New diamino-diheterophenol ligands coordinate iron(iii) to make structural and functional models of protocatechuate 3,4-dioxygenase

Article abstract of DOI:10.1039/c3dt53431f

Three diamino, dihetero-phenol ligands were synthesized by sequential Mannich condensations. These ligands were combined with FeCl₃ to produce three five-coordinate Fe(iii) complexes that are structural models for the enzyme 3,4-PCD. The three

Full text of DOI:10.1039/c3dt53431f

Dalton
Transactions
View Article Online
View Journal | View Issue
COMMUNICATION
New diamino-diheterophenol ligands coordinate
iron(III) to make structural and functional models
of protocatechuate 3,4-dioxygenase†
Joshua R. Farrell,*a Jonathan A. Niconchuk,a Peter R. Renehan,a
Christine S. Higham,a Eric Yoon,a Mark V. Andrews,a Janet L. Shaw,b Anil Cetin,b
James Engleb and Christopher J. Zieglerb
Cite this: Dalton Trans., 2014, 43,
6610
Received 5th December 2013,
Accepted 8th March 2014
DOI: 10.1039/c3dt53431f
www.rsc.org/dalton
Three diamino, dihetero-phenol ligands were synthesized by
sequential Mannich condensations. These ligands were combined
with FeCl3 to produce three five-coordinate Fe(III) complexes that
are structural models for the enzyme 3,4-PCD. The three Fe(III)
complexes were characterized by elemental analysis, single crystal
X-ray diffraction studies, UV-vis spectroscopy, and cyclic volta-
mmetry. Combining the Fe(III) complexes with 3,5-di-t-butylcatechol
and O2 resulted in oxidative cleavage similar to the function of
3,4-PCD.
Chart 1 Ligands and complexes in this study.
Protocatechuate 3,4-dioxygenase (3,4-PCD) is a well-studied
example of an intradiol dioxygenase whose role in nature is to
catalyze the ring cleavage of 3,4-dihydroxybenzoate, resulting
in β-carboxy-cis,cis-muconate.1–7 As a dioxygenase, 3,4-PCD
incorporates both atoms from molecular oxygen between the
vicinal hydroxyl groups of a catechol substrate. X-ray crystallo-
graphic data has shown that the active-site of 3,4-PCD consists of
a high-spin five-coordinate ferric ion bound by tyrosine and histi-
dine in the two axial positions and tyrosine, histidine, and
hydroxide ligands in the equatorial plane.8,9 Upon binding of
the catechol substrate, the active-site shifts to a square-pyramidal
geometry where the axial tyrosine and the hydroxide ligands
have been replaced by the double deprotonated catecholate.
Many iron(III) model complexes of the 3,4-PCD active-site
have been synthesized and characterized.10–19 A majority of the
early examples utilized nitrogen based ligand systems, but
more recently mixed nitrogen/phenolate ligands have been
used to try and model the structure and function of the
enzyme.13,16–18 Various groups have been able to either syn-
thesize good structural models (i.e. produce five-coordinate
iron(III) complexes with two nitrogen and two phenolate
donors)13,14,19,20 or produce monomeric iron(III) complexes
that oxidize molecules like 3,5-di-t-butylcatechol to match the
function of 3,4-PCD,16,17,20 but to our knowledge no group has
been able to combine these two roles in a single complex.
Palaniandavar and coworkers reported a series of iron(III)
complexes prepared from bis(phenolate)diamine ligands (1–4,
Chart 1).20 Use of ligand 3 led to the isolation of a monomeric
trigonal bipyramidal iron(III) complex, which unfortunately did
not exhibit any intradiol cleavage of catechol. The other three
ligands (1, 2, and 4) were used to make monomeric octahedral
iron(III) complexes that did exhibit oxidation chemistry similar
to 3,4-PCD. Our group has recently been preparing a series of
multiphenolate/amine ligands, which led us to wonder if we
could prepare mixed phenolate ligands to better model bio-
inorganic systems.21–23 Using stepwise Mannich condensations
we have prepared and characterized three heterophenolate/
diamine ligands 6–8 and used them to form the first mono-
meric trigonal bipyramidal iron(III) complexes that carry out
the intradiol cleavage of catechols.
