A family of structural and functional models for the active site of a unique dioxygenase: Acireductone dioxygenase (ARD)

Article abstract of DOI:10.1016/j.jinorgbio.2020.111253

We report the synthesis and biomimetic activity of a family of model complexes with relevance to acireductone dioxygenase (ARD), an enzyme that displays dual function based on metal identity found in the methionine salvage pathway (MSP). Three complexes with related structural motifs were synthesized and characterized derived from phenolate, and pyridine N₄O Schiff-base ligands. They display pseudo-octahedral Ni(II)-N₄O ligand coordination with water at the sixth site, in close alignment to the structure in the resting state of ARD. The three featured complexes exhibit carbon?carbon bond cleavage activation of lithium acetylacetonate, which was used as a model enzyme substrate. Computationally derived mechanistic routes for the observed reactivity consistent with experimental conditions are herein proposed. The mechanism suggests the possibility of Ni(II)-substrate interactions, followed by oxygen insertion. These results constitute only the third functional model system of ARD, in an attempt to further advance biomimetic contributions to the ongoing debate of ARD's unique metal mediated, regioselective oxidative cleavage.

Full text of DOI:10.1016/j.jinorgbio.2020.111253

JournalofInorganicBiochemistry212(2020)111253
Contents lists available at ScienceDirect
Journal of Inorganic Biochemistry
journal homepage: www.elsevier.com/locate/jinorgbio
A family of structural and functional models for the active site of a unique
dioxygenase: Acireductone dioxygenase (ARD)
Glenn A. Bladea, Riffat Parveenc, Jennifer L. Jaimesa, Wrenell Ilustrea, Diego Saldañaa,
⁎
Denisa A. Ivana, Vincent M. Lynchb, Thomas R. Cundaric, Santiago Toledoa,
a Department of Chemistry, St. Edward's University, 3001 South Congress Ave, Austin, TX 78704, United States of America
b Department of Chemistry, The University of Texas at Austin, 120 Inner Campus Dr Stop G2500, Austin, TX 78712, United States of America
c Department of Chemistry, CASCaM, University of North Texas, 1508 W. Mulberry Street, Denton, TX 76203, United States of America
A R T I C L E I N F O
A B S T R A C T
Keywords:
We report the synthesis and biomimetic activity of a family of model complexes with relevance to acireductone
dioxygenase (ARD), an enzyme that displays dual function based on metal identity found in the methionine
salvage pathway (MSP). Three complexes with related structural motifs were synthesized and characterized
derived from phenolate, and pyridine N4O Schiff-base ligands. They display pseudo-octahedral Ni(II)-N4O ligand
coordination with water at the sixth site, in close alignment to the structure in the resting state of ARD. The three
featured complexes exhibit carbon‑carbon bond cleavage activation of lithium acetylacetonate, which was used
as a model enzyme substrate. Computationally derived mechanistic routes for the observed reactivity consistent
with experimental conditions are herein proposed. The mechanism suggests the possibility of Ni(II)-substrate
interactions, followed by oxygen insertion. These results constitute only the third functional model system of
ARD, in an attempt to further advance biomimetic contributions to the ongoing debate of ARD's unique metal
mediated, regioselective oxidative cleavage.
Modeling chemistry
Schiff-base Ni coordination complexes
Metalloenzyemes
Oxygenases
Inorganic chemistry
Computational studies
1. Introduction
methionine precursor) and formic acid. This reaction occurs when iron
is bound to the active site of the enzyme and is known as the “on-
In this article, a novel set of structural and functional biomimetic
models for the enzyme nickel acireductone dioxygenase (Ni-ARD) is
introduced. These complexes represent only the third example of a re-
active biomimetic model for the enzyme, and the first that will permit
the systematic exploration of the role that the electronic environment of
the metal plays in function, mechanism, and ultimately its role in
promoting enzymatically derived disease pathways for Ni-ARD. This
work thus contributes to the very relevant and ongoing debate con-
cerning the role and mechanism of action that ARD has in disease.
Acireductone Dioxygenase (ARD) plays a key role in the ubiquitous
methionine salvage pathway (MSP) in animals, plants, and bacteria [1].
The MSP uses methylthioadenosine (MTA) derived from S-adenosyl
methionine (SAM) after polyamine synthesis in animals and ethylene in
plants. MTA is an inhibitor of polyamine synthesis and transmethyla-
tion reactions [2]. It is known that the inhibition of MTA halts DNA
replication and elevated amounts of polyamines are associated with
tumor formation [3]. ARD catalyzes the penultimate step in MSP, the
oxidative cleavage of the substrate acireductone (1, 2-dihydroxo-3-keto-
5-(thiomethyl)pent-1-ene) into 2-keto-4-(thiomethyl) butyrate (KMBT-
pathway” route. In contrast, when nickel is bound to the active site, an
“off-pathway” shunt leads to the formation of 3-(methylthio)propionate
(MTP), formate, and carbon monoxide (Fig. 1). This difference in the
regioselective oxidation of substrate can be ascribed to the identity of
the metal ion. ARD is the only known example of a metalloenzyme
displaying such behavior. Recently,
a mouse analogue of ARD
(MmARD) was structurally characterized at high resolution [4], and 3
human analogue of ARD (HsARD) is now available at low resolution
[5]. Both of these mammalian analogues showed via in vitro studies to
perform the same metal dependent chemistry as that observed in the
bacterial ARD [4,6,7]. Furthermore, metal promiscuity binding studies
showed that MmARD and HsARD display identical “off-pathway”
chemistry when bound to Co2+ or Mn2+ [6,7]. It has been hypothe-
sized that the “off-pathway” function of ARD, which is observed in the
presence of metals relevant in mammals (cobalt, iron and manganese),
could lead to cells having anti-apoptotic behavior in cancerous cells
mediated by the production of CO, a known anti-apoptotic signaling
molecule [8,9], and neurotransmitter in mammals [10]. The oxidative
cleavage of acireductone substrate is generally accepted to follow a
⁎ Corresponding author.
E-mail address: stoledoc@stedwards.edu (S. Toledo).
https://doi.org/10.1016/j.jinorgbio.2020.111253
Received 12 July 2020; Received in revised form 15 August 2020; Accepted 6 September 2020
Availableonline14September2020
0162-0134/©2020ElsevierInc.Allrightsreserved.

G.A. Blade, et al.
JournalofInorganicBiochemistry212(2020)111253
Fig. 1. Key features of the methionine salvage pathway. “On and off” ARD pathways display distinct product formation based solely on the identity of the metal
center. Metal promiscuity studies show that both Mn and Co follow the “off pathway” route.
Lewis acid mediated activation followed by oxygen insertion, and not a
more traditional path of oxygen activation promoted by other dioxy-
genases [11].
not affect regioselectivity, which is in contrast to the chelate hypoth-
esis. A fourth working hypothesis was recently deduced from computa-
tional and experimental work done on human ARD (HsARD) [7]. These
studies suggest that human ARD goes through an oxygen activation
step, where oxygen is bound to the active site followed by substrate
In the resting state of the enzyme, the metal ion in ARD is co-
ordinated to the protein backbone in a pseudo-octahedral environment,
by three histidine and one glutamate residues, with the remaining co-
ordination sites occupied by two water molecules (Fig. 2). The structure
of the active site is conserved regardless of its enzymatic context [1].
