solvation
eﬀect of acetonitrile was introduced in geometry optimizations
and energy through the IEFPCM-SMD polarizable continuum
model.43 Dispersion eﬀects were also included using the
Grimme D3 correction.44 The geometries have been edited
with the Chemcraft program.45 The located stationary points
were characterized by analytical frequency calculations at the
same level of theory as geometry optimizations. Gibbs energy
values (G) were obtained by including thermal, solvation, and
Grimme corrections to the potential energy computed with the
6-31+G** 6d basis set on equilibrium geometries.
Molecular Dynamics Simulations. MD simulations in
explicit water were performed using the AMBER 16 package46
on our in-house GPU cluster Galatea. 1Bio and 2Bio parameters
for the MD simulations were generated within the antechamber
module of AMBER 16 using the general AMBER force ﬁeld
(GAFF).47 We used the bonded model for the Co (quadruplet
state as the lowest in energy) and ligand moiety; in particular,
we used the Seminario method approach to obtain the metal
parameters needed for the simulation as implemented in Prof.
Ryde’s program.48 The optimization, frequencies, and partial
charge were set to ﬁt the electrostatic potential generated at
the UB3LYP/6-31G(d) level by the restrained electrostatic
potential (RESP) model using Gaussian 09. The charges were
calculated according to the Merz−Singh−Kollman scheme. We
docked 1Bio and 2Bio complexes in the biotin pocket using the
pair-ﬁtting tool in Pymol (http://www.pymol.org) using the
biotin moiety of the crystal structure (PDB 1MK5) as a
reference and the optimized structure of the complex creating
the tetramer enzyme. Amino acid protonation states and the
data needed for the electrostatic map were set up using the
Propka3.0 server (http://nbcr-222.ucsd.edu/pdb2pqr_2.0.0/
).49Then the enzyme was solvated in a pre-equilibrated
truncated cuboid box with a 10 Å buﬀer of TIP3P50 water
molecules using the AMBER16 leap module, resulting in the
addition of ca. 9000 solvent molecules. The systems were
neutralized by addition of explicit counterions (Na+ and Cl−).
All subsequent calculations were carried out using the widely
tested Stony Brook modiﬁcation of the Amber 99 force ﬁeld
(ﬀ14SB).51 The structure used was 1MK5 from the protein
data bank (PDB).
A two-stage geometry optimization was performed. The ﬁrst
stage minimizes the positions of solvent molecules and ions
imposing positional restraints on solute by a harmonic
potential with a force constant of 500 kcal mol−1 Å−2, and
the second stage is an unrestrained minimization of all of the
atoms in the simulation cell. The systems are gently heated
using six 50 ps steps, incrementing the temperature 50 K each
step (0−300 K) under constant volume and periodic boundary
conditions. Water molecules were treated with the SHAKE
algorithm such that the angle between the hydrogen atoms is
5843

DOI: 10.1021/acscatal.8b04981
ACS Catal. 2019, 9, 5837−5846

Research Article

ACS Catalysis
kept ﬁxed. Long-range electrostatic eﬀects were modeled using
the particle-mesh-Ewald method.52
A 8 Å cutoﬀ was applied to Lennard−Jones and electrostatic
interactions. Harmonic restraints of 10 kcal·mol−1 were applied
to the solute, and the Langevin equilibration scheme was used
to control and equalize the temperature. The time step was
kept at 1 fs during the heating stages, allowing potential
inhomogeneities to self-adjust. Each system was then
equilibrated without restrains for 2 ns with a 2 fs time step
at a constant pressure of 1 atm and temperature of 300 K. After
the systems were equilibrated in the NPT ensemble, 3
independent 250 ns MD simulations for 1Bio and 2Bio (i.e.,
750 μs accumulated) were performed under the NVT
ensemble and periodic-boundary conditions.

■

Supercomputing Center-Centro Nacional de Supercomputación.