The Mannich condensation is a multicomponent reaction
(MCR) between a non-enolizable aldehyde (in this case,
formaldehyde), a 1° or 2° amine, and an enolizable carbonyl
compound. Our group and many others have used 2,4-di-
substituted phenols as the enolizable carbonyl compound (which
limits reactivity to the 6-position of the phenol) to make
amino-diphenol complexes where both phenols are the same,
as in 3 and 4 in Chart 1.24–26 Having prepared a series of
aDepartment of Chemistry, College of the Holy Cross, Worcester, MA 01610, USA
bDepartment of Chemistry, University of Akron, Akron, OH 44325, USA.
E-mail: jfarrell@holycross.edu
†Electronic supplementary information (ESI) available: X-ray crystallographic
file (CIF) for 8–11, additional experimental, structural, and electrochemical
details. CCDC 961571, 961572, 961573 and 961574. For ESI and crystallographic
data in CIF or other electronic format see DOI: 10.1039/c3dt53431f
6610 | Dalton Trans., 2014, 43, 6610–6613
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Communication
refluxing methanol. Compounds 14 and 15 are prone to retro-
Mannich reactions, which can lead to scrambling of the
phenols. Phenol scrambling was minimized in the lower temp-
erature methanol reflux and becomes more problematic under
higher temperature conditions. Final compounds were isolated
by precipitation from methanol solutions. No chromatography
was required at any time during the synthesis. Only three
examples were prepared for this study, but 14 and 15 could
serve as starting materials for a large number of diamino
heterodiphenol ligands similar to 6–8. This synthetic process
provides greater control over the steric and electronic pro-
perties of this family of ligands and also introduces chirality
into any four-coordinate metal complexes. A single crystal
X-ray diffraction study confirmed the solid state structure of 8
and a thermal ellipsoid diagram and the crystallographic
details can be found in the ESI.†
Mononuclear trigonal bipyramidal iron(III) complexes 9–11
were prepared by combining FeCl3, triethylamine, and the
appropriate ligands 6, 7, or 8 in methanol. After refluxing for
two hours, the methanol was removed in vacuo. The resulting
purple solids were combined with acetonitrile, filtered, recrys-
tallized, and isolated in yields ranging from 67 to 82%. Com-
pounds 9–11 were isolated as FeClL complexes where L = 6, 7,
or 8 respectively. Single crystal X-ray structures indicate that
9–11 exist as distorted trigonal bipyramidal complexes, Fig. 1
and Table 1. Bond lengths, angles, and τ values for complexes
9–13 are compiled in Table S1 found in the ESI.† For each
complex the chloride and the internal nitrogen heteroatom
(N1 in Fig. 1) act as the axial ligands and the two phenols and
the substituted dimethylamine (N2 in Fig. 1) act as the equa-
torial ligands. Complexes 9–11 have similar structures to 1220
and 13,19 which contain either two 2,4-dimethylphenol or two
2-t-butyl-4-methylphenol moieties, respectively. The difference
between the two Fe–O bond lengths in 12 and 13 (which have
two identical phenols coordinated to iron) are 0.013–0.020 Å.
Complex 9, in which both phenol moieties have two alkyl sub-
stituents, exhibits similar behavior, but 10 and 11 both have
differences in Fe–O bond lengths of 0.27–0.36 Å, where the
Scheme 1 Synthetic approach for ligands 6–8.
tetrahydrosalens using Mannich condensations, instead of the
more common reductive amination, led us wonder if we could
prepare complexes such as 6–8 by a two-step Mannich conden-
sation to prepare hetero-diphenols. A single example of this
type of compound appears in the literature and it was prepared
by a statistical reaction where one equivalent of 2,4-dimethyl-
phenol and 2,4-di-t-butyl phenol were combined with N,N-
dimethylethylene diamine and formaldehyde resulting in
three products (3, 6, and a compound similar to 3 where the
methyl groups on the phenols are t-butyl groups).25 These
three compounds were then separated by a series of recrystalli-
zations and column chromatography. Our strategy was to syn-
thesize one-armed intermediates such as 14 and 15 (Scheme 1)
that could then be used to make a series of heterophenolate
diamines such as 6–8.