The actual mechanism for activation of acireductone is currently
under debate among four working hypotheses [7,12,13,14]. The first,
deemed the chelate hypothesis, proposes that the regioselectivity of the
reaction, and hence the differences of product formation, result from
the mode of substrate binding to different metals, Fe vs. Ni. This dif-
ference in substrate coordination promotes a different oxygenation
pathway by encouraging only certain intermediates to be accessible
[12]. A second working mechanistic proposal, challenging the chelate
hypothesis comes from recent computational work. These studies sug-
gest that differences in the profile of oxygenation product formation are
influenced by the charge transfer differences innate to Fe and Ni. Where
iron is more capable of donating electron density to the substrate
creating a pseudo-redox state that is simply not possible with nickel
[13]. A third hypothesis is derived from biomimetic modeling work [14].
These studies suggest that the regioselective behavior of ARD is
modulated by the access of water to the active site of the enzyme. The
models' biomimetic reactivity appeared to be susceptible to the pre-
sence or absence of water leading to off-pathway type products only
when water was rigorously removed regardless of the type of metal
used. In this work it was shown that coordination mode of substrate did
oxidation in the presence of human relevant metals Fe2+ and Co2+
.
This is in direct contrast with earlier studies indicating no evidence for
redox activity of the metal at the active site [15].
Taken together this body of work highlights the need for further
clarification on how the enzyme mediates the proposed regioselective
oxidation, and more importantly the need for more and improved ac-
tive site models of ARD. This is relevant because of ARD's possible in-
volvement in mammalian systems with the anti-apoptotic behavior in
cancerous cells mediated by the production of CO. Those studying ARD
agree that understanding the mechanistically driven regioselectivity on
these enzymatic reactions is essential to clarifying ARD's “off-pathway”
chemistry, its involvement in disease, and other moonlighting functions
[1,4,15,16,17].
To our knowledge, there are only two functional biomimetic models
for ARD. The Ni(II) complex [(6-Ph2TPA)Ni(CH3CN)2](ClO4)2 uses the
N4 chelating ligand 6-Ph2TPA (N,N-bis[(6-phenyl-2-pyridyl)methyl]-N-
(2-pyridylmethyl)amine). Extensive biomimetic reactivity studies have
been reported for this model using acireductone substrate analogues
[14,17,18,19]. This work contributed observations aimed at clarifying
the ongoing mechanistic uncertainty, specifically challenging the che-
late hypothesis [14] mentioned above. Recent work with macrocyclic
N4 tetradentate ligands (N,N′-di-benzyl-2,11-diaza[3,3](2,6)pyr-
idinophane), and their corresponding Ni(II) complexes, reported the
Fig. 2. Left: Active site of Ni-ARD (PDB ID: 5I91) [4]. Center: Skeletal structure of the active site. Right: Skeletal drawing of the ARD model complexes presented in
this work.
2

G.A. Blade, et al.
JournalofInorganicBiochemistry212(2020)111253
oxidative cleavage of an ARD substrate analogue under atmospheric
oxygen and ambient conditions [20]. Although these are functional
models, the N4 systems lack all of the structural and electronic char-
acteristics to accurately mimic the electronic environment of ARD's
active site, specifically the introduction of the oxygen moiety corre-
sponding to the glutamate donor in ARD. It is important to note that
additional work reported the use of an N3 chelating ligand (2,6-bis(1-
methylbenzimidazolyl)pyridine) to make Ni(II) complexes that were
shown to be reactive with substituted diketones towards CeC oxidative
bond cleavage analogous to the enzymatic reactivity [21]. These re-
sults, however, were recently put into question suggesting that activa-
tion of electron deficient diketone substrates was not reproducible in
these complexes.[22]. The biomimetic precedents argue in favor of
models that more closely align with the electronic and structural
characteristics of the enzyme. In the present work we introduce the
synthesis and characterization of a family of structural models closely
related to ARD, which more closely mimic the electronic structure of
the active site and constitute proof of principle for future systematic
structure-function studies of ARD's oxidative reactivity using model
complexes. We report here preliminary results of their biomimetic
carbon‑carbon cleavage reactivity towards a substrate mimic, and a
proposed mechanism for the observed reactivity. This work expands
ongoing biomimetic studies to advance the aforementioned debate re-
garding the unique reactivity of ARD.
corresponding aldehyde and NaBH(OAc)3 were added to the reaction
mixture and allowed to react for an additional 30 min. The reaction was
then quenched with 15 mL of 2 M NaOH and extracted with di-
chloromethane and washed with saturated sodium chloride solution,
followed by drying over anhydrous sodium sulfate. The solvent was
removed under reduced pressure, then further dried under vacuum,
resulting in a brown oil in 88% yield. The acetyl protecting group was
removed by adding 20 mL of 5 M aqueous hydrochloric acid and al-
lowed to reflux for 24 h under an N2 atmosphere. The resulting solution
was treated with careful addition of 5 M NaOH and the organics were
extracted with dichloromethane. Afterwards, the organics were dried
over anhydrous KOH and dried under reduced pressure. Spectroscopic
details of ligand precursors can be found in the supporting information
document.
2.4. Synthesis of metal complexes
2.4.1. [NiII(OPhN4(6-H-DPEN)(H2O)](OTf) (1)
The synthesis and characterization of complex 1 was published in
earlier work [23].
2.4.2. [NiII(OPhNO2N4(6-Me-DPEN)(H2O)](OTf) (2)
Synthesis of the metal complex 2 was formed via an in-situ re-
ductive amination reaction. The 6-Me-DPEN ligand precursor (192 mg,
0.751 mmol) was dissolved in 5 mL dichloroethane with 4 Å molecular
sieves. One equivalent of the corresponding aldehyde, 2-hydroxy-5-ni-
trobenzaldehyde (126 mg, 0.751 mmol) was added and the mixture was
allowed to react for 3 h. The solvent was then removed under reduced
pressure and further dried under vacuum. 5 mL of MeOH and one
equivalent of N,N-diisopropylethylamine (0.0759 mL, 0.751 mmol)
were added and allowed to stir for 20 min. Nickel tri-
fluoromethanesulfonate (268 mg, 0.751 mmol) was then added and the
reaction was allowed to stir under dynamic nitrogen at room tem-
perature for two days. The resulting solution was filtered through a frit
under vacuum and then pumped down under reduced pressure. The
resulting oil was washed with diethyl ether three times and then
pumped down once again. A green powder was isolated.
2. Experimental section
2.1. Materials and methods
The preparation and manipulation of air-sensitive compounds were
performed using standard Schlenk techniques under an N2 atmosphere.
Reagents and solvents were purchased from commercial suppliers of the
highest available purity and used without further purification unless
otherwise noted. Solvents (MeOH, MeCN, 2-Butanone, and Et2O) were
dried using 4 Å molecular sieves. Solvents were degassed via freeze-
pump-thaw cycles.
2.2. Physical methods
Yield = 74.2% (351.7 mg, 0.557 mmol). Electronic absorption
spectrum (MeOH): λ (nm) (ε (M−1 cm−1): for complex 2: 235 (5128),
371 (3896), 575 (99). IR: C=N: 1654 cm−1 We were not able to isolate
complex 2 as an analytically pure sample. HRMS (ESI-MS, m/z): calcd.
for [C23H24N5NiO3]+: found: 476.1226, calculated: 476.1227.
1H NMR spectra were recorded on a Varian 300 MHz spectrometer
at room temperature and referenced to a residual deuterated solvent.
