A nickel-containing model system of acireductone dioxygenases that utilizes a C(1)-H acireductone substrate

Article abstract of DOI:10.1002/ejic.201402254

A mononuclear Ni^II complex bearing the monoanion of 1-acetoxy-3-phenylpropane-2,3-dione (4) as a ligand has been prepared {[(6-Ph₂TPA)Ni{PhC(O)C(O)CHOC(O)CH₃)]ClO₄, 5; 6-Ph₂TPA = N,N-bis[(6-phenyl-2-pyridyl)methyl]-N-(2-pyridylmethyl)amine}. This complex was characterized by ¹H NMR, UV/Vis and IR spectroscopy, mass spectrometry, and elemental analysis. Exposure of solutions of 5 to O₂ did not result in any reaction over the course of hours. Deprotection of 5 by the addition of NaOCH₃ in methanol generated a Ni^II species (6) that contains a coordinated dianionic C(1)-H acireductone. Exposure of 6 to O₂ led to regioselective oxidative cleavage reactivity akin to that found for the Ni^II-containing acireductone dioxygenase enzyme. The strategy outlined herein is the first synthetic approach that enables examination of the oxidative reactivity of a synthetic Ni^II species containing a dianionic C(1)-H acireductone ligand.

Full text of DOI:10.1002/ejic.201402254

FULL PAPER
DOI:10.1002/ejic.201402254
A Nickel-Containing Model System of Acireductone
Dioxygenases that Utilizes a C(1)-H Acireductone
Substrate
CLUSTER
ISSUE
[a]
[a]
Caleb J. Allpress and Lisa M. Berreau*
Keywords: Enzyme mimics / Metalloenzymes / Nickel / Oxygen / Cleavage reactions / Regioselectivity
A mononuclear NiII complex bearing the monoanion of 1-
acetoxy-3-phenylpropane-2,3-dione (4) as a ligand has been
of 5 by the addition of NaOCH
in methanol generated a
Ni species (6) that contains a coordinated dianionic C(1)-H
acireductone. Exposure of 6 to O led to regioselective oxi-
3
II
prepared {[(6-Ph
6-Ph TPA
pyridylmethyl)amine}. This complex was characterized by H
NMR, UV/Vis and IR spectroscopy, mass spectrometry, and
2
TPA)Ni{PhC(O)C(O)CHOC(O)CH
N,N-bis[(6-phenyl-2-pyridyl)methyl]-N-(2-
3
)]ClO
4
,
2
II
5
;
2
=
dative cleavage reactivity akin to that found for the Ni -con-
taining acireductone dioxygenase enzyme. The strategy out-
lined herein is the first synthetic approach that enables ex-
1
II
elemental analysis. Exposure of solutions of 5 to O
2
did not
amination of the oxidative reactivity of a synthetic Ni spe-
result in any reaction over the course of hours. Deprotection
cies containing a dianionic C(1)-H acireductone ligand.
Introduction
enzymes are found at the only known branch point within
this pathway and thus are of current interest. In Klebsiella
pneumonia, the on-pathway reaction is catalyzed by the
iron-containing enzyme Fe-ARDЈ, whereas the off-pathway
Methionine is an essential amino acid that plays an im-
portant role in protein structure, biosynthetic pathways,
and is the start codon for translation in eukaryotic cells.
[
1]
reaction is catalyzed by the nickel-containing enzyme Ni-
Its function in biosynthetic pathways is dominated by its
enzymatic derivatization by adenosine triphosphate (ATP)
to form S-adenosylmethionine (SAM).[ SAM has numer-
ous important biological functions due in part to its sulfo-
nium cation with three S–C bonds,[ including 1) as a co-
substrate for methyltransferases, which leads to the loss of
the methyl group and the formation of S-adenosylhomome-
[13–15]
ARD.
These enzymes cleave the C–C bond(s) of their
substrate with differing regiospecificity (Scheme 1) and are
particularly interesting from a chemical standpoint as the
only constitutive difference between the enzymes is the
2,3]
3,4]
II
II [15]
identity of the metal ion at the active site (Fe or Ni ).
The reaction catalyzed by Ni-ARD also produces CO, a
well-known cellular signaling molecule with therapeutic po-
[
5]
[6]
thionine, 2) as a co-factor in radical SAM enzymes, and
) as a source of n-propylamine during polyamine biosyn-
thesis, leading to the formation of methylthioadenosine
[16]
tential, thereby combining a regulatory junction with the
3
production of a cellular signal. Efforts to gain an under-
standing of the chemical factors that control the regiospec-
[
7]
(
MTA). Polyamines such as spermine and spermidine are
associated with cell growth, and defects in polyamine bio-
[
8]
synthesis regulation are associated with oncogenesis.
Therefore metabolic pathways that regulate polyamine syn-
thesis are a potentially therapeutically important area of
study. One such pathway is the methionine salvage pathway
(MSP), a ubiquitous enzymatic pathway in which the meth-
ylthio unit is recycled into methionine after SAM has been
[
7,9]
converted into MTA during polyamine biosynthesis.
A pair of enzymes known as acireductone dioxygenases
(
ARD and ARDЈ) catalyze the dioxygenolytic cleavage of
[10–12]
an acireductone intermediate within the MSP.
These
[
a] Department of Chemistry and Biochemistry, Utah State
University,
0
300 Old Main Hill, Logan, UT 84322-0300, USA
E-mail: lisa.berreau@usu.edu
http://lisaberreau.org/
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejic.201402254.
Scheme 1. Carbon–carbon bond-cleavage reactions of nickel- and
iron-containing acireductone dioxygenases.
Eur. J. Inorg. Chem. 2014, 4642–4649
4642
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
ificity of the acireductone dioxygenase cleavage reactions
are therefore warranted.