■

(1) (a) Barber, J.; Tran, P. D. From Natural to Artificial
Photosynthesis. J. R. Soc., Interface 2013, 10, 20120984. (b) Lewis,
N. S.; Nocera, D. G. Powering the Planet: Chemical Challenges in
Solar Energy Utilization. Proc. Natl. Acad. Sci. U. S. A. 2006, 103,
15729−15735.
(2) (a) McKone, J. R.; Marinescu, S. C.; Brunschwig, B. S.; Winkler,
J. R.; Gray, H. B. Earth-abundant Hydrogen Evolution Electrocatalysts. Chem. Sci. 2014, 5, 865−878. (b) Du, P.; Eisenberg, R.
Catalysts Made of Earth-abundant Elements (Co, Ni, Fe) for Water
Splitting: Recent Progress and Future Challenges. Energy Environ. Sci.
2012, 5, 6012−6021. (c) Artero, V.; Chavarot-Kerlidou, M.;
Fontecave, M. Splitting Water with Cobalt. Angew. Chem., Int. Ed.
2011, 50, 7238−7266. (d) Le Goff, A.; Artero, V.; Jousselme, B.;
Tran, P. D.; Guillet, N.; Metaye, R.; Fihri, A.; Palacin, S.; Fontecave,
M. From Hydrogenases to Noble Metal-free Catalytic Nanomaterials
for H2 Production and Uptake. Science 2009, 326, 1384−1387.
(3) (a) Lubitz, W.; Ogata, H.; Rudiger, O.; Reijerse, E. Hydrogenases. Chem. Rev. 2014, 114, 4081−4148. (b) Simmons, T. R.;
Berggren, G.; Bacchi, M.; Fontecave, M.; Artero, V. Mimicking
Hydrogenases: From Biomimetics to Artificial Enzymes. Coord. Chem.
Rev. 2014, 270−271, 127−150. (c) Tard, C.; Pickett, C. J. Structural
and Functional Analogues of the Active Sites of the [Fe]-, [NiFe]-,
and [FeFe]-Hydrogenases. Chem. Rev. 2009, 109, 2245−2274.
(4) (a) Mulder, D. W.; Shepard, E. M.; Meuser, J. E.; Joshi, N.; King,
P. W.; Posewitz, M. C.; Broderick, J. B.; Peters, J. W. Insights into
[FeFe]-hydrogenase Structure, Mechanism, and Maturation. Structure
2011, 19, 1038−1052. (b) Knörzer, P.; Silakov, A.; Foster, C. E.;
Armstrong, F. A.; Lubitz, W.; Happe, T. Importance of the Protein
Framework for Catalytic Activity of [FeFe]-hydrogenases. J. Biol.
Chem. 2012, 287, 1489−1499. (c) Berggren, G.; Adamska, A.;
Lambertz, C.; Simmons, T. R.; Esselborn, J.; Atta, M.; Gambarelli, S.;
Mouesca, J. M.; Reijerse, E.; Lubitz, W.; Happe, T.; Artero, V.;
Fontecave, M. Aiomimetic Assembly and Activation of [FeFe]hydrogenases. Nature 2013, 499, 66−69. (d) Esselborn, J.; Lambertz,
C.; Adamska-Venkatesh, A.; Simmons, T.; Berggren, G.; Noth, J.;
Siebel, J.; Hemschemeier, A.; Artero, V.; Reijerse, E.; Fontecave, M.;
Lubitz, W.; Happe, T. Spontaneous Activation of [FeFe]-hydrogenases by an Inorganic [2Fe] Active Site Mimic. Nat. Chem. Biol.
2013, 9, 607−609.
(5) (a) Zee, D. Z.; Chantarojsiri, T.; Long, J. R.; Chang, C. J. Metalpolypyridyl Catalysts for Electro- and Photochemical Reduction of
Water to Hydrogen. Acc. Chem. Res. 2015, 48, 2027−2036.
(b) Queyriaux, N.; Jane, R. T.; Massin, J.; Artero, V.; ChavarotKerlidou, M. Recent Developments in Hydrogen Evolving Molecular
Cobalt(II)-polypyridyl Catalysts. Coord. Chem. Rev. 2015, 304−305,
3−19.
(6) (a) Jian, J. X.; Liu, Q.; Li, Z. J.; Wang, F.; Li, X. B.; Li, C. B.; Liu,
B.; Meng, Q. Y.; Chen, B.; Feng, K.; Tung, C. H.; Wu, L. Z. Chitosan
Confinement Enhances Hydrogen Photogeneration from a Mimic of
the Diiron Subsite of [FeFe]-hydrogenase. Nat. Commun. 2013, 4, 1−
9. (b) Singleton, M. L.; Reibenspies, J. H.; Darensbourg, M. Y. A
Cyclodextrin Host/Guest Approach to a Hydrogenase Active Site
Biomimetic Cavity. J. Am. Chem. Soc. 2010, 132, 8870−8871.
(7) (a) Orain, C.; Quentel, F.; Gloaguen, F. Photocatalytic
Hydrogen Production Using Models of the Iron-iron Hydrogenase
Active Site Dispersed in Micellar Solution. ChemSusChem 2014, 7,
638−643. (b) Pullen, S.; Fei, H.; Orthaber, A.; Cohen, S. M.; Ott, S.
Enhanced Photochemical Hydrogen Production by a Molecular
Diiron Catalyst Incorporated into a Metal-organic Framework. J. Am.
Chem. Soc. 2013, 135, 16997−17003. (c) Yu, T.; Zeng, Y.; Chen, J.;
Li, Y. Y.; Yang, G.; Li, Y. Exceptional Dendrimer-based Mimics of
Diiron Hydrogenase for the Photochemical Production of Hydrogen.
Angew. Chem., Int. Ed. 2013, 52, 5631−5635.
(8) (a) Faiella, M.; Roy, A.; Sommer, D.; Ghirlanda, G. De
NovoDesign of Functional Proteins: Toward Artificial Hydrogenases.