UV–vis spectra were recorded in a 1 cm quartz cuvette on a Varian Cary
100 Bio spectrophotometer. Attenuated Total reflectance-Fourier-
Transform Infrared spectra (ATR-FTIR) were recorded with an Agilent/
Cary 630 FTIR KBr or ZnSe engine. The electrochemical data were
obtained using CH Instrument 600E Electrochemical workstation. ESI-
Mass spectral data were collected using Agilent Technologies 6530
Accurate-Mass Q-TOF LC/MS equipped with a Jet Stream electrospray
ionization (ESI) source. Elemental analyses were performed by
Galbraith Atlantic Microlabs, Norcross, GAt
2.4.3. [NiII(OPhNO2N4(6-H-DPEN)(H2O)](OTf) (3)
The synthesis of complex 3 followed the procedure outlined for
complex 2 with the alteration of using the 6-H-DPEN ligand precursor
(63.5 mg, 0.262 mmol). A yellow powder was isolated. Yield = 80%
(129.1 mg, 0.2096 mmol). Electronic absorption spectrum (MeOH): λ
(nm) (ε (M−1 cm−1): for complex 3: 237 (9386), 381 (6879), 567
(115). IR: C=N: 1650 cm−1 Elemental Analysis for NiO7N5C22H22SF3
Calcd: C, 42.88%; H, 3.60%; N, 11.37%. Found: C, 43.14%; H, 3.49%;
N, 11.34%. HRMS (ESI-MS, m/z): calcd. for [C21H20N5NiO3]+: found:
448.09170, calculated: 448.0914.
2.3. Synthesis of ligand precursors 6-H-DPEN (LP1) and 6-Me-DPEN (LP2)
The ligand precursors N,N-Bis(2-pyridilmethyl)ethane-1,2-diamine
(6-H-DPEN, L1) and N,N-Bis(6-Me-2-pyridilmethyl)ethane-1,2-diamine
(6-Me-DPEN, L2) were synthesized using slight modifications to pre-
viously reported synthetic procedures.23-25 N-acetylethylenediamine
(141 μL, 1.03 mmol), and the corresponding aldehyde (2-pyr-
idinecarboxaldehyde or 6-Me-2-pyridinecarboxaldehyde, 139 μL,
1.03 mmol) were dissolved in 10 mL of 1,2-dichloroethane and allowed
to stir under an N2 atmosphere at room temperature. The reaction
progress was monitored via thin layer chromatography using ninhydrin
stain to screen for any unreacted amine. After ~30 min, an equivalent
of NaBH(OAc)3 (0.4375 g, 2.89 mmol) was added to the reaction
mixture. Continual monitoring via TLC was done as the mixture was
allowed to react for 2 h. After this period an additional equivalent of the
2.5. X-ray crystallography
Single crystals of 1–3 were used for X-ray crystallographic studies.
This analysis was performed at the University of Texas at Austin X-ray
facility laboratories. X-ray data collection was done on a Rigaku AFC12
diffractometer with a Saturn 724+ CCD using a graphite mono-
chromator with MoKα radiation (λ = 0.71073 Å). The data were col-
lected at 100 K using a Rigaku XStream Cryostream low temperature
device. Crystal data, refinement parameters, and additional details can
be found in the supporting information for this manuscript.
3

G.A. Blade, et al.
JournalofInorganicBiochemistry212(2020)111253
2.6. Oxidative reactivity studies
complexes 1-3 aim to modulate the metal's Lewis acidity via structural
modifications. Specifically complexes 2 and 3 introduce a 4-nitro sub-
stituent to the phenolate donor moiety, and 2 substitutes an additional
6-Me group on the pyridine donors of the ligand (Scheme 1). Recent
work has demonstrated the importance of analogous modulation of the
ligand's sterics and Lewis acidity plays on reactivity of model complexes
[25,26]. The metal complexes in this work aim at addressing questions
about the specific role that Lewis acidity might play in the observed on-
and off-pathway functions of ARD.
In a typical experiment, to a Schlenk reaction flask were added
~20 mg of the appropriate nickel complex (~0.035 mmol), 5 mol
equivalents of lithium acetylacetonate, and these were dissolved in 1 ml
CD3OD and heated to 35 °C under a dynamic atmosphere of oxygen gas.
This was done by flowing oxygen gas via the Schlenk line onto the
reaction flask. The reactions were allowed to stir for at least 24 h. with
ongoing in-situ monitoring thereafter via 1H NMR by removing the
reaction mixture and directly inserting into the NMR tube.
The resulting novel N4O nickel complexes were isolated as yellow-
green crystals from slow vapor diffusion crystallizations of diethyl ether
into 2-butanone solutions of the corresponding complex. The molecular
structures of complexes 1, 2, and 3 were obtained via X-ray diffraction
and are shown in Fig. 3.
2.7. Computational methods
Computational studies employed the Gaussian 09 software package.
For geometry optimizations - and to obtain enthalpic and free energy
corrections - the B3LYP functional was utilized in conjunction with the
SBKJC pseudopotential/valence basis set for nickel while the
6–31 + G(d) all-electron basis set was used for main group elements. At
the optimized geometries, single point energy calculations with a larger
basis set on the main group elements, 6–311++G(d,p), were con-
ducted along with solvent (acetonitrile and methanol were modeled
within the SMD continuum solvation model) and dispersion (GD3-BJ)
corrections. Intrinsic reaction coordinate calculations were conducted
to confirm all transition states. All reported energetics are in kcal/mol
and assume standard temperature and pressure.
The primary coordination sphere around the Ni center of each of the
three complexes is a Schiff-base, N4O donor framework with a water
molecule occupying the sixth coordination site. The introduction of an
O-donor moiety in these complexes is the first attempt at modeling the
contribution of the glutamate oxygen of the ARD active site.
Furthermore, these complexes constitute the first examples of water-bound
model systems emulating the resting state of the enzyme. Key structural
parameter comparisons across the complexes and with the active site of
ARD are summarized in Table 1. Overall the mean NieN bond lengths
are only slightly shorter for the model complexes (0.02 Å difference)
when compared to the enzyme's average NieN bonds [4]. The mean Ni-
OPh is 0.08 Å shorter than the Ni-OGlu bond length, as expected given
donor basicity differences between a glutamate O-donor vs. a phenolate
O-donor. The Ni-OH2 average distance in the enzyme is only 0.03 Å
longer than the mean Ni-OH2 of the model compounds; this particular
site is the one proposed to be labile in order to permit substrate binding.
In spite of differences in coordination between the enzyme's active site
and these structural models, given that these complexes carry an ad-
ditional nitrogen donor moiety (N4O vs N3O, Fig. 2), we propose the
incorporation of an oxygen donor, and the unique feature of water
binding in complexes 1-3 provide sufficient structural similarities to
ARD to justify these as structural models of ARD's active site. No bio-
mimetic ARD activity has been reported for N3O or N4O chelating li-
gands.
3. Results and discussion
3.1. Synthesis and characterization of metal complexes [NiII(OPhN4(6-H-
DPEN)(H2O)](OTf) (1), [NiII(OPhNO2N4(6-Me-DPEN)(H2O)](OTf)
(2), and [NiII(OPhNO2N4(6-H-DPEN)(H2O)](OTf) (3)
This work builds from our previously published structural model of
ARD (1) [23], and takes advantage of its modular ligand design that
permits the synthesis of a family of structurally related compounds.