The original explanation for the difference in regiospec-
ificity exhibited by Fe-ARDЈ and Ni-ARD inferred that
changes in the binding mode of the acireductone to the
metal center, from a five-membered chelate ring in Fe-
ARDЈ to a six-membered chelate ring in Ni-ARD, would
respectively activate the C(1) and C(2) or C(1) and C(3)
carbon atoms towards reaction with dioxygen
[
17]
(Scheme 1).
Subsequent collapse of the resulting dioxet-
ane rings would lead to the observed regiospecificity of the
products. This “chelate hypothesis” was supported by
NMR studies of the tertiary structures of the enzymes,
wherein it was found that the position of a tryptophan resi-
due in the Ni-ARD active site could favor coordination of Scheme 2. Reactivity of 1 and 2 with O2.
[
18]
the six-membered chelate binding mode of the substrate.
However, direct spectral analyses of the enzyme–substrate
ES) adducts, including by UV/Vis spectroscopy and XAS,
methylthiopentene), presumably due to the presence of the
(
[
25]
methylthio unit.
As a target we have chosen 3-oxo-3-
did not give conclusive evidence of a change in binding
mode between Fe-ARDЈ and Ni-ARD.[19] In addition, a re-
phenylpropene-1,2-diol (phenyl acireductone), previously
reported as a substrate for the enzyme, which has UV/Vis
cent mixed QM/DMD study of the acireductone dioxygen-
ases found that there does not appear to be a difference
properties that are amenable for spectroscopic studies (λ
max
[
19]
in binding mode between the two enzymes.[
20]
= 320 nm as a monoanion).
Previously reported syn-
Rather, it is
II
thetic routes to phenyl acireductone focused on generating
a phosphorylated precursor and subsequently using the E1
enolase/phosphatase enzyme from the MSP to dephospho-
proposed that the lower occupation of the Fe d orbitals
II
(
by two electrons compared with Ni ) and the greater redox
II
flexibility of Fe facilitates a key O–O bond-cleavage step
that is not accessible to Ni-ARD.
[
25]
rylate this precursor.
The resulting acireductone
monoanion was then employed in situ for enzymatic stud-
ies. The difficulty in isolating phenyl acireductone is a result
of its propensity to oxidize under the alkaline conditions
used in its synthesis.
To generate model systems of enhanced relevance to the
acireductone dioxygenases, our goal was to generate nickel
and iron complexes containing a coordinated dianionic
phenyl acireductone moiety. Our strategy is outlined in
Scheme 3 and involves the generation of a protected acired-
uctone. We hypothesized that the binding of this protected
acireductone to a metal center would allow us to generate
a well-defined complex. Subsequent deprotection and expo-
As an alternative to enzymatic studies as a method for
investigating the mechanistic details of the acireductone di-
oxygenases, we synthesized functional small molecular
model complexes of the proposed enzyme–substrate ad-
ducts of Ni-ARD and Fe-ARDЈ (1 and 2, respectively,
[
21,22]
Scheme 2).
Upon exposure to O , these synthetic com-
2
plexes undergo dioxygenolytic carbon–carbon bond cleav-
[
22,23]
age via an intermediate trione/hydroperoxide pair.
was found that the regioselectivity of the reaction of 1 and
with O was identical in dry solvent, consistent with the
It
2
2
chelate hypothesis. However, in the presence of H O a
2
change in regioselectivity was observed in the reaction of 2,
but not of 1, which suggests that the chelate hypothesis is
not sufficient to explain the chemistry in these model sys-
sure to O would then enable us to investigate the role that
2
the metal center plays in directing the regioselectivity of aci-
[
22]
tems. We therefore proposed a new reaction pathway for
in which the ferrous center promotes hydration of the tri-
2
ketone intermediate, thereby facilitating the change in re-
gioselectivity (Scheme 2).
Unfortunately, this model system is not truly biomimetic
as it uses an acireductone that is not a substrate for the
acireductone dioxygenase enzymes and involves O reacti-
2
vity at the monoprotonation level of the coordinated model
[
10]
substrate.
Computational studies have additionally sug-
gested significant differences in the electronic structures and
reactivities of acireductones that have a phenyl group or a
hydrogen atom at the C(1) position.[
23,24]
To provide better
insight into the acireductone dioxygenase reaction path-
ways, we endeavored to use a C(1)-H acireductone that is a
substrate for the native enzyme. To date there are no re-
ported non-enzymatic syntheses of the native substrate of
Scheme 3. Strategy for generating a protected C(1)-H acireductone
complex and its subsequent deprotection to investigate the reactiv-
the acireductone dioxygenases (1,2-dihydroxy-3-oxo-S- ity of a coordinated dianionic acireductone with O
2
.
Eur. J. Inorg. Chem. 2014, 4642–4649
4643
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
II
reductone cleavage. The results of this approach for Ni are vealed the presence of 0.5 equiv. of CH Cl , which was also
2
2
1
described herein.
confirmed by a H NMR spectrum of the elemental analy-
sis sample. This solvent was retained despite trituration and
extensive drying of the solid. We note that other transition-
Results
metal complexes of the 6-Ph TPA chelate ligand similarly
2
[
26,27]
exhibit a strong affinity for CH Cl .
2
2
Synthesis of Protected Phenyl Acireductone
Complex 5 was analyzed by HRMS and exhibits a [M –
+
ClO ] molecular ion, in excellent agreement with the calcu-
4
3-Phenyl-2-propyn-1-ol was acylated using acetic anhy-
lated exact mass and isotopic distribution (see Figure S5 in
the Supporting Information). The UV/Vis absorption spec-
trum of 5 contains an absorption band arising from the
coordinated enolate with a maximum at 350 nm. Consistent
with a proposed five-membered-ring coordination motif,
the FTIR spectrum of 5 contains a carbonyl band at
dride to generate 3-phenyl-2-propynyl acetate (3; Scheme 4;
see Figures S1 and S2 in the Supporting Information). Oxi-
dation of the alkyne 3 with NaIO using RuCl as a catalyst
generated the diketone 4 (Scheme 4; see Figures S3 and S4).