ASSOCIATED CONTENT

S
* Supporting Information

The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acscatal.8b04981.

■

REFERENCES

Synthesis and characterization of ligands, catalysts, and
SA adducts; determination of electro- and photocatalytic
performance; DFT and molecular dynamic simulations
(PDF)

AUTHOR INFORMATION

Corresponding Authors

*E-mail: gghirlanda@asu.edu
*E-mail: jlloret@iciq.es
ORCID

Arnau Call: 0000-0003-0970-7195
Sílvia Osuna: 0000-0003-3657-6469
Julio Lloret-Fillol: 0000-0002-4240-9512
Author Contributions
‡

A.C. and C.C.: These authors ontributed equally.

Notes

The authors declare no competing ﬁnancial interest.

■

ACKNOWLEDGMENTS
We thank the European Commission for the ERC-CG-2014648304 (J.Ll.-F) and ERC-2015-StG-679001 (S.O) projects.
The Spanish Ministry of Science is acknowledged for a FPU
fellowship to A.C. (AP2012-6436) and C.C. and the
Generalitat de Catalunya for a Ph.D. fellowship to A.R.R.
(2015-FI-B-00165). We also thank Catexel for a generous gift
of tritosyl-1,4,7-triazacyclononane. Financial support from the
ICIQ Foundation and CELLEX Foundation through the
CELLEX-ICIQ high-throughput experimentation platform and
the Starting Career Program is gratefully acknowledged. We
also thank the CERCA Programme (Generalitat de Catalunya)
for ﬁnancial support and MINECO (Severo Ochoa Excellence
Accreditation 2014−2018; SEV-2013-0319) and MINECO
project CTQ2016-80038-R. S.O. thanks the Spanish MINECO
CTQ2016-80038-R, Ramón y Cajal contract (RYC-201416846), the European Community for CIG project (PCIG14GA-2013-630978). This work was supported in part by NSF
award 1508301 to G.G. We acknowledge SOLEIL and
DIAMOND for provision of synchrotron radiation facilities,
and we thank Dr. Gautier Landrot for assistance in using
beamline SAMBA. We are grateful for the computer resources,
technical expertise, and assistance provided by the Barcelona
5844