Complexes 2 and 3 are structural derivatives of 1 and were synthesized
via the in-situ condensation of ligand precursor [N′,N′-bis(2-pyr-
idilmethyl)ethane-1, 2-diamine; 6-H-DPEN (LP1) and N′,N′-bis(6-me-
thyl-2-pyridilmethyl)ethane-1, 2-diamine; 6-Me-DPEN (LP2)], with the
corresponding aldehyde (salicylaldehyde 1, and 2-hydroxy-4-ni-
trobenzaldeyde 2,3 in the presence of N,N-diisopropylethylamine
(DIPEA) as an auxiliary base (Scheme 1). The ligand precursors LP1,
and LP2 were synthesized in a two-step reductive amination via slight
modification of existing synthetic procedures (see Supporting In-
formation) [23–25]. Given that ARD's oxidative cleavage of substrate is
proposed to occur via Lewis acid mediated substrate activation,
When evaluating structural differences between the complexes as a
result of the introduction of nitro groups or methylated-pyridyl deri-
vatives, these modifications don't appear to have minor effects on the
overall structures. It was hypothesized that nitro and methylated deri-
vatives will lead to weaker metal-ligand interactions due to electron
withdrawing and steric effects respectively. These modifications were
introduced to impact the Lewis acidity at the metal center, and there-
fore indirectly affect eventual metal substrate interactions. Experiment
Scheme 1. Synthetic procedure of biomimetic metal complexes 1 (R = H, R1 = H), 2 (R = CH3, R1 = NO2), 3 (R = H, R1 = NO2) from ligand precursors LP1 and
LP2.
4

G.A. Blade, et al.
JournalofInorganicBiochemistry212(2020)111253
Fig. 3. Crystal structures of [NiII(OPhN4(6-H-DPEN)(H2O)](OTf) (1), [NiII(OPhNO2N4(6-Me-DPEN)(H2O)](OTf) (2), and [NiII(OPhNO2N4(6-H-DPEN)(H2O)](OTf) (3).
Thermal ellipsoids are drawn at the 50% probability level. H atoms and counterions are omitted for clarity.
approximately 375 nm (~18,000 M−1 cm−1
)
and 290 nm
Table 1
(~4000 M−1 cm−1), and weak d-d visible band at ~575 nm
(~45 M−1 cm−1). Based on literature examples, the band around
375 nm can be assigned to a ligand to metal charge transfer transition
primarily associated with the phenolate ligand [19,27]. The weak [d-d]
bands are characteristic of hexacoordinated nickel(II) complexes. These
are not solvent dependent, thus suggesting that the hexacoordinate
structure of the complexes remains intact in solution.
Key structural parameter comparisons for complexes 1, 2, 3 and ARD active site
from Mm-ARD (PDB ID: 5I91) [4].
1
2
3
Mm-ARD
Ni-N1(N-tert.)
Ni-N2(im)
2.120(3)
2.001(3)
2.104 (3)
2.097(2)
2.024(2)
2.075(2)
2.098(2)
2.047(2)
2.159(2)
2.184 (2)
2.001(2)
2.125 (2)
2.099(7)
2.010(7)
2.084(8)
2.087(8)
2.009(6)
2.139(6)
Avg. Ni-N
2.1
Avg. Ni-O(H2O)
Ni-O (Glu)
2.15
2.1
Ni-N3(Py-right)
Ni-N4 (Py-left)
Ni-O(Ph)
The 1H NMR spectra of complexes 1, 2, and 3 all display para-
magnetically shifted signals consistent with features corresponding to
penta- and hexacoordinated Ni(II) complexes (Supporting information
Figs. S3, S5, S9) [28].
Ni-O(H2O)
shows the non-substituted derivative 1 has the shortest Ni-OH2 bond
length (0.05 Å and 0.06 Å shorter as compared to 2 and 3 respectively).
Complex 2, containing 6-methyl-pyridyl donors, shows the largest Ni-
3.2. Biomimetic reactivity observations of model complexes
N
py bond lengths consistent with a steric effect imparted by the methyl
In previous work, it was observed that complex 1 activated the
crystallization solvent 2-butanone in the presence of oxygen giving rise
to the formation of acetic acid, a probable carbon‑carbon bond cleavage
product [23]. This led to a systematic study to test the biomimetic
oxidative activity of complex 1, by probing its reactivity in the presence
of oxygen towards a proof of principle model substrate, lithium acet-
ylacetonate (LiACAC). This substrate serves as an analogue to ARD's
native substrate 1, 2-dihydroxy-3-keto-5-methylthiopent-1-ene (acir-
eductone). Although LiACAC's symmetry does not allow for testing the
regioselective chemistry observed in ARD, it has been used in the study
of analogous model systems as a model compound for testing oxidative
reactivity [20–22]. Furthermore, this substrate is biologically relevant
to other related oxygenases specifically acetylacetone dioxygenase
(Dke1) [29].
substituents. Finally, the bonded water appears to prefer a trans NImine
coordination in complex 1, and trans NPy coordination in complexes 2
and 3. Whether this is simply an artifact of the structural character-
ization thermodynamic preference, or if this position for the labile site
remains in solution is yet to be determined. Density functional theory
(DFT) simulations indicate, however, that the differences in free en-
ergies between the trans-Npy and trans-Nimine coordination isomers is
very small, ca. 1 kcal/mol or less, suggesting that the particular isomer
observed in the solid-state is likely quite sensitive to the crystal en-
vironment.
Cyclic voltammetry was performed on all three complexes
(Supporting information Figs. S4, S7 and S10). These display cathodic
redox potentials for the presumed Ni(II) to Ni(III) redox couple.
Complex 1 (unsubstituted), displays an irreversible wave with the most
anodically shifted potential at 510 mV (vs. vs. Ferrocene-Fc/
Ferrocenium-Fc+). The redox potential of 2 is cathodically shifted and
displays a quasi-reversible wave with an E1/2 of 826 mV (vs. Fc/Fc+).
Complex 3 falls in between with a positive reduction potential E1/2 of
707 mV (vs. Fc/Fc+) and is also quasi-reversible. These reduction
potentials are consistent with our structural observations and ligand
design goals aimed at modulating the Lewis acidity of the complexes.
Unsubstituted 1 is more thermodynamically able to access the Ni(III)
state in comparison to the complexes containing sterically hindered and
electron withdrawing modifications (2,3). The substitutions in 2 and 3
should increase the Lewis acidity of the metal by removing electron
density at the metal center or preventing the approach of the pyridine
donors as evidenced in the crystal structure of 2. It is important to note
that no other reduction/oxidation waves were observed suggesting that
there is no redox-activity for the phenolic ligand framework.
In a typical oxygenation reaction, a 0.035 M solution of the nickel
complex in deuterated methanol was mixed in a 1:5 ratio with LiACAC
in a dynamic atmosphere of oxygen and heated to 35 °C while mon-
itored for changes via 1H NMR of the reaction mixture. The 1H NMR
spectra for the reaction mixtures between 1 and LiACAC under oxyge-
nation and anaerobic conditions are shown in Fig. 4 (top and bottom
respectively). Under dynamic O2, the decrease of the peak corre-
sponding to substrate LiACAC was observed, as measured by compar-
ison to an internal standard, concomitant with the appearance of sev-
eral signals in the region between 1–9 ppm of the spectrum (Fig. 4 top).
This is in direct contrast with the anaerobic reaction (Fig. 4 bottom)
which retains the relative ratio of internal standard to LiACAC and lacks
specifically the features at 1.37, 1.91 ppm; the diamagnetic signals
between 6.6 and 7.7 ppm; and the sharp singlet at 9.88 ppm. These
results highlight the contrast between the reaction in the presence and
absence of oxygen suggesting possible oxygen involvement. A control
experiment under identical conditions of LiACAC under an O2 atmo-
sphere, in the absence of metal, resulted in the isolation of starting
The UV–Visible spectra of 1, 2, and 3 is available in the supporting
information. All three complexes display charge transfer bands at
5

G.A. Blade, et al.