Careful monitoring of this reaction by TLC was necessary
due to the propensity of 4 to over-oxidize to give carboxylic
acids, presumably by an initial hydrolytic cleavage of the
acetyl protecting group. Owing to the formation of byprod-
ucts, purification by column chromatography was required.
The diketone tautomeric form of 4 was confirmed by the
4
3
–
1
1
749 cm , attributable to the acetyl ester group, and a band
–1
at 1600 cm , due to the non-enolized ketone carbonyl
group vicinal to the phenyl ring. The H NMR spectrum of
1
5
(see Figure S6), collected using paramagnetic parameters,
1
3
exhibits features consistent with those exhibited by other
presence of two ketone C NMR signals (Figure S4), an
integral of two protons for the methylene group by H
NMR, and by a UV absorption band at 255 nm.
II
1
Ni –enolate complexes containing 6-Ph TPA as an ancil-
2
[
21,28]
lary ligand.
Anaerobic Reactivity
Addition of 5 equiv. of NaOCH to a solution of 5 in
3
MeOH under anaerobic conditions led to the decay of the
3
50 nm absorption with concomitant growth of a new ab-
–1
–1
sorption band centered at 390 nm (ε = 5400 m cm , as-
suming that conversion is to a single absorbing species; Fig-
ure 1). This spectral change is consistent with nucleophilic
substitution of the acetyl group to generate methyl acetate
and a new compound 6 containing a coordinated dianionic
acireductone (Scheme 6). In contrast, addition of a strong,
Scheme 4. Synthesis of 1-acetoxy-3-phenylpropane-2,3-dione (4), a
protected acireductone.
non-nucleophilic base, such as LiHMDS in CH CN, did
not lead to any change in the position of the 350 nm ab-
3
sorption maximum. We note that the conversion of 5 into
6
does not proceed with an isosbestic point, which suggests
II
Synthesis and Characterization of the Ni Complex
Deprotonation of 4 by LiHMDS in Et O led to the for-
2
mation of a pale-yellow slurry. Combination of this with
[23]
[
(6-Ph TPA)Ni(CH CN) ](ClO )
{6-Ph TPA = N,N-
2
3
2
4 2
2
bis[(6-phenyl-2-pyridyl)methyl]-N-(2-pyridylmethyl)amine}
led to the formation of [(6-Ph TPA)Ni{PhC(O)C(O)-
2
CHOC(O)CH }]ClO (5), which was isolated in moderate
3
4
yield as a powder (Scheme 5). Attempts to produce X-ray-
quality single crystals of 5 have thus far not proven success-
ful. Elemental analysis of the powdered sample of 5 re-
Figure 1. UV/Vis spectra showing the effect of the addition of
Scheme 5. Synthesis of 5.
NaOCH
3
to a methanolic solution of 5 under anaerobic conditions.
Eur. J. Inorg. Chem. 2014, 4642–4649
4644
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
that this conversion may involve the formation of an inter- nances consistent with the presence of the free 6-Ph TPA
2
mediate that is spectroscopically similar to 5 and 6. Re- ligand (see Figure S8), albeit broadened due to the presence
moval of the solvent from the reaction of 5 with NaOCH3 of a paramagnetic species in solution. These results suggest
under reduced pressure and subsequent analysis of the the initial formation of a new acireductone dianion complex
product by solid-state FTIR spectroscopy in a KBr pellet in which the 6-Ph TPA ligand is coordinated to the metal
2
–1
revealed the disappearance of the band at 1749 cm , con- center, followed by its subsequent demetalation. The in-
sistent with removal of the acetyl group and its subsequent volvement of a transient intermediate containing the coor-
volatilization as methyl acetate. When the reaction was car- dinated 6-Ph TPA ligand is consistent with our anaerobic
2
ried out at higher concentrations, a white precipitate of Na- UV/Vis studies, which suggest the formation of an interme-
ClO was also formed as the reaction proceeds.
diate in the conversion of 5 into 6. Overall, these studies
suggest that 6 is a Ni (enediolate) species that could rea-
4
II
sonably be either a mononickel solvate species or a polynu-
[
29]
clear cluster, as has been observed previously.
The dis-
placement of the ancillary ligand in both systems is as-
sumed to be due to the poor affinity of the phenylpyridyl
II
donors of 6-Ph TPA for Ni relative to the affinity of the
2
enolate moieties of the acireductone dianion. Notably,
paramagnetically shifted chelate ligand resonances were not
observed when 5 was deprotected under O (see Figure S9).
2
II
Scheme 6. Reaction of 5 with excess NaOCH
3
under anaerobic This suggests that if a 6-Ph TPA-ligated Ni –acireductone
2
conditions.
dianion complex is initially formed, it is rapidly intercepted
by O2.
Evidence that 6 contains a dianionic form of phenyl aci-
reductone comes from a comparison of the observed spec-
tral features with those reported in previous studies of this
[
19]
acireductone.
Specifically, the 40 nm bathochromic shift
observed upon treatment of 5 with NaOCH is exactly the
3
same magnitude of shift as was previously found for the
conversion of the free phenyl acireductone mononanion
(
λmax = 320 nm) into the dianion (λmax = 360 nm) in phos-
[19]
phate buffer under anaerobic conditions.
An identical
Scheme 7. Previously reported reactivity of 1 with excess base re-
4
0 nm redshift is also observed when the phenyl acireduc-
sulting in the formation of a hexanickel enediolate cluster species.
tone monoanion is exposed to the Ni-ARD enzyme under
anaerobic conditions, which was taken as evidence of a co-
Additional efforts to characterize 6 were made by using
ordinated acireductone dianion in the enzyme–substrate ESI-MS. Analysis of an anaerobic CH
complex. The observed difference in absorption maxima be- within 10 min following the addition of NaOCH
OH solution of 5
(5 equiv.)