DOI: 10.1021/acscatal.8b04981
ACS Catal. 2019, 9, 5837−5846

Research Article

ACS Catalysis
Biopolymers 2013, 100, 558−571. (b) Bos, J.; Browne, W. R.;
Driessen, A. J. M.; Roelfes, G. Supramolecular Assembly of Artificial
Metalloenzymes Based on the Dimeric Protein LmrR as Promiscuous
Scaffold. J. Am. Chem. Soc. 2015, 137, 9796−9799.
(9) (a) Soltau, S. R.; Niklas, J.; Dahlberg, P. D.; Poluektov, O. G.;
Tiede, D. M.; Mulfort, K. L.; Utschig, L. M. Aqueous Light Driven
Hydrogen Production by a Ru−ferredoxin−Co Biohybrid. Chem.
Commun. 2015, 51, 10628−10631. (b) Onoda, A.; Kihara, Y.;
Fukumoto, K.; Sano, Y.; Hayashi, T. Photoinduced Hydrogen
Evolution Catalyzed by a Synthetic Diiron Dithiolate Complex
Embedded within a Protein Matrix. ACS Catal. 2014, 4, 2645−2648.
(c) Bacchi, M.; Berggren, G.; Niklas, J.; Veinberg, E.; Mara, M. W.;
Shelby, M. L.; Poluektov, O. G.; Chen, L. X.; Tiede, D. M.; Cavazza,
C.; Field, M. J.; Fontecave, M.; Artero, V. Cobaloxime-based Artificial
Hydrogenases. Inorg. Chem. 2014, 53, 8071−8082. (d) Silver, S. C.;
Niklas, J.; Du, P.; Poluektov, O. G.; Tiede, D. M.; Utschig, L. M.
Protein Delivery of a Ni Catalyst to Photosystem I for Light-driven
Hydrogen Production. J. Am. Chem. Soc. 2013, 135, 13246−13249.
(e) Berggren, G.; Adamska, A.; Lambertz, C.; Simmons, T. R.;
Esselborn, J.; Atta, M.; Gambarelli, S.; Mouesca, J. M.; Reijerse, E.;
Lubitz, W.; Happe, T.; Artero, V.; Fontecave, M. Biomimetic
Assembly and Activation of [FeFe]-hydrogenases. Nature 2013,
499, 66−69. (f) Utschig, L. M.; Silver, S. C.; Mulfort, K. L.; Tiede, D.
M. Nature-driven Photochemistry for Catalytic Solar Hydrogen
Production: a Photosystem I-transition Metal Catalyst Hybrid. J. Am.
Chem. Soc. 2011, 133, 16334−16337.
(10) (a) Kandemir, B.; Chakraborty, S.; Guo, Y.; Bren, K. L.
Semisynthetic and Biomolecular Hydrogen Evolution Catalysts. Inorg.
Chem. 2016, 55, 467−477. (b) Slater, J. W.; Shafaat, H. S. Nickelsubstituted Rubredoxin as a Minimal Enzyme Model for Hydrogenase. J. Phys. Chem. Lett. 2015, 6, 3731−3736. (c) Sommer, D. J.;
Vaughn, M. D.; Clark, B. C.; Tomlin, J.; Roy, A.; Ghirlanda, G.
Reengineering Cyt b for Hydrogen Production: A facile Route to
Artificial Hydrogenases. Biochim. Biophys. Acta, Bioenerg. 2016, 1857,