JournalofInorganicBiochemistry212(2020)111253
Fig. 4. 1H NMR in CD3OD of the reaction mixture between 1 with LiACAC under O2 (top); 1 and LiACAC under N2 (Bottom).
material LiACAC and no evidence of the presence of analogous peaks to
those observed in the metal oxygenated reaction. Similarly, O2 depen-
dent reactivity was observed for complexes 2 and 3 where N2 and O2
reactions display different sets of peaks growing although slight dif-
ferences between these reactions and those observed in reactions with 1
warrant further studies and will be reported in a future communication.
is no structural enzymatic data to confirm that indeed the substrate in
ARD is a bidentate chelate 1. Furthermore when human relevant metals
are bound to ARD, Fe, and Co, monodentate binding is the proposed
binding mode of substrate [7]. Computational (vide infra) and experi-
mental observations (NMR diamagnetic aromatic signals Fig. 4) do not
rule out a bidentate binding mode of substrate if one of the pyridine
arms were to dissociate.
The second step in the mechanism is proposed as the dioxygen ad-
dition to the enolate bound species to form a peroxo type intermediate.
This intermediate could then attack the carbonyl bound to nickel to
give a second intermediate, a 1, 2-dioxetane, which eventually col-
lapses with the production of pyruvaldehyde and an acetate bound
nickel complex (Scheme 2). Pyruvaldehyde has been shown to de-
compose via a variety of pathways, under the given experimental
conditions, including the decomposition to CO2 and acetic acid, as well
as hydration oligomeric products [30].
3.3. Proposed mechanism for reactivity
To account for these experimental observations indicating an
oxygen dependence of the reaction we propose an oxygen mediated
carbon‑carbon bond cleavage of the acetylacetonate substrate (Scheme
2). The first step could correspond to the binding of acetylacetonate to
the nickel complex with displacement of water. A binding event is
supported by NMR titration experiments that indicate changes in the
paramagnetic spectra of the nickel complexes upon addition of 1.25 eq.
of LiACAC substrate (Fig. S11). Although, in ARD substrate binding of
acireductone for Ni-ARD is proposed to be bidentate, at this time there
Several observations provide support for the hypothesized reaction
mechanism. Evolution of bubbles was observed in the aerobic reactions,
Scheme 2. Proposed reaction mechanism for the oxygenation LiACAC mediated by the Ni-N4O complexes.
6

G.A. Blade, et al.
JournalofInorganicBiochemistry212(2020)111253
which was more evident after the first 24 h of reaction. To test for the
possibility of this gas evolution corresponding to CO2, the headspace of
the reaction flask was purged into a vessel containing a saturated so-
lution of Ca(OH)2, which will react with CO2 to produce CaCO3 as a
precipitate. After purging the headspace of the flask into the Ca(OH)2
solution, small amounts of a milky white precipitate resulted. After
ATR-FTIR analysis of the powder its identity was confirmed as CaCO3
when compared to a control (Fig. S12). To reiterate, CO2 production
supports the formation of pyruvaldehyde, a product exclusive to oxy-
genation of the substrate.
complexes 1, 2, 3 and acetylacetonate - in SMD-methanol (continuum
solvent model) are shown in Fig. 5.
The computational studies explore the feasibility of an oxidative
activation of the 2,3 carbon‑carbon bond of acetylacetonate via a sub-
strate-bound
species,
which
displaces
the
labile
water.
Thermodynamically, this complex is favorable with respect to its water-
bound parent by an average of 7.2 kcal/mol when considering all three
complexes, consistent with NMR titration studies (Fig. S11). The dis-
placement energies for the NO2-substituted complexes 2 and 3 are
slightly more exergonic (blue and red values), consistent with the
greater Lewis acidity they are proposed to impart to the nickel center.
Computational analysis of a bidentate binding of substrate suggest that
k1- and k2-acac coordination modes (the latter being generated by
displacement of a pyridyl arm) are essentially degenerate, implying that
both linkage isomers may exist in equilibrium in solution. The rate
limiting step of the modeled reaction passes through a transition state
(TS-1, Fig. 5) that is an association between the 2,3‑carbon‑carbon
enolate bond of the bound substrate and dioxygen to form a peroxo type
intermediate (Int1) with reasonable activation barriers of 19–23 kcal/
mol, depending on the ligand substituents, consistent with a manage-
able activation barrier under mild temperature experimental condi-
tions. The formation of the second intermediate, a 1, 2 dioxetane
complex (Int2), is preceded by intramolecular attack of the peroxo on
the bound carbonyl carbon of the enolate. It is hypothesized that this
step in the mechanism is facilitated by an intramolecular interaction
between the distal oxygen in the peroxo intermediate and a hydrogen
atom on the backbone of the ligand. This interaction starts with TS1
and its observed in both Int1 and TS-2 (calculated bond lengths 2.11 Å,
1.76 Å, and 1.91 Å respectively). This interaction may be key to guiding
the peroxo towards formation of the dioxetane (Int2). To further test
this interaction, different conformational states of the substrate bound
to the metal were explored and no other conformation yielded the
Additionally, we propose that the peak at 1.91 ppm observed in the
NMR spectrum of the aerobic reactions with 1 (Fig. 4 top) can be ten-
tatively assigned to the -CH3 of the acetate bound nickel(II) complex,
another product of the oxygenation of LiACAC. We propose this based
on comparisons of this peak to a peak observed in the independent
synthesis of an acetate bound nickel(II) complex of 1 (Fig. S13). The
additional signals that grown during the oxidative reactivity could not
be definitively assigned although we hypothesize, they might corre-
spond to pyruvaldehyde hydration/solvation/oligomeric type products
[30].
Finally, in most of the oxygenation reactions the formation of a
small amount of an off-white cream-colored precipitate was observed as
the reaction proceeded. The powder was isolated and analyzed via ATR-
FTIR. The IR spectrum of the precipitate is shown in Fig. S14. The
spectrum displays strong bands at 950 cm−1, consistent with C-O-C
stretching, a band at 1379 cm− and a broad peak at 3332 cm−1 cor-
responding to an OeH stretch, all of these expected to be present with
pyruvaldehyde hydrolytic type oligomers. This reactivity of pyr-
uvaldehyde is well known in the literature [30a,e,f]..
To further test the feasibility of the proposed mechanism density
functional theory (DFT) calculations were performed. The proposed
reaction coordinate diagram, and corresponding free energies for the
various intermediates and transition states for the reaction between
…
…
observed HL Operoxo interaction. The HL Operoxo interaction is also
Fig. 5. Calculated free energy profile for the oxidation of acac catalyzed by [NiII(OPhN4(6-H-DPEN)(H2O))]+ (1), [NiII(OPhNO2N4(6-Me-DPEN)(H2O)](OTf) (2), and
[NiII(OPhNO2N4(6-H-DPEN)(H2O)](OTf) (3). Quoted values are free energies in kcal/mol and were derived using DFT as outlined in Computational Methods.
Multiplicities for each species is indicated by the superscript to the left of the name. The product of the reaction (not pictured) is an acetate-bound Ni(II) species.
7

G.A. Blade, et al.
JournalofInorganicBiochemistry212(2020)111253
intriguing as it could suggest secondary coordination sphere involve-
ment and directing effects in the observed regioselective activation of
the native substrate in the enzyme. The intermediates in the proposed
mechanism such as the formation of a peroxo intermediate, and a 1, 2
dioxetane have been documented for other model complexes [31].