3
3
tween 6 and the ES complex of Ni-ARD (390 vs. 360 nm, did not give a molecular ion consistent with either the de-
+
respectively) is likely due to the differences in the primary protected dianion-bound form of 5 (expected [M] : m/z =
II
II
coordination environment of the Ni center.
663) or with the formation of Ni (enediolate) cluster spe-
cies that could be identified. Instead, only isotope clusters
consistent with the presence of the starting material 5 (m/z
1H NMR Studies
+
+
=
705), [H(6-Ph TPA)] and [Na(6-Ph TPA)] species, and
2 2
oxidation products (see below) were identified.
We previously reported that the treatment of 1
(Scheme 2) with additional base under anaerobic conditions
resulted in the loss of the 6-Ph TPA chelate ligand and the
2
Aerobic Reactivity
formation of a hexanickel enediolate cluster species
[
29]
(
Scheme 7). We followed up our anaerobic UV/Vis stud-
Exposure of CH CN or MeOH solutions of 5 to O did
3
2
1
ies of the present system with H NMR investigations to not lead to the decomposition of the complex over the
1
evaluate the chelate ligand binding properties of the C(1)- course of 24 hours, as revealed by UV/Vis and H NMR
H-containing phenyl acireductone dianion species 6. Treat- spectroscopy. This is the expected result because the coordi-
ment of 5 with 5 equiv. of NaOCH under anaerobic condi- nated phenyl acireductone should not react with O due to
3
2
tions in CD OD led to the initial growth of a new species the presence of the acetyl protecting group. However, once
3
with resonances at δ = 49.2 and 46.2 ppm, followed by the deprotected, this complex should react rapidly with O . In
2
gradual disappearance of all of the paramagnetically shifted this regard, a methanolic solution of 5 was deprotected by
peaks associated with the methylene and pyridyl ring pro- the addition of 5 equiv. of NaOCH under anaerobic condi-
3
tons of the 6-Ph TPA ligand (see Figure S7 in the Support- tions to form 6. Once this reaction was complete, as moni-
2
ing Information). Analysis of the diamagnetic region of the tored by UV/Vis spectroscopy, the resulting solution was
spectrum recorded at the end of the reaction showed reso- purged with O . This led to the rapid decay of the absorp-
2
Eur. J. Inorg. Chem. 2014, 4642–4649
4645
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
tion at 390 nm (Figure 2, left), consistent with oxidative had proceeded to completion showed that 0.85 equiv. of CO
cleavage of the acireductone dianion. As an alternative ap- had been produced in the reaction (Scheme 8). CO is the
proach to examining O reactivity, a methanolic solution of expected product if both the C(1)–C(2) and C(2)–C(3)
2
5
was prepared and purged with O . This solution was then bonds of the phenyl acireductone substrate are cleaved. The
2
mixed with 5 equiv. of NaOCH and the subsequent decay yield of CO gas suggests that this cleavage reaction is the
3
monitored by UV/Vis spectroscopy (Figure 2, right). The major reaction pathway, which is consistent with the forma-
+
observed decay of the absorption at 350 nm, without any tion of the [(6-Ph TPA)Ni(O CH)] and [(6-Ph TPA)Ni(-
2
2
2
+
significant growth of the peak at 390 nm, suggests a rapid O CPh)] species as the major products, as were detected
oxidative cleavage of the acireductone dianion as it is by ESI-MS and H NMR spectroscopy. Additional evi-
2
1
formed.
dence for the formation of the carboxylic acid products was
obtained following removal of the Ni ion from the reac-
II
tion mixture and analysis of the organic products. Formate
was detected by using a previously reported method for the
detection of low-molecular-weight carboxylates involving
[
13]
derivatization with 4Ј-phenylphenacyl bromide and ben-
zoic acid was detected by GC–MS. Notably, no evidence
was found by GC–MS of a tricarbonyl-type intermediate
(
benzoylglyoxal) akin to that involved in reaction the reac-
tion of complex 1 bearing the bulky acireductone
-hydroxy-1,3-diphenylpropane-1,3-dione as ligand
In addition, no evidence was found for the
2
[
22,23]
(Scheme 2).
formation of benzoylformic acid, which would be the ex-
pected product if only C(1)–C(2) cleavage (Fe-ARDЈ-type)
reaction occurred.
Figure 2. Left: Spectral change produced upon the introduction of
O
2
into a solution of 5 in MeOH that had previously been treated
with 5 equiv. of NaOCH ; a decrease in the band at 390 nm is ob-
served. Right: Spectral change produced upon treatment of an O
3
2
-
3
purged solution of 5 with 5 equiv. of NaOCH ; a decrease in the
band at is observed.
In the presence of O , the hexanickel enediolate cluster
2
Scheme 8. Products of the oxidative cleavage reaction resulting
from treatment of 5 with NaOCH in the presence of O .
3 2
species produced upon treatment of 1 with base underwent
II
reaction to yield a Ni –benzoate species and CO. The for-
mer exists as a coordination complex with the 6-Ph TPA
2
[21]
chelate ligand, [(6-Ph TPA)Ni(O CPh) ]. After extended
2
2
2
Discussion
stirring under oxygen, methanol solutions of 6 were ana-
lyzed by ESI-MS and found to contain isotope clusters con-
At the outset of this study, our goal was to develop an
approach for generating a mononuclear Ni complex of a
protected C(1)-H acireductone that upon deprotection
would enable reactivity studies of relevance to the enzy-
matic chemistry of Ni-ARD. In this regard, we have suc-
ceeded in developing a high-yielding, simple synthetic route
sistent
with
the
formation
of
[(6-Ph TPA)Ni-
2
II
+
+
(
O CH)] and [(6-Ph TPA)Ni(O CPh)] species (see Fig-
2
2
2
ures S10 and S11 in the Supporting Information), along
+
with [H(6-Ph TPA)] . Divalent nickel formate and benzoate
2
complexes containing these cationic species have previously
been prepared and comprehensively characterized in our
II
to a protected C(1)-H acireductone precursor and the Ni
[
30] 1
laboratory.