598−603. (d) Sommer, D. J.; Vaughn, M. D.; Ghirlanda, G. Protein
Secondary-shell Interactions Enhance the Photoinduced Hydrogen
Production of Cobalt Protoporphyrin IX. Chem. Commun. 2014, 50,
15852−15855. (e) Kleingardner, J. G.; Kandemir, B.; Bren, K. L.
Hydrogen Evolution from Neutral Water under Aerobic Conditions
Catalyzed by Cobalt Microperoxidase-11. J. Am. Chem. Soc. 2014,
136, 4−7. (f) Saint-Martin, P.; Lespinat, P. A.; Fauque, G.; Berlier, Y.;
Legall, J.; Moura, I.; Teixeira, M.; Xavier, A. V.; Moura, J. J. G.
Hydrogen Production and Deuterium-proton Exchange Reactions
Catalyzed by Desulfovibrio Nickel(II)-substituted Rubredoxins. Proc.
Natl. Acad. Sci. U. S. A. 1988, 85, 9378−9380.
(11) Weber, P. C.; Ohlendorf, D. H.; Wendoloski, J. J.; Salemme, F.
R. Structural Origins of High-affinity Biotin Binding to Streptavidin.
Science 1989, 243, 85−88.
(12) Wilson, M. E.; Whitesides, G. M. Conversion of a Protein to a
Homogeneous Asymmetric Hydrogenation Catalyst by Site-specific
Modification with a Diphosphinerhodium(I) Moiety. J. Am. Chem.
Soc. 1978, 100, 306−307.
(13) (a) Liang, A. D.; Serrano-Plana, J.; Peterson, R. L.; Ward, T. R.
Artificial Metalloenzymes Based on the Biotin−Streptavidin Technology: Enzymatic Cascades and Directed Evolution. Acc. Chem. Res.
2019, 52, 585. (b) Schwizer, F.; Okamoto, Y.; Heinisch, T.; Gu, Y.;
Pellizzoni, M. M.; Lebrun, V.; Reuter, R.; Kohler, V.; Lewis, J. C.;
Ward, T. R. Artificial Metalloenzymes: Reaction Scope and
Optimization Strategies. Chem. Rev. 2018, 118, 142−231. (c) Heinisch, T.; Ward, T. R. Artificial Metalloenzymes Based on the BiotinStreptavidin Technology: Challenges and Opportunities. Acc. Chem.
Res. 2016, 49, 1711−1721. (d) Chatterjee, A.; Mallin, H.; Klehr, J.;
Vallapurackal, J.; Finke, A. D.; Vera, L.; Marsh, M.; Ward, T. R. An
Enantioselective Artificial Suzukiase Based on the Biotin−streptavidin
Technology. Chem. Sci. 2016, 7, 673−677. (e) Dundas, C. M.;
Demonte, D.; Park, S. Streptavidin-biotin Technology: Improvements
and Innovations in Chemical and Biological Applications. Appl.
Microbiol. Biotechnol. 2013, 97, 9343−9353. (f) Ward, T. R. Artificial
Metalloenzymes Based on the Biotin-Avidin Technology: Enantiose-