Finally, the dioxetane complex (Int2) collapses in a highly ex-
ergonic process to form pyruvaldehyde and an acetate bound species
(Prod). The exothermicity of product formation observed in the cal-
culations matches values from previous enzymatic [13], as well as
biomimetic computational work [18] in analogous substrates per-
forming CeC bond cleavage. Evidence for the accessibility of an acetate
bound product was published in earlier work [23], and synthetically in
this work via 1H NMR (Fig. S13). Pyruvaldehyde, the other product of
the reaction, is known to decompose further in the presence of oxygen
to form another equivalent of acetic acid and carbon dioxide [30b]
consistent with experimental observations. As shown in Fig. 5, similar
computational results were observed for the two derivatives complexes
2 and 3 with slight differences in the calculated energies of inter-
mediates and transition states, suggesting modest thermodynamic and
kinetic impact on the reactivity imparted by the structural modifica-
tions.
oxidative cleavage. Elsberg et al. showed the lack of oxidative reactivity
of a collection of N3-tridentate Ni(II) unsubstituted diketonate com-
plexes [22]. This they concluded is consistent with the literature which
suggests the activation of diketonates is only possible for electron rich
diketonate ligands. These results rebutted the findings of previous work
by Ramasubramanian et al. [21] The observed reactivity and compu-
tational studies in this work suggest that an unsubstituted diketonate
ligand (acetylacetonate) is capable of undergoing oxidative cleavage
when reacted with complexes 1-3. These contrasting results point to-
wards the key differences of the complexes in previous reports and our
work. Complexes 1-3 have an N4O ligand environment in contrast to all
other examples with N3 and N4 chelates. It could be hypothesized that
the presence of the oxygen donor in the chelating ligand might allow for
activation of such substrates by imparting additional electron density to
the nickel center supplanting the need for electron rich substrates. Also,
it is important to point out that there are key differences in the reaction
conditions between these studies and our work e.g. oxygenation times,
temperature, presence or absence of water. We currently don't have
enough experimental evidence to make a direct comparison for all the
conditions. This current proof of principle study opens the door to
suggest role that ligand environment might play in these substrate ac-
tivations, but it also highlights the need for further studies that test the
influence of ligand environments analogous to the enzyme and their
influence on oxidative reactivity.
The calculations were also scrutinized for evidence of redox activity
at the metal center and the supporting ligand framework. Analysis of
the spin density yielded no evidence of changes to the spin density of
the nickel(II) center. For all supporting ligand types and all stationary
points - ground and transition states - the calculated (Mulliken popu-
lation analysis) - was uniformly 1.80(2) e−, consistent with a high-spin
Ni(II) with a small degree of metal-ligand covalency. The only excep-
tions are, of course, the NieO2 adducts, where spin density is deloca-
lized onto the oxygen atoms of dioxygen, with more spin density on the
terminal (non-coordinated) vs. the coordinated oxygen of the O2 ligand,
as expected from the classical Drago spin-pairing model of dioxygen
binding [32]. The calculations are thus consistent with cyclic voltam-
metry experiments that indicate highly cathodic redox potentials and
no redox activity at the ligand.
This research indicates the possibility that the model complexes in
this study are capable of biomimetic carbon‑carbon bond cleavage re-
actions. Additional kinetic and mechanistic labeling studies are on-
going. The preliminary observations presented here are promising and
constitute the first example of N4O-Ni(II) complexes displaying such
reactivity, and are a needed addition to the ongoing biomimetic work
relevant to ARD. Future work includes complete mechanistic studies of
the reaction, further testing of oxygenation activity towards more clo-
sely structurally aligned substrates for ARD already published in the
literature [18,19,20], and probing the influence of ligand donating
abilities on reactivity. These advances will contribute more directly to
the debate over how the metal, substrate coordination, and electronic
structure of the enzyme's active site impact the unique regioselectivity
of ARD.
While the outlined computationally derived oxygenation me-
chanism is of most relevance to biomimetic oxygenase activity, a hy-
drolytic/solvolysis retro-Claisen type reaction should also be con-
sidered since it has been reported in the literature that diketones such
as acetylacetonate can undergo carbon‑carbon hydrolytic retro-Claisen
bond cleavage in the presence of metals [33] as well as other Lewis
acids [34]. To our knowledge, no such reactivity has been reported with
nickel complexes. The proposed hydrolysis could result from water at-
tack on the bound carbonyl of acetylacetonate followed by formation of
a diol, tautomerization, and final carbon‑carbon bond cleavage re-
sulting in the production of acetic acid and an acetone enolate that
should rapidly protonate under the reaction conditions to yield acetone.
Similarly, with methanol (reaction solvent) as the nucleophile this
would result in the formation of methyl acetate and acetone. It is im-
portant to reiterate that no retro-Claisen products were observed ex-
perimentally under the aforementioned reaction conditions in either
the aerobic or anaerobic regimes for any of the complexes. When the
experiment was carried out under anaerobic conditions in D2O no hy-
drolytic products were observed either, although the complex only
displayed partial solubility in D2O.
4. Conclusions
A novel family of structural and functional models of acireductone
dioxygenase (ARD) were introduced and structurally characterized.
This research, therefore, expands the availability of biomimetic models
of an intriguing and important enzyme. In spite of structural differences
between these models and the resting state of ARD, these complexes are
the first to incorporate an oxygen binding moiety to the metal co-
ordination environment, and a nickel-bound water, both analogous to
that of the ARD resting state. Building on the structural characteriza-
tion, the dioxygen mediated carbon‑carbon cleavage of a model sub-
strate (LiACAC) promoted by these NiII-N4O complexes was explored
experimentally and computationally. Experimental observations pro-
vide indirect evidence for the proposed mechanism that invokes sub-
strate binding to nickel that provides a path for substrate activation via
oxidation by dioxygen, in biomimetic fashion. The computationally
supported mechanism invokes reactive intermediates observed in other
dioxygenase model systems and enzymes with experimentally relevant
barriers of activation. The observed reactivity, to our knowledge, is the
first examples of biomimetic activity of an N4O nickel system.
Furthermore, the propensity of these complexes towards binding of
substrate in two possible modes, mono and bidentate fashion, suggests a
unique possible pathway for oxidative cleavage not yet reported in the
literature of nickel oxidative biomimetic systems. Further work with
human relevant metals (e.g., Fe and Co) will be necessary given the
recent work in HsARD [7] where monodentate substrate binding is
proposed to undergo oxidative cleavage mediated by an oxygen-bound
For purposes of further assessing the observed lack of a solvolysis/
hydrolytic pathway under reaction conditions computational studies
were performed to probe the feasibility of a hydrolytic mechanism (see
Supporting Information). DFT results suggest that expected inter-
mediates from the hydrolysis of the Ni-acac complex are > 20 kcal/mol
uphill in free energy relative to the reactants in Fig. 5, and so not
competitive from a thermodynamic perspective.
As it was mentioned earlier the oxidative reactivity of unsubstituted
diketonate substrates by Nickel(II) complexes has been reported
[21,22,35]. The conclusions from this work showed that unsubstituted
diketonates e.g. acetylacetonate, are generally unreactive towards
8

G.A. Blade, et al.
JournalofInorganicBiochemistry212(2020)111253
oxidant. Detailed mechanistic studies are underway in our laboratory to
fully understand this observed oxidative reactivity. Additionally, we
have recently developed a related family of N3O systems to allow for
possible bidentate binding of substrate. With these complexes, whose
synthesis, characterization, and reactivity will be reported in due
course, we hope to expand in a systematic fashion understanding of
oxygenases in general, and more specifically how acireductone dioxy-
genase might carry out its regioselective, metal-dependent, oxidative
cleavage. Studying the mechanism of the complexes introduced here
and derivatives thereof can thus contribute to answering many ques-
tions that remain with regards to ARD's mechanism of action and its
involvement in human disease.
(e) L. Marton, A. Pegg, Polyamines as targets for therapeutic intervention, Annu.