H NMR analysis of the reaction product
complex (5) of its monoanion. We note that this methodol-
ogy should be extendable to the synthesis of a variety of
protected C(1)-H acireductones and thus represents a sig-
nificant step forward in the synthetic modeling of acireduc-
tone dioxygenases.
mixture provided additional evidence for the presence of
these carboxylate species with characteristic paramagneti-
cally shifted resonances in the range of 10–50 ppm (Fig-
ure S12). Importantly, the observation of these oxidative
II
cleavage products suggests Ni -ARD-type regioselective
We found that the treatment of 5 with NaOCH to de-
protect the acireductone moiety, followed by exposure of a
3
C(1)–C(2) and C(2)–C(3) cleavage with release of CO.
solution of the complex to O , results in the formation of
2
the same products as a Ni-ARD-type reaction (CO, form-
ate, benzoate). This is significant as it is the first example
Organic Product Analysis
Analysis of the head-space gas once the reaction of 5 of a model system that undergoes oxidative cleavage of a
with 5 equiv. of NaOCH in MeOH in the presence of O2 C(1)-H acireductone substrate to give products akin to
3
Eur. J. Inorg. Chem. 2014, 4642–4649
4646
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
II
those produced by the Ni -containing acireductone dioxy- or by using standard Schlenk techniques. The 6-Ph
2
TPA {N,N-
bis[(6-phenyl-2-pyridyl)methyl]-N-(2-pyridylmethyl)amine} ligand
genase enzyme. Notably, in the absence of a metal ion, the
dianionic form of phenyl acireductone undergoes oxidative
[
32]
was synthesized according to a previously published procedure.
cleavage to give the products of C(1)–C(2) cleavage (formate Physical Methods: H and 13C NMR spectra of organic compounds
1
and benzoylformic acid).[ The fact that no benzoylformic were recorded by using a Bruker ARX-400 spectrometer; chemical
19]
1
shifts are referenced to the residual solvent peak in CD
2
HCN ( H
acid is found in the product mixture starting from 5 indi-
cates that the Ni center in this system influences the re-
gioselectivity of the reaction. These results distinguish this
chemistry from our prior studies of a bulky acireductone
that was predisposed to C(1)–C(2) and C(2)–C(3) (Ni-
ARD-type) cleavage chemistry. In our former studies, it was
1
3
1
II
NMR: 1.94 ppm, quintet; C: 1.39 ppm, quintet). H NMR spec-
tra of paramagnetic complexes were collected by using the above
spectrometer and parameters as described previously.[ UV/Vis
data were collected by using a Hewlett-Packard HP8453A spec-
trometer at ambient temperature. IR spectra were recorded with a
Shimadzu FTIR-8400 spectrometer as neat oils or KBr pellets.
GC–MS data was obtained with a Shimadzu GC/MS-QP5000 gas
28]
II
only under wet conditions in the presence of Fe that we
found evidence for metal-dependent regioselective carbon– chromatograph/mass spectrometer with a GC-17A gas chromato-
carbon bond cleavage. This led us to propose a mechanistic graph, using an Alltech EC5 30 mϫ 25 mmϫ 25 μm thin-film cap-
II
pathway in which the Fe center had an enhanced ability to illary column and the following temperature program: Tinitial: 70 °C
–
1
(
5 min); temperature gradient: 23 °Cmin ; TFinal: 250 °C (10 min).
promote the hydration of a vicinyl triketone intermediate,
leading to a differentiation in regioselectivity.[22] In the cur-
GC-TCD data of the reaction head-space gas were collected by
using an Agilent 3000A Micro gas chromatograph. Mass spectro-
scopic data were collected by the Mass Spectrometry Facility, Uni-
versity of California, Riverside. Elemental analyses of compounds
II
rent system, in the absence of water, the Ni center has a
directing effect on the chemistry, which makes this model
II
system particularly relevant to Ni -ARD.
3
–5 were performed by Atlantic Microlabs Inc., Norcross, GA.
In terms of the reaction pathway leading to carbon–car-
bon bond cleavage, we note that the lack of an observed
tricarbonyl intermediate (benzoylglyoxal) does not ex-
plicitly rule out the involvement of such a species. However,
it may be difficult to detect this possible intermediate due
3
-Phenyl-2-propynyl Acetate (3): 3-Phenyl-2-propyn-1-ol (1.00 g,
7
.57 mmol) and NEt
3
(1 mL) were dissolved in Ac
. The resulting orange solution was diluted
with EtOAc (150 mL) and washed repeatedly with 100 mL portions
of H O. The organic layer was passed through an activated char-
2
O (5 mL) and
stirred for 16 h under N
2
2
to the lack of a facile benzoyl migration that would produce coal filter and dried with anhydrous Na SO . The solvent was then
2
4
a stable side-product (phenylglyoxal) (see Scheme S1 in the removed under reduced pressure to yield a yellow oil (1.10 g, 83%).
1
Supporting Information).
3
H NMR (400 MHz, CD CN): δ = 7.46–7.35 (m, 5 H), 4.87 (s, 2
1
3
H), 2.07 (s, 3 H) ppm. C NMR (100 MHz, CD
132.54, 129.94, 129.58, 122.90, 86.47, 84.54, 53.23, 20.88 ppm.