One armed diaminophenols 14 and 15 were prepared by
combining one equivalent of formaldehyde and N,N-dimethyl-
ethylenediamine in refluxing methanol for an hour followed
by addition of the appropriate phenol and refluxing overnight.
Compounds 14 and 15 are susceptible to further Mannich con-
densations. Side-product formation was minimized by the pre-
reaction of amine with formaldehyde, but 14 and 15 were iso-
lated in 55% and 38% yields respectively. The decreased solu-
bility of 14 compared to 15 likely leads to its formation in
higher yield as the product falls out of solution prior to under-
going deleterious further reactions. Compounds 14 and 15
were used as precursors for ligands 6–8, by reaction of the one-
armed diaminophenol with one equivalent of formaldehyde
and one equivalent of the appropriate second phenol in
Table 1 Crystal data and structural refinement parameters for 8–11
Compound
8
9
10
11
Chemical formula
FW
Crystal system
Space group
a/Å
b/Å
c/Å
α/°
C
20H26Cl2N2O2
C28H42ClFeN2O2
529.94
Monoclinic
P21/c
11.477(4)
17.017(6)
14.198(5)
90.00
C26H36Cl3FeN2O2
570.77
Monoclinic
P21/c
11.359(15)
16.93(2)
14.284(18)
90.00
C20H24Cl3FeN2O2
486.61
Monoclinic
P21/c
8.471(6)
24.118(16)
12.526(9)
90.00
397.33
Orthorhombic
P212121
8.377(2)
13.897(3)
16.716(3)
90.00
β/°
90.00
90.00
1946.2(7)
4
1.356
0.351
0.0461
0.0985
96.999(6)
90.00
2752.2(17)
4
1.279
.672
96.78(2)
90.00
2728(6)
4
1.390
.872
109.195(13)
90.00
2417(3)
4
1.337
1.085
χ/°
V/Å3
Z
Dc/g cm−3
μ/mm−1
R1 (on Fo2, I > 2σ(I))
wR2 (on Fo2, I > 2σ(I))
0.0379
0.0933
0.0478
0.1082
0.0634
0.1561
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 6610–6613 | 6611

View Article Online
Communication
Dalton Transactions
Scheme 2 Catechol oxidation products.
to these peaks. Significant changes in the UV-visible spectra
are observed upon addition of 3,5-di-t-butylcatecholate to
9–12. There is a pronounced difference between those com-
plexes that have a 2,4-dichlorophenolate group (10 and 11)
with a large red shift in the lower energy charge-transfer
bands.
Cyclic voltammetry was used to examine the Fe(III/II) redox
potential of acetonitrile solutions of complexes 9–12. Each
complex exhibited one reversible cathodic peak and the
Fe(III/II) redox potentials followed the trend of decreasing Lewis
acidity about the iron(III) center with 10 > 11 > 9 > 12 (−0.561 V,
−0.575 V, −0.624 V, and −0.640 V respectively vs. Fc/Fc+).
This data shows that the replacement of methyl substituents
by t-butyl substituents on the phenolate groups destabilizes
Fe(III) compared to Fe(II) by ∼15 mV. Similarly replacement
of an alkyl substituted phenolate by a chloro substituted
phenolate is worth ∼64 mV of destabilization.
Compounds 9–11 were combined with 1 equivalent of 3,5-
di-t-butylcatechol and 2 equivalents of piperidine in DMF and
exposed to molecular oxygen for 24 hours. After an acidic
workup, the products were investigated by GCMS analysis. As
opposed to 12,20 which was found to be non-reactive as an
intradiol dioxygenase, 9–11 all exhibited intradiol dioxygenase
reactivity as shown by the formation of compounds 16, 17 and
18. Compounds 17 and 18 are the products formed when
either piperidine or dimethylamine (an impurity found in
DMF) react with the acid anhydride product (16) of the reac-
tion. Compounds 16–18 typically made up 75–85% of the
product mixture and were the three most prevalent products
(Scheme 2).