lective Catalysis and Beyond. Acc. Chem. Res. 2011, 44, 47−57.
(g) Letondor, C.; Ward, T. R. Artificial Metalloenzymes for
Enantioselective Catalysis: Recent Advances. ChemBioChem 2006, 7,
1845−1852.
(14) Keller, S. G.; Probst, B.; Heinisch, T.; Alberto, R.; Ward, T. R.
Photo-driven Hydrogen Evolution by an Artificial Hydrogenase
Utilizing the Biotin-Streptavidin Technology. Helv. Chim. Acta
2018, 101, e1800036.
(15) (a) Call, A.; Codolà, Z.; Acuña-Parés, F.; Lloret-Fillol, J. Photoand Electrocatalytic H2 Production by New First-row Transitionmetal Complexes Based on an Aminopyridine Pentadentate Ligand.
Chem. - Eur. J. 2014, 20, 6171−6183. (b) Call, A.; Franco, F.;
Kandoth, N.; Fernández, S.; González-Béjar, M.; Pérez-Prieto, J.; Luis,
J. M.; Lloret-Fillol, J. Understanding Light-driven H2 Evolution
through the Electronic Tuning of Aminopyridine Cobalt Complexes.
Chem. Sci. 2018, 9, 2609−2619.
(16) Skander, M.; Humbert, N.; Collot, J.; Gradinaru, J.; Klein, G.;
Loosli, A.; Sauser, J.; Zocchi, A.; Gilardoni, F.; Ward, T. R. Artificial
Metalloenzymes: (Strept)avidin as Host for Enantioselective Hydrogenation by Achiral Biotinylated Rhodium-diphosphine Complexes. J.
Am. Chem. Soc. 2004, 126, 14411−14418.
(17) Wilson, M. E.; Nuzzo, R. G.; Whitesides, G. M. Bis(2diphenylphosphinoethyl)amine. A flexible Synthesis of Functionalized
Chelating Diphosphines. J. Am. Chem. Soc. 1978, 100, 2269−2270.
(18) (a) Planas, O.; Whiteoak, C. J.; Martin-Diaconescu, V.; Gamba,
I.; Luis, J. M.; Parella, T.; Company, A.; Ribas, X. Isolation of Key
Organometallic Aryl-Co(III) Intermediates in Cobalt-Catalyzed
C(sp2)−H Functionalizations and New Insights into Alkyne
Annulation Reaction Mechanisms. J. Am. Chem. Soc. 2016, 138,
14388−14397. (b) Planas, O.; Roldán-Gómez, S.; Martin-Diaconescu,
V.; Luis, J. M.; Company, A.; Ribas, X. Mechanistic Insights into the
SN2-type Reactivity of Aryl-Co(III) Masked-carbenes for C−C Bond
Forming Transformations. Chem. Sci. 2018, 9, 5736−5746. (c) Sarangi, R.; Cho, J.; Nam, W.; Solomon, E. I. XAS and DFT Investigation
of Mononuclear Cobalt(III) Peroxo Complexes: Electronic Control of
the Geometric Structure in CoO2 versus NiO2 Systems. Inorg. Chem.
2011, 50, 614−620. (d) Fujita, E.; Furenlid, L. R.; Renner, M. W.
Direct XANES Evidence for Charge Transfer in Co−CO2 Complexes.
J. Am. Chem. Soc. 1997, 119, 4549−4550.
(19) Nishida, Y.; Ida, K.; Kida, S. Investigation on Low-spin
Cobalt(II) Complexes.X. [1] An ESR Study on Six-coordinate
Complexes. Inorg. Chim. Acta 1980, 38, 113−116.
(20) (a) The overpotential is deﬁned as the diﬀerence between the
applied potential and the thermodynamic potential for the reduction
of protons at given conditions. The overpotential is deﬁned as the
diﬀerence between the applied potential and the thermodinamic
potential for the reduction of protons at given conditions. (b) The
thermodyinamic potential for the electrocatalytic water reduction to
H2 at given pH (SCE, 1 atm of H2) is E(pH) = −0.241 V − 0.059 V ×
pH. The thermodynamic potential for the electrocatalytic water
reduction to H2 at given pH (SCE, 1 atm of H2) is E(pH) = −0.241 V
− 0.059 V × pH.
(21) (a) McCrory, C. C.; Uyeda, C.; Peters, J. C. Electrocatalytic
Hydrogen Evolution in Acidic Water with Molecular Cobalt
Tetraazamacrocycles. J. Am. Chem. Soc. 2012, 134, 3164−3170.
(b) Sun, Y.; Bigi, J. P.; Piro, N. A.; Tang, M. L.; Long, J. R.; Chang, C.
J. Molecular Cobalt Pentapyridine Catalysts for Generating Hydrogen
from Water. J. Am. Chem. Soc. 2011, 133, 9212−9215. (c) McNamara,
W. R.; Han, Z.; Yin, C.-J. M.; Brennessel, W. W.; Holland, P. L.;
Eisenberg, R. Cobalt-dithiolene Complexes for the Photocatalytic and
Electrocatalytic Reduction of Protons in Aqueous Solutions. Proc.
Natl. Acad. Sci. U. S. A. 2012, 109, 15594−15599. (d) Abe, T.;
Kaneko, M. pH-dependent Electrocatalysis for Proton Reduction by
Bis(2,2′:6′,2″-terpyridine) Cobalt(II) Complex Embedded in Nafion® Membrane. J. Mol. Catal. A: Chem. 2001, 169, 177−183.
(e) Slater, J. W.; Marguet, S. C.; Monaco, H. A.; Shafaat, H. S. Going
beyond Structure: Nickel-Substituted Rubredoxin as a Mechanistic
Model for the [NiFe] Hydrogenases. J. Am. Chem. Soc. 2018, 140,
10250−10262.
5845