FTIR (neat): ν˜ = 2234 (νCC), 1734 (νCO) cm . C11
calcd. C 75.84, H 5.79; found C 75.77, H 5.83.
3
CN): δ = 171.01,
A complicating challenge in the chemistry of synthetic
II
Ni –acireductone dianion species is supporting chelate li-
–
1
H
10
O
2
(174.20):
gand loss, which in a prior system resulted in the formation
of a hexanickel–enediolate cluster species that reacted with
[29]
1-Acetoxy-3-phenylpropane-2,3-dione
.00 mmol) was dissolved in H
RuCl (0.037 mmol) to form a yellow solution. 3-Phenyl-2-propynyl
acetate (3; 0.215 g, 1.23 mmol) dissolved in a mixture of CCl
10 mL) and CH CN (10 mL) was added to the aqueous solution,
and the resulting slurry stirred for 15 min, monitoring carefully by
TLC. The slurry was then diluted with CH Cl (150 mL) and fil-
tered. The filtrate was cooled in an ice bath and then carefully
(4):
NaIO4
(1.07 g,
O . Upon deprotection of 5 under anaerobic conditions,
2
5
2
O (15 mL) and combined with
we found preliminary evidence in terms of paramagnetically
1
3
shifted H NMR chelate ligand resonances to suggest that
4
the formation of 6 may initially involve the generation of
(
3
a 6-Ph TPA-ligated acireductone dianion complex, which
2
subsequently undergoes loss of the chelate ligand, likely to
2
2
II
give a Ni (enediolate) cluster species. Elucidation of the
structural features of this 6-Ph TPA-ligated complex will mixed with Na SO (100 mL, 1.0 m). The resulting suspension was
2
2
3
depend upon our ability to stabilize it relative to loss of the placed in a separatory funnel and the aqueous layer extracted with
chelate ligand. Experiments directed towards addressing CH Cl (3ϫ 100 mL). The organic layers were combined, dried
with anhydrous Na SO , and the solvent removed under reduced
2
2
this issue are underway in our laboratory. In addition, the
nature of the cluster species that may form and play a role
2
4
pressure. The crude yellow product mixture was purified by column
chromatography using a silica gel solid phase and eluting with 4:1
hexanes/EtOAc (180 mg, 71%). H NMR (300 MHz, CD
8.01 [d, J(H,H) = 8.2 Hz, 2 H], 7.73 [t, J(H,H) = 7.2 Hz, 1 H],
.57 [t, J(H,H) = 7.9 Hz, 2 H], 5.21 (s, 2 H), 2.13 (s, 3 H) ppm.
C NMR (100 MHz, CD
32.94, 131.07, 129.92, 67.19, 20.49 ppm. FTIR (neat): ν˜ = 1753
(νCO), 1730 (νCO), 1669 (ν ) cm . UV/Vis (Et O): λ (ε) =
max
in the reactivity of 6 towards O is also under investigation.
2
1
3
CN): δ
Although the characterization of 6 is a key ongoing chal-
lenge, the results presented herein conclusively demonstrate
the feasibility of a new synthetic approach that will enable
examination of the oxidative chemistry of C(1)-H acireduct-
one dianion complexes.
3
3
=
3
7
13
3
CN): δ = 196.23, 191.26, 171.24, 135.95,
1
–
1
CO
2
–
1
–1
2
4
10 4
55 nm (12000 m cm ). C11H O (206.20): calcd. C 64.07, H
.89; found C 64.19, H 5.27.
Experimental Section
Caution! Perchlorate salts of metal complexes with organic ligands
are potentially very explosive. Only small amounts of material should
General Methods: All reagents were obtained from commercial
sources and used without additional purification unless stated
otherwise. Solvents were dried according to published procedures
and purified by distillation under N
[
33]
be prepared, and these should be handled with extreme caution.
[(6-Ph TPA)Ni{PhC(O)C(O)CHOC(O)CH }]ClO (5): A CH
(5 mL) solution of [(6-Ph TPA)Ni(CH CN) ](ClO (0.078 mmol)
was stirred with 1-acetoxy-3-phenylpropane-2,3-dione (16 mg,
2
3
4
3
CN
prior to use.[ Air-sensitive
31]
2
2
3
2
4 2
)
reactions were performed in an MBraun Unilab glovebox under N
2
Eur. J. Inorg. Chem. 2014, 4642–4649
4647
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
0
.078 mmol) under N
tion was then added to LiHMDS (14 mg, 0.082 mmol) in Et
1 mL) and stirred overnight. The solvent was then removed under
reduced pressure and the solid redissolved in CH Cl and the solu-
tion filtered through a glass wool/Celite plug to remove insoluble
LiClO . The CH Cl solution was concentrated under reduced
pressure and then layered with hexanes to yield a pale-yellow pre-
cipitate. The precipitate was collected, triturated with Et O, and
then dried under vacuum for 48 h (36 mg, 54%). FTIR (KBr): ν˜ =
749 (νCO), 1609 (νCO), 1451, 1355, 1094 (νClO4), 765, 623
2
until it was completely dissolved. This solu-
NaOCH
3
(0.05 mmol) in MeOH (1.0 mL) was injected into the
2
O
round-bottomed flask by using a gas-tight syringe and the resulting
solution stirred for 12 h. The head-space gas (10 mL) was removed
by means of a gas-tight syringe and analyzed by GC-TCD. The
yield of CO generated in the reaction was determined from a cali-
(
2
2
4
2
2
bration curve generated by using gas mixtures of O
2
and CO.
Supporting Information (see footnote on the first page of this arti-
2
1
13
cle): H and C NMR and mass spectra; schematic description of
a possible reaction pathway for benzoylglyoxal.
1
ClO4) cm . UV/Vis (MeOH): λmax (ε) = 350 (6100 m–1 cm ).