Rev. Pharmacol. Toxicol. 35 (1) (1995) 55–91.
[4] A. Deshpande, K. Wagenpfeil, T. Pochapsky, G. Petsko, D. Ringe, Metal-dependent
function of a mammalian acireductone dioxygenase, Biochem 55 (9) (2016)
1398–1407.
[5] RCSB Protein Data Bank, 4QGN, http://www.rcsb.org/pdb/explore.do?
structureId=4QGN, (2015) Last accessed February, 18 2020.
[6] A. Deshpande, T. Pochapsky, G. Petsko, D. Ringe, Dual chemistry catalyzed by
human acireductone dioxygenase, Protein Eng., Des. Sel. 30 (3) (2017) 109–206.
[7] A. Miłaczewska, E. Kot, J. Amaya, T. Makris, M. Zając, J. Korecki, A. Chumakov,
B. Trzewik, S. Kędracka-Krok, W. Minor, M. Chruszcz, On the structure and reaction
mechanism of human acireductone dioxygenase, Chem.–Eur. J. 24 (20) (2018)
5225–5237.
[8] L. Gunther, P. Berberat, M. Haga, S. Brouard, R. Smith, M. Soares, F. Bach,
E. Tobiasch, Carbon monoxide protects pancreatic β-cells from apoptosis and im-
proves islet function/survival after transplantation, Diabetes 51 (2002) 994–999.
[9] S. Thom, D. Fisher, Y. Xu, K. Notarfrancesco, H. Ischiropoulos, Adaptive responses
and apoptosis in endothelial cells exposed to carbon monoxide, Proc. Natl. Acad.
Sci. U. S. A. 97 (3) (2000) 1305–1310.
Accession codes
[10] R. Johnson, F. Johnson, The effects of carbon monoxide as a neurotransmitter, Curr.
Opin. Neurol. 13 (6) (2000) 709–713.
CCDC: 1577098, 1985088 and 1985383 contain the supplementary
crystallographic data for this paper. These data can be obtained free of
charge via www.ccdc.cam.ac.uk/data_request/cif, or by emailing data_
request@ccdc.cam.ac.uk, or by contacting The Cambridge
Crystallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK;
fax: +441,223 336,033.
[11] (a) M. Abu-Omar, A. Loaiza, N. Hontzeas, Reaction mechanisms of mononuclear
non-heme iron oxygenases, Chem. Rev. 105 (6) (2005) 2227–2252;
(b) E. Solomon, D. Heppner, E. Johnston, J. Ginsbach, J. Cirera, M. Qayyum,
M. Kieber-Emmons, C. Kjaergaard, R. Hadt, L. Tian, Copper active sites in biology,
Chem. Rev. 114 (7) (2014) 3659–3853.
[12] T. Ju, R. Goldsmith, S. Chai, M. Maroney, S. Pochapsky, T. Pochapsky, One protein,
two enzymes revisited: a structural entropy switch interconverts the two isoforms of
acireductone dioxygenase, J. Mol. Biol. 363 (4) (2006) 823–834.
[13] C. Valdez, N. Gallup, A. Alexandrova, Co2+ acireductone dioxygenase: Fe2+
mechanism, Ni2+ mechanism, or something else? Chem. Phys. Lett. 604 (2014)
77–82.
Declaration of competing interest
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influ-
ence the work reported in this paper.
[14] C. Allpress, K. Grubel, E. Szajna-Fuller, A. Arif, L. Berreau, Regioselective aliphatic
carbon–carbon bond cleavage by a model system of relevance to iron-containing
acireductone dioxygenase, J. Am. Chem. Soc. 135 (2) (2013) 659–668.
[15] Y. Dai, T. Pochapsky, R. Abeles, Mechanistic studies of two dioxygenases in the
methionine salvage pathway of Klebsiella pneumoniae, Biochem 40 (21) (2001)
6379–6387.
Acknowledgements
[16] M. Sparta, C. Valdez, A. Alexandrova, Metal-dependent activity of Fe and Ni acir-
eductone dioxygenases: how two electrons reroute the catalytic pathway, J. Mol.
Biol. 425 (16) (2013) 3007–3018.
This research was supported by the National Institute of General
Medical Sciences of the National Institutes of Health under Award
Number SC2GM130438. The content is solely the responsibility of the
authors and does not necessarily represent the official views of the
National Institutes of Health. Departmental Welch Foundation grant
(BH-0018-20151106), NSF-STEP (0969153), NSF-IUSE (1525490),
DOE-McNair Post-Baccalaureate Achievement Program (P217A170002)
for faculty and undergraduate student support. Research at UNT was
supported in part by the NSF through grant CHE-1464943. NSF support
for the UNT CASCaM HPC cluster via CHE-1531468 is gratefully ac-
knowledged. We also would like to thank the University of Texas Mass
Spectrometry Facility, especially Dr. Ian Riddington, for the work on
the high-resolution ESI-MS of samples.
[17] E. Szajna, A. Arif, L. Berreau, Aliphatic carbon− carbon bond cleavage reactivity of
a mononuclear Ni (II) cis-β-keto− enolate complex in the presence of base and O2:
a model reaction for acireductone dioxygenase (ARD), J. Am. Chem. Soc. 127 (49)
(2005) 17186–17187.
[18] L. Berreau, B. Tomasz, G. Katarzyna, C. Allpress, J. Wikstrom, M. Germain,
E. Rybak-Akimova, D. Tierney, Mechanistic studies of the O2-dependent aliphatic
carbon− carbon bond cleavage reaction of a nickel enolate complex, Inorg. Chem.
50 (3) (2011) 1047–1057.
[19] C. Allpress, L. Berreau, A nickel-containing model system of acireductone dioxy-
genases that utilizes a C (1)-H acireductone substrate, Eur. J. Inorg. Chem. (27)
(2014) 4642–4649.
[20] S. Raje, K. Mani, P. Kandasamy, R.J. Butcher, R. Angamuthu, Bioinspired oxidative
cleavage of aliphatic C-C bonds utilizing aerial oxygen by nickel acireductone di-
oxygenase mimics, Eur. J. Inorg. Chem. (16) (2019) 2164–2167.
[21] R. Ramasubramanian, K. Anandababu, M. Kumar, R. Mayilmurugan, Nickel (ii)
complexes of a 3N ligand as a model for diketone cleaving unusual nickel (ii)-di-
oxygenase enzymes, Dalton Trans. 47 (12) (2018) 4049–4053.
[22] J. Elsberg, A. Peterson, A. Fuller, L. Berreau, N 3-ligated nickel (II) diketonate
complexes: synthesis, characterization and evaluation of O2 reactivity, Dalton
Trans. 49 (2020) 7564–7575.
Appendix A. Supporting information
X-ray data collection and refinement details, UV-Visible spectra, 1H
NMR characterization and reactivity spectra, elemental analysis, and
ESI-mass spectral data. Supplementary data to this article can be found
online at https://doi.org/10.1016/j.jinorgbio.2020.111253.
[23] D. Ivan, A. Gremillion, A. Sanchez, S. Sanchez, V. Lynch, S. Toledo, The first
structural model for the resting state of the active site of nickel acireductone di-
oxygenase (Ni-ARD), Inorg. Chem. Commun. 89 (2018) 37–40.
[24] S.A. Toledo, Synthesis and Reactivity of an Expanded Family of Superoxide
Reductase (SOR) Model Complexes Using N-Heterocyclic, Thiolate-Containing
Ligands: Towards a Better Understanding of Structural-Functional Relationships,
University of Washington, 2009.
References
[25] M. Schatz, M. Leibold, S. Foxon, M. Weitzer, F. Heinemann, F. Hampel, O. Walter,
S. Schindler, Syntheses and characterization of copper complexes of the ligand (2-
aminoethyl) bis (2-pyridylmethyl) amine (uns-penp) and derivatives, Dalton Trans.
(8) (2003) 1480–1487.
[1] A. Deshpande, T. Pochapsky, D. Ringe, The metal drives the chemistry: dual func-
tions of acireductone dioxygenase, Chem. Rev. 117 (15) (2017) 10474–10501.
[2] D. Morris, L. Marton, C. Dave, M. Dekker (Ed.), Polyamines in Biology and
Medicine, 1981, pp. 43–73.
[26] P. Yan Poon, M. Dedushko, X. Sun, G. Yang, S. Toledo, E. Hayes, A. Johansen,
M. Piquette, J. Rees, S. Stoll, E. Rybak-Akimova, How metal ion Lewis acidity and
steric properties influence the barrier to dioxygen binding, peroxo O–O bond
cleavage, and reactivity, J. Am. Chem. Soc. 141 (38) (2019) 15046–15057.
[27] (a) A. Roth, E. Spielberg, W. Plass, Kit for unsymmetric dinucleating double-Schiff-
base ligands: facile access to a versatile new ligand system and its first hetero-
bimetallic copper− zinc complex, Inorg. Chem. 46 (11) (2007) 4362–4364;
(b) T. Nagataki, K. Ishii, Y. Tachi, S. Itoh, Ligand effects on Ni II-catalysed alkane-
hydroxylation with m-CPBA, Dalton Trans. 11 (2007) 1120–1128.
[28] E. Szajna, P. Dobrowolski, A. Fuller, A. Arif, L. Berreau, NMR studies of mono-
nuclear octahedral Ni (II) complexes supported by tris ((2-pyridyl) methyl) amine-
type ligands, Inorg. Chem. 43 (13) (2004) 3988–3997.
[3] (a) A. Pegg, Polyamine metabolism and its importance in neoplastic growth and as
a target for chemotherapy, Cancer Res. 48 (4) (1988) 759–774;
(b) T. Li, H. Yang, H. Peng, M. Xia, J. Mato, S.C. Lu, Effects of S-adeno-
sylmethionine and methylthioadenosine on inflammation-induced colon cancer in
mice, Carcinogenesis. 33 (2) (2012) 427–435;
(c) M. Tomasi, C. Cossu, Y. Spissu, A. Floris, M. Ryoo, A. Iglesias-Ara, Q. Wang,
S.J. Pandol, N. Bhowmick, E. Seki, E. Posadas, S-adenosylmethionine and methyl-
thioadenosine inhibit cancer metastasis by targeting microRNA 34a/b-methionine
adenosyltransferase 2A/2B axis, Oncotarget. 8 (45) (2017) 78851;
(d) M. Tomasi, M. Ryoo, A. Skay, I. Tomasi, P. Giordano, J. Mato, S. Lu, Polyamine
and methionine adenosyltransferase 2A crosstalk in human colon and liver cancer,
Exp. Cell Res. 319 (12) (2013) 1902–1911;
[29] G. Straganz, A. Glieder, L. Brecker, D. Ribbons, W. Straganz Steiner, D. Grit,
Acetylacetone-cleaving enzyme Dke1: a novel CC-bond-cleaving enzyme from
9

G.A. Blade, et al.
JournalofInorganicBiochemistry212(2020)111253
Acinetobacter johnsonii, Biochem. J. 581 (2003) 573–581.
hydride (HNO) replaces dioxygen in nitroxygenase activity of manganese quercetin
dioxygenase, Proc. Natl. Acad. Sci. U.S.A. 108 (47) (2011) 18926–18931;
(c) Fiedler, A.; Fischer,A. Oxygen activation by mononuclear Mn, Co, and Ni
centers in biology and synthetic complexes. J. Biol. Inorg. Chem. 2017, 22(2-3),
407-424.
[30] (a) I. Nemet, D. Vikić-Topić, L. Varga-Defterdarović, Spectroscopic studies of me-
thylglyoxal in water and dimethylsulfoxide, Bioorg. Chem. 32 (6) (2004) 560–570;
(b) L. Vanoye, A. Aloui, M. Pablos, R. Philippe, A. Percheron, Favre-Réguillon, A.
de Bellefon, C., A safe and efficient flow oxidation of aldehydes with O2, Org. Lett.
15 (23) (2013) 5978–5981;
[32] R. Drago, B. Corden, Spin-pairing model of dioxygen binding and its application to
various transition-metal systems as well as hemoglobin cooperativity, Acc. Chem.
Res. 13 (10) (1980) 353–360.
(c) J.L. Axson, K. Takahashi, D. De Haan, V. Vaida, Gas-phase water-mediated
equilibrium between methylglyoxal and its geminal diol, Proc. Natl. Acad. Sci.
U.S.A. 107 (15) (2010) 6687–6692;
[33] (a) G. Zhang, Z. Hu, F. Belitz, Y. Ou, N. Pirkl, L. Gooßen, Rhodium‐Catalyzed
Annelation of Benzoic Acids with α, β‐Unsaturated Ketones with Cleavage of C− H,
CO− OH, and C− C Bonds, Angew. Chem. Int. Ed. 58 (19) (2019) 6435–6439;
(b) S. Biswas, S. Maiti, U. Jana, An Efficient Iron‐Catalyzed Carbon–Carbon
Single‐Bond Cleavage via Retro‐Claisen Condensation: A Mild and Convenient
Approach to Synthesize a Variety of Esters or Ketones, Eur. J. Org. Chem. 2010 (15)
(2010) 2861–2866.
(d) F. Yasmeen, N. Sauret, J. Gal, P. Maria, L. Massi, W. Maenhaut, M. Claeys,
Characterization of oligomers from methylglyoxal under dark conditions: a
pathway to produce secondary organic aerosol through cloud processing during
nighttime, Atmos. Chem. Phys. 10 (8) (2010) 3803–3812;
(e) K. Loeffler, C. Koehler, N. Paul, D. De Haan, Oligomer Formation in
Evaporating Aqueous Glyoxal and Methyl Glyoxal Solutions, Environ. Sci. Technol.
40 (20) (2006) 6318–6323;
[34] M. Hussein, V. Huynh, R. Hommelsheim, R. Koenigs, T. Nguyen, An efficient
method for retro-Claisen-type C–C bond cleavage of diketones with tropylium cat-
alyst, Chem. Commun. 54 (92) (2018) 12970–12973.
(f) S. Bulman, P. Goswami, G. Tronci, S. Russell, C. Carr, Investigation into the
potential use of poly(vinyl alcohol)/methylglyoxal fibers as antibacterial wound
dressing components, J. Biomater. Appl. 29 (8) (2015) 1193–1200.
[31] (a) X. Han, K. Klausmeyer, P. Farmer, Characterization of the initial intermediate
formed during photoinduced oxygenation of the ruthenium (II) bis (bipyridyl) fla-
vonolate complex, Inorg. Chem. 55 (15) (2016) 7320–7322;
[35] M. Martin, M. Hörner, B. Boneberger, F. Nunes, Synthesis and molecular structure
of Bis (ethylenediamine) acetatonickel (II) hexafluoridophosphate complex [Ni (en)
2 (CH3CO2)] PF6. An unexpected acetylacetonato cleavage reaction, Anorg. Alg.
Chem. 637 (9) (2011) 1229–1233.
(b) M. Kumar, A. Zapata, A. Ramirez, S. Bowen, W. Francisco, P. Farmer, Nitrosyl
10