DOI: 10.1021/acscatal.8b04981
ACS Catal. 2019, 9, 5837−5846

Research Article

ACS Catalysis
(22) (a) Slattery, S. J.; Blaho, J. K.; Lehnes, J.; Goldsby, K. A. pHDependent Metal-based Redox Couples as Models for Protoncoupled Electron Transfer Reactions. Coord. Chem. Rev. 1998, 174,
391−416. (b) Save′ant, J.-M. Elements of Molecular and Biomolecular
Electrochemistry: An Electrochemical Approach to Electron Transfer
Chemistry; Wiley-Interscience: Hoboken, NJ, 2006. (c) Horvath, S.;
Fernandez, L. E.; Appel, A. M.; Hammes-Schiffer, S. pH-dependent
Reduction Potentials and Proton-coupled Electron Transfer Mechanisms in Hydrogen-producing Nickel Molecular Electrocatalysts.
Inorg. Chem. 2013, 52, 3643−3652.
(23) Anxolabéhère-Mallart, E.; Costentin, C.; Fournier, M.; Nowak,
S.; Robert, M.; Saveant, J. M. Boron-capped Tris(glyoximato) Cobalt
Clathrochelate as a Precursor for the Electrodeposition of Nanoparticles Catalyzing H2 Evolution in Water. J. Am. Chem. Soc. 2012,
134, 6104−6107.
(24) Bigi, J. P.; Hanna, T. E.; Harman, W. H.; Chang, A.; Chang, C.
J. Electrocatalytic Reduction of Protons to Hydrogen by a Watercompatible Cobalt Polypyridyl Platform. Chem. Commun. 2010, 46,
958−960.
(25) Léger, C.; Elliott, S. J.; Hoke, K. R.; Jeuken, L. J. C.; Jones, A.
K.; Armstrong, F. A. Enzyme Electrokinetics: using Protein Film
Voltammetry to Investigate Redox Enzymes and their Mechanisms.
Biochemistry 2003, 42, 8653−8662.
(26) (a) Cline, E. D.; Adamson, S. E.; Bernhard, S. Homogeneous
Catalytic System for Photoinduced Hydrogen Production Utilizing
Iridium and Rhodium Complexes. Inorg. Chem. 2008, 47, 10378−
10388. (b) Curtin, P. N.; Tinker, L. L.; Burgess, C. M.; Cline, E. D.;
Bernhard, S. Structure-activity Correlations among Iridium(III)
Photosensitizers in a Robust Water-reducing System. Inorg. Chem.
2009, 48, 10498−10506. (c) Hanss, D.; Freys, J. C.; Bernardinelli, G.;
Wenger, O. S. Cyclometalated Iridium(III) Complexes as Photosensitizers for Long-range Electron Transfer: Occurrence of a
Coulomb Barrier. Eur. J. Inorg. Chem. 2009, 2009, 4850−4859.
(d) Ladouceur, S.; Fortin, D.; Zysman-Colman, E. Role of
Substitution on the Photophysical Properties of 5,5′-diaryl-2,2′bipyridine (bpy*) in [Ir(ppy)2(bpy*)]PF6 Complexes: a Combined
Experimental and Theoretical Study. Inorg. Chem. 2010, 49, 5625−
5641.
(27) Simond, M. R.; Ballerat-Busserolles, K.; Coulier, Y.; Rodier, L.;
Coxam, J. Y. Dissociation Constants of Protonated Amines in Water
at Temperatures from 293.15 to 343.15 K. J. Solution Chem. 2012, 41,
130−142.
(28) Natali, M. Elucidating the Key Role of pH on Light-Driven
Hydrogen Evolution by a Molecular Cobalt Catalyst. ACS Catal.
2017, 7, 1330−1339.
(29) Han, Z.; McNamara, W. R.; Eum, M. S.; Holland, P. L.;
Eisenberg, R. A nickel thiolate catalyst for the long-lived photocatalytic production of hydrogen in a noble-metal-free system. Angew.
Chem., Int. Ed. 2012, 51, 1667−1670.
(30) (a) Stayton, P. S.; Freitag, S.; Klumb, L. A.; Chilkoti, A.; Chu,
V.; Penzotti, J. E.; To, R.; Hyre, D.; Le Trong, I.; Lybrand, T. P.;
Stenkamp, R. E. Streptavidin−biotin Binding Energetics. Biomol. Eng.
1999, 16, 39−44. (b) DeChancie, J.; Houk, K. N. The Origins of
Femtomolar Protein-ligand Binding: Hydrogen-bond Cooperativity
and Desolvation Energetics in the Biotin-(Strept)avidin Binding Site.
J. Am. Chem. Soc. 2007, 129, 5419−5429. (c) The structure used was
taken from the protein data bank (PDB); 1MK5.
(31) Schiller, T. L.; Keddie, D. J.; Blakey, I.; Fredericks, P. M.
Surface-enhanced Raman Encoded Polymer Stabilized Gold Nanoparticles: Demonstration of Potential for Use in Bioassays. Eur. Polym.
J. 2017, 87, 508−518.
(32) Data collection with APEX II, version v2013, 4-1; Bruker AXS
Inc.: Madison, WI, 2007.
(33) Data reduction with Bruker SAINT, version V8.30c; Bruker AXS
Inc.: Madison, WI, 2007.
(34) SADABS, V2012/1; Bruker AXS Inc.: Madison, WI,
2001.Blessing, R. H. An empirical correction for absorption
anisotropy. Acta Crystallogr. 1995, A51, 33-38.