–1
–1
(ν
C
41
H
35ClN NiO ·0.5CH Cl
4
8
2
2
: calcd. C 58.74, H 4.28, N 6.61; found Acknowledgments
+
35 4 4
C 59.17, H 4.67, N 7.09. HRMS (ESI): calcd. for [C41H N NiO ]
+
705.2012 [M – ClO
4
] ; found 705.2010.
This research was supported by the National Science Foundation
NSF) (grant numbers CHE-0848858 and CHE-1301092 to
(
Treatment of 5 with NaOCH
3
: For UV/Vis experiments, an approxi-
L. M. B.). C. A. and L. M. B. thank Jay Riley Argue and Sushma
Saraf for technical assistance.
mate 0.2 mm MeOH stock solution of 5 was prepared. A 2.4 mL
aliquot of this solution (ca. 0.48 μmol) was placed in a quartz UV/
Vis cell and then combined with a 200 μL aliquot of a 12 mm solu-
tion of NaOCH
μmol) was dissolved in CD
15 μmol) also dissolved in CD OD (0.25 mL). The two solutions
were then combined.
3
(2.4 μmol). For NMR experiments, 5 (ca. 2.5 mg,
3
(
3
OD (0.75 mL) and NaOCH
3
[
[
1] a) J. D. Finkelstein, J. Nutr. Biochem. 1990, 1, 228–237; b) C. C.
Valley, A. Cembran, J. D. Perlmutter, A. K. Lewis, N. P. Lab-
ello, J. Gao, J. N. Sachs, J. Biol. Chem. 2012, 287, 34979–34991;
c) P. Cavuoto, M. F. Fenech, Cancer Treat. Rev. 2012, 38, 726–
3
Treatment of 5 with LiHMDS: An approximate 0.2 mm CH
stock solution of 5 was prepared. A 2.4 mL aliquot of this solution
ca. 0.48 μmol) was placed in a quartz UV/Vis cell. This solution
3
CN
736; d) T. Nakamoto, Gene 2009, 432, 1–6.
2] M. A. Pajares, G. D. Markham, in: Advances in Enzymology
and Related Areas of Molecular Biology (Ed.: E. J. Toone),
Wiley, Hoboken, NJ, 2011, vol. 78, p. 449–521.
(
was then combined with a 200 μL aliquot of a 12 mm Et
of LiHMDS (2.4 μmol).
2
O solution
[3] M. H. Stipanuk, Ann. Rev. Nutr. 1986, 6, 179–209.
[
[
4] W. A. M. Loenen, Biochem. Soc. Trans. 2006, 34, 330–333.
5] a) H. L. Schubert, R. M. Blumenthal, X. Cheng, Trends Bio-
chem. Sci. 2003, 28, 329–335; b) A.-W. Struck, M. L. Thomp-
son, L. S. Wong, J. Micklefield, ChemBioChem 2012, 13, 2642–
2655.
O
2
Reactivity: Solutions for monitoring by UV/Vis or 1H NMR
were prepared as described above. Oxygen was introduced by purg-
ing the solutions with dry O gas for 30 s, and then the reaction
vessels were sealed. In an alternative set of experiments, solutions
of 5 were purged with O and then subsequently treated with
equiv. of NaOCH
2
2
[6] a) P. A. Frey, A. D. Hegeman, F. J. Ruzicka, Crit. Rev. Biochem.
Mol. Biol. 2008, 43, 63–88; b) K. A. Shisler, J. B. Broderick,
Curr. Opin. Struct. Biol. 2012, 22, 701–710.
5
3
.
Identification of Organic Products: To analyze the products of the
reaction, a MeOH solution containing 5 (2 mg/mL, ca. 2.5 mm)
was prepared. The solution was then purged with O for 30 s. The
aerated solution was then combined with solid NaOCH (0.6 mg
per mL of solution) to give a NaOCH concentration of ca.
2.5 mm (5 equiv.). The solution was then repurged with O for
0 s, sealed with a rubber septum, and stirred for 12 h. The solvent
[7] M. Sauter, B. Moffatt, M. C. Saechao, R. Hell, M. Wirtz, Bio-
O
2
chem. J. 2013, 451, 145–154.
2
[8] a) M. E. Martinez, T. G. O’Brien, K. E. Fultz, N. Babbar, H.
Yerushalmi, N. Qu, Y. Guo, D. Boorman, J. Einspahr, D. S.
Alberts, E. W. Gerner, Proc. Natl. Acad. Sci. USA 2003, 100,
3
3
7
859–7864; b) R. M. Pascale, M. M. Simile, M. R. De Miglio,
1
3
2
F. Feo, Alcohol 2002, 27, 193–198.
[
9] a) E. Albers, IUBMB Life 2009, 61, 1132–1142; b) A. Sekow-
ska, V. Dénervaud, H. Ashida, K. Michoud, D. Haas, A. Yok-
ota, A. Danchin, BMC Microbiol. 2004, 4, 9; c) M. A. Avila,
E. R. Gar c´ ia-Trevijano, S. C. Lu, F. J. Corrales, J. M. Mato,
Int. J. Biochem. Cell Biol. 2004, 36, 2125–2130.
10] T. C. Pochapsky, T. Ju, M. Dang, R. Beaulieu, G. M. Pagani,
B. OuYang, in: Metal Ions in Life Sciences (Eds.: A. Sigel, H.
Sigel, R. K. O. Sigel), Wiley-VCH, Weinheim, Germany, 2007,
vol. 2, p. 473–500.
was then removed under reduced pressure and the products ana-
lyzed as described below.