Fig. 1 Thermal ellipsoid plots of the single crystal X-ray crystallographic
structures of 9–11. Ellipsoids are drawn at 50% and hydrogen atoms are
omitted for clarity.
Fe–O bond length for the 2,4-dichlorophenol is longer
(∼1.88 Å) than the 2,4-dialkylphenol bond length (1.84–
1.85 Å). This differentiation in phenol binding relates to the
enzyme, in that the proposed first step for the oxidation of
catechols is loss of one of the tyrosine phenols. All five com-
plexes 9–13 have similar trigonality indexes27 (τ values range
from 0.76–0.79) which make these structures closer to a tri-
gonal bipyramid than the enzyme active site which has a
τ value of 0.44. Other complexes using salen ligand sets have
come closer to the τ value of the native enzyme,13,14 but lack
the matching reactivity.
UV-visible spectroscopy was used to look at the electronic
spectra of the free ligands (3 and 6–8), the iron(III) complexes
(9–12), as well as the iron(III) complexes in the presence of 3,5-
di-t-butylcatechol, Table 2. For 9–12 there were two higher
intensity bands in the near UV and two bands in the visible
region of the spectra. The highest intensity band is associated
with the π–π* transition of the phenol/phenolate groups
(ranging from 281–287 nm) and there is a small red-shift when
chloro-substituents are present (compared to alkyl substitu-
ents) and a small blue shift upon binding of iron(III). The
lower energy bands are associated with ligand to metal charge-
transfer transitions from phenolate groups to the iron(III)
center. The lower energy charge-transfer bands are broad, pre-
sumably from having two different phenol ligands contributing
Conclusions
We have elucidated a new route to diamino/heterophenol com-
plexes that provide a means to differentiate the binding
Table 2 Electronic spectral data for iron(III) complexes in CH3CN
λmax (nm) (ε, M−1 cm−1
)
11
10
9
12
Free ligand
FeClL complex
287 (4730)
471 (2120)
390 (2030) br
320 (4260) sh
284 (8190)
620 (1630)
427 (2880)
296 (10 000)
285 (4370)
470 (2390) sh
400 (2540)
324 (4880)
280 (8710)
619 (1350)
422 (2240)
296 (8940)
284 (4460)
478 (2220)
390 (2030)
320 (3710) sh
281 (7400)
478 (4500)
340 sh
283 (4740)
473 (3920)
390 (3580)
321 (6560)
281 (13 600)
477 (2700)
340 sh
FeClL + 3,5-di-t-butylcatechol
282 (15 900)
283 (9820)
6612 | Dalton Trans., 2014, 43, 6610–6613
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Communication
strength of the two phenols to a metal center by varying the 11 P. C. A. Bruijnincx, M. Lutz, A. L. Spek, W. R. Hagen, G. van
electronic and steric properties of those moieties. These new
ligands provide asymmetry at the metal center that could be
Koten and R. J. M. K. Gebbink, Inorg. Chem., 2007, 46,
8391.
useful for new catalysts and have been shown to be useful in 12 M. Costas, M. P. Mehn, M. P. Jensen and L. Que, Jr., Chem.
modeling the active-sites of biological systems in this study. Rev., 2004, 104, 939.
When complexes 6–8 are bound to iron(III), all three form tri- 13 H. Fujii and Y. Funahashi, Angew. Chem., Int. Ed., 2002, 41,
gonal bipyramidal complexes that are a close match to the 3638.
structure of the active-site of 3,4-PCD and exhibit dioxygenase 14 T. Kurahashi, K. Oda, M. Sugimoto, T. Ogura and H. Fujii,
reactivity with a catechol substrate. Our lab is investigating a Inorg. Chem., 2006, 45, 7709.
wider series of ligands and the reactivity of the resulting Fe(III) 15 F. Li, M. Wang, P. Li, T. Zhang and L. Sun, Inorg. Chem.,
complexes. Those results will be reported soon.
2007, 46, 9364.
16 R. Mayilmurugan, K. Visvaganesan, E. Suresh and
M. Palaniandavar, Inorg. Chem., 2009, 48, 8771.
17 M. Velusamy, R. Mayilmurugan and M. Palaniandavar,
Inorg. Chem., 2004, 43, 6284.
Notes and references
18 C.-H. Wang, J.-W. Lu, H.-H. Wei and M. Takeda, Inorg.
Chim. Acta, 2007, 360, 2944.
1 T. Borowski and P. E. M. Siegbahn, J. Am. Chem. Soc., 2006,
128, 12941.
19 K. Hasan, C. Fowler, P. Kwong, A. K. Crane, J. L. Collins
and C. M. Kozak, J. Chem. Soc., Dalton Trans., 2008, 2991.
20 M. Velusamy, M. Palaniandavar, R. S. Gopalan and
G. U. Kulkarni, Inorg. Chem., 2003, 42, 8283.
21 T. J. Boyle, H. D. Pratt III, L. A. M. Ottley, T. M. Alam,
S. K. McIntyrem, M. A. Rodriguez, J. R. Farrell and
C. F. Campana, Inorg. Chem., 2009, 48, 9191.
22 J. R. Farrell, J. Niconchuk, C. S. Higham and
B. W. Bergeron, Tetrahedron Lett., 2007, 48, 8034.
23 C. S. Higham, D. P. Dowling, J. L. Shaw, A. Çetin,
C. J. Ziegler and J. R. Farrell, Tetrahedron Lett., 2006, 47,
4419.
24 S. Groysman, I. Goldberg, M. Kol, E. Genizi and
Z. Goldschmidt, Organometallics, 2004, 23, 1880.
25 S. Groysman, E. Y. Tshuva, I. Goldberg, M. Kol,
A. Goldschmidt and M. Shuster, Organometallics, 2004, 23,
5291.
2 M. I. Davis, A. M. Orville, F. Neese, J. M. Zaleski,
J. D. Lipscomb and E. I. Solomon, J. Am. Chem. Soc., 2002,
124, 602.
3 R. W. Frazee, A. M. Orville, K. B. Dolbeare, H. Yu,
D. H. Ohlendorf and J. D. Lipscomb, Biochemistry, 1998, 37,
2131.
4 M. Y. M. Pau, M. I. Davis, A. M. Orville, J. D. Lipscomb and
E. I. Solomon, J. Am. Chem. Soc., 2007, 129, 1944.
5 M. P. Valley, C. K. Brown, D. L. Burk, M. W. Vetting,
D. H. Ohlendorf and J. D. Lipscomb, Biochemistry, 2005, 44,
11024.
6 M. W. Vetting, D. A. D’Argenio, L. N. Ornston and
D. H. Ohlendorf, Biochemistry, 2000, 39, 7943.
7 D. H. Ohlendorf, J. D. Lipscomb and P. C. Weber, Nature,
1988, 336, 403.
8 A. M. Orville, J. D. Lipscomb and D. H. Ohlendorf, Biochem-
istry, 1997, 36, 10052.
26 E. Y. Tshuva, M. Versano, I. Goldberg, M. Kol, H. Weitman
and Z. Goldschmidt, Inorg. Chem. Commun., 1999, 2, 371.
27 A. W. Addison, T. N. Rao, J. Reedijk, J. Van Rijn and
G. C. Vershcoor, J. Chem. Soc., Dalton Trans., 1984, 1349.
9 A. E. True, A. M. Orville, L. L. Pearce, J. D. Lipscomb and
L. Que Jr., Biochemistry, 1990, 29, 10847.
10 M. M. Abu-Omar, A. Loaiza and N. Hontzeas, Chem. Rev.,
2005, 105, 2227.
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 6610–6613 | 6613