(35) TWINABS, Version 2012/1; Bruker AXS scaling for twinned
crystals: Blessing, R. H. An empirical correction for absorption
anisotropy. Acta Crystallogr., Sect. A: Found. Crystallogr. 1995, A51,
33−38.
(36) SHELXT; V2014/4, 2014: Sheldrick, G. M. SHELXTintegrated space-group and crystal determination. Acta Crystallogr.,
Sect. A: Found. Adv. 2015, A71, 3−8.
(37) SHELXle: Hübschle, C. B.; Sheldrick, G. M.; Dittrich, B.
ShelXle: a Qt graphical user interface for SHELXL. J. Appl. Crystallogr.
2011, 44, 1281−1284.
(38) SHELXL; SHELXL-2014/7, 2014: Sheldrick, G. M. SHELXTintegrated space-group and crystal determination. Acta Crystallogr.,
Sect. A: Found. Adv. 2015, A71, 3−8.
(39) Stoll, S.; Schweiger, A. EasySpin, a Comprehensive Software
Package for Spectral Simulation and Analysis in EPR. J. Magn. Reson.
2006, 178, 42−55.
(40) (a) Newville, M. IFEFFIT: Interactive XAFS Analysis and
FEFF Fitting. J. Synchrotron Radiat. 2001, 8, 322−324. (b) Ravel, B.;
Newville, M. ATHENA, ARTEMIS, HEPHAESTUS: Data Analysis
for X-ray Absorption Spectroscopy using IFEFFIT. J. Synchrotron
Radiat. 2005, 12, 537−541.
(41) Gaussian 09, Revision E.01; Gaussian, Inc.: Wallingford, CT,
2009.
(42) (a) Becke, A. D. J. Chem. Phys. 1993, 98, 1372−1377.
(b) Becke, A. D. J. Chem. Phys. 1993, 98, 5648. (c) Lee, C. T.; Yang,
W. T.; Parr, R. G. Phys. Rev. B: Condens. Matter Mater. Phys. 1988, 37,
785.
(43) Marenich, A. V.; Cramer, C. J.; Truhlar, D. G. Universal
Solvation Model Based on Solute Electron Density and on a
Continuum Model of the Solvent Defined by the Bulk Dielectric
Constant and Atomic Surface Tensions. J. Phys. Chem. B 2009, 113,
6378−6396.
(44) Grimme, S. Semiempirical GGA-type Density Functional
Constructed with a Long-range Dispersion Correction. J. Comput.
Chem. 2006, 27, 1787−1799.
(45) www.chemcraftprog.com.
(46) Case, D. A.; Darden, T. A.; Cheatham, T. E.; Simmerling, C. L.;
Wang, J.; Duke, R. E.; Luo, R.; Crowley, M.; Walker, R. C.; Zhang,
W.; Merz, K. M.; Wang, B.; Hayik, S.; Roitberg, A.; Seabra, G.;
Kolossváry, I.; Wong, K. F.; Paesani, F.; Vanicek, J.; Wu, X.; Brozell, S.
R.; Steinbrecher, T.; Gohlke, H.; Yang, L.; Tan, C.; Mongan, J.;
Hornak, V.; Cui, G.; Mathews, D. H.; Seetin, M. G.; Sagui, C.; Babin,
V.; Kollman, P. A. AMBER 16; University of California: San
Francisco, CA, 2016.
(47) Wang, J.; Wolf, R. M.; Caldwell, J. W.; Kollman, P. A.; Case, D.
A. Development and Testing of a General Amber Force Field. J.
Comput. Chem. 2004, 25, 1157−1174.
(48) Hu, L.; Ryde, U. Comparison of Methods to Obtain ForceField Parameters for Metal Sites. J. Chem. Theory Comput. 2011, 7,
2452−2463.
(49) Dolinsky, T. J.; Nielsen, J. E.; McCammon, J. A.; Baker, N. A.
PDB2PQR: an Automated Pipeline for the Setup of PoissonBoltzmann Electrostatics Calculations. Nucleic Acids Res. 2004, 32,
W665−667.
(50) Jorgensen, W. L.; Chandrasekhar, J.; Madura, J. D.; Impey, R.
W.; Klein, M. L. Comparison of Simple Potential Functions for
Simulating Liquid Water. J. Chem. Phys. 1983, 79, 926−935.
(51) Wang, J.; Cieplak, P.; Kollman, P. A. How Well Does a
Restrained ElectrostaticPotential (RESP) Model Perform in Calculating Conformational Energies of Organic and Biological Molecules? J.
Comput. Chem. 2000, 21, 1049−1074.
(52) Darden, T.; York, D.; Pedersen, L. Particle mesh Ewald: An
Nlog(N) Method for Ewald Sums in Large Systems. J. Chem. Phys.
1993, 98, 10089−10092.

5846

DOI: 10.1021/acscatal.8b04981
ACS Catal. 2019, 9, 5837−5846