Detection of Formate: The crude reaction mixture was redissolved
in a mixture of CH CN (10 mL) and benzene (10 mL). 4Ј-Phenyl-
3
phenacyl bromide (10 equiv.) and 18-crown-6 (5 equiv.) were then
added to the reaction mixture. The resulting slurry was heated at
reflux under nitrogen for 18 h. The solvent was then removed under
[
2 2 2 2
reduced pressure and the solid extracted with CH Cl . The CH Cl
[
[
11] M. J. Maroney, S. Ciurli, Chem. Rev. 2014, 114, 4206–4228.
12] C. J. Allpress, L. M. Berreau, Coord. Chem. Rev. 2013, 257,
3005–3029.
solution was then passed through a short plug of silica and ana-
lyzed by GC–MS. 4Ј-Phenylphenacyl formate was detected in the
reaction mixture.
[13] J. W. Wray, R. H. Abeles, J. Biol. Chem. 1995, 270, 3147–3153.
[
14] J. W. Wray, R. H. Abeles, J. Biol. Chem. 1993, 268, 21466–
1469.
15] Y. Day, P. C. Wensink, R. H. Abeles, J. Biol. Chem. 1999, 274,
193–1195.
Detection of Benzoic Acid: The crude reaction mixture was redis-
solved in a small amount of CH CN and passed through a short
3
silica column, eluting with ethyl acetate. The organic products were
then analyzed by GC–MS and benzoic acid was identified by com-
parison of the molecular ion, fragmentation pattern, and retention
time with those of an authentic sample.
2
[
1
[16] R. Motterlini, L. E. Otterbein, Nat. Rev. Drug Discovery 2010,
9, 728–743.
[17] T. C. Pochapsky, S. S. Pochapsky, T. Ju, H. Mo, F. Al-Mjeni,
M. J. Maroney, Nat. Struct. Biol. 2002, 9, 966–972.
Analysis of Gaseous Products: Complex 5 (0.01 mmol) was dis-
solved in MeOH (1.0 mL) and the solution placed in a 50 mL
round-bottomed flask equipped with a stirring bar. The flask was
[
18] a) T. Ju, R. B. Goldsmith, S. C. Chai, M. J. Maroney, S. S. Po-
chapsky, T. C. Pochapsky, J. Mol. Biol. 2006, 363, 823–834; b)
T. C. Pochapsky, S. S. Pochapsky, T. Ju, C. Hoefler, J. Liang, J.
Biomol. NMR Spectroscopy 2005, 34, 117–127; c) S. S. Pochap-
purged with O
2
and sealed with a rubber septum. A solution of
Eur. J. Inorg. Chem. 2014, 4642–4649
4648
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
sky, J. C. Sunshine, T. C. Pochapsky, J. Am. Chem. Soc. 2008,
30, 2156–2157.
19] a) Y. Dai, T. C. Pochapsky, R. H. Abeles, Biochemistry 2001,
0, 6379–6387; b) S. C. Chai, T. Ju, M. Dang, R. B. Goldsmith,
M. J. Maroney, T. C. Pochapsky, Biochemistry 2008, 47, 2428–
438.
20] M. Sparta, C. E. Valdez, A. N. Alexandrova, J. Mol. Biol. 2013,
25, 3007–3018.
21] E. Szajna, A. M. Arif, L. M. Berreau, J. Am. Chem. Soc. 2005,
27, 17186–17187.
[26] a) C. J. Allpress, A. M. Arif, D. T. Houghton, L. M. Berreau,
Chem. Eur. J. 2011, 17, 14962–14973; b) K. Grubel, K. Rudzka,
A. M. Arif, K. L. Klotz, J. A. Halfen, L. M. Berreau, Inorg.
Chem. 2010, 49, 82–96.
1
[
4
[
[
[
27] E. Szajna-Fuller, K. Rudzka, A. M. Arif, L. M. Berreau, Inorg.
Chem. 2007, 46, 5499–5507.
28] E. Szajna, P. Dobrowolski, A. L. Fuller, A. M. Arif, L. M.
Berreau, Inorg. Chem. 2004, 43, 3988–3997.
29] K. Rudzka, K. Grubel, A. M. Arif, L. M. Berreau, Inorg.
Chem. 2010, 49, 7623–7625.
2
[
[
[
[
4
1
22] C. J. Allpress, K. Grubel, E. Szajna-Fuller, A. M. Arif, L. M.
Berreau, J. Am. Chem. Soc. 2013, 135, 659–668.
[30] E. Szajna-Fuller, B. M. Chambers, A. M. Arif, L. M. Berreau,
Inorg. Chem. 2007, 46, 5486–5498.
[31] W. L. F. Armarego, D. D. Perrin, Purification of Laboratory
Chemicals, 4th ed., Butterworth–Heinemann, Boston, 1996.
[32] M. M. Makowska-Grzyska, E. Szajna, C. Shipley, A. M. Arif,
M. H. Mitchell, J. A. Halfen, L. M. Berreau, Inorg. Chem.
2003, 42, 7472–7488.
23] L. M. Berreau, T. Borowski, K. Grubel, C. J. Allpress, J. P.
Wikstrom, M. E. Germain, E. V. Rybak-Akimova, D. L. Tier-
ney, Inorg. Chem. 2011, 50, 1047–1057.
[
[
24] T. Borowski, A. Bassan, P. E. M. Siegbahn, THEOCHEM
2006, 772, 89–92.
25] a) R. W. Myers, J. W. Wray, S. Fish, R. H. Abeles, J. Biol.
Chem. 1993, 268, 24785–24791; b) Y. Zhang, M. H. Heinsen, [33] W. C. Wolsey, J. Chem. Educ. 1973, 50, A335–A337.
M. Kostic, G. M. Pagani, T. V. Riera, I. Perovic, L. Hedstrom,
Received: March 31, 2014
Published Online: August 25, 2014
B. B. Snider, T. C. Pochapsky, Bioorg. Med. Chem. 2004, 12,
3847–3855.
Eur. J. Inorg. Chem. 2014, 4642–4649
4649
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim