The Synthesis and Characterization of Aromatic Hybrid Anderson-Evans POMs and their Serum Albumin Interactions: The Shift from Polar to Hydrophobic Interactions

Article abstract of DOI:10.1002/chem.201502458

Four aromatic hybrid Anderson polyoxomolybdates with Fe³⁺ or Mn³⁺ as the central heteroatom have been synthesized by using a pre-functionalization protocol and characterized by using single-crystal X-ray diffraction, FTIR, ESI-MS,

Full text of DOI:10.1002/chem.201502458

DOI: 10.1002/chem.201502458
Full Paper
&
Protein Interactions
The Synthesis and Characterization of Aromatic Hybrid Anderson–
Evans POMs and their Serum Albumin Interactions: The Shift from
Polar to Hydrophobic Interactions
Emir Al-Sayed,[a] Amir Blazevic,[a] Alexander Roller,[b] and Annette Rompel*[a]
Dedicated to Professor Dr. Wolfgang Tremel on the occasion of his 60th birthday
Abstract: Four aromatic hybrid Anderson polyoxomolyb-
dates with Fe3+ or Mn3+ as the central heteroatom have
been synthesized by using a pre-functionalization protocol
and characterized by using single-crystal X-ray diffraction,
{(OCH2)3CNHCOC6H5}2] (Na-MnMo6-bzn), and Na3[MnMo6O18-
{(OCH2)3CNHCOC8H7}2] (Na-MnMo6-cin). The hydrolytic stabil-
ity of the sodium salts was examined by applying ESI-MS in
the pH range of 4 to 9. Sodium dodecylsulfate–polyacryl-
amide gel electrophoresis (SDS-PAGE) showed that human
and bovine serum albumin (HSA and BSA) remain intact in
solutions that contain up to 100 equivalents of the sodium
salts over more than 4 d at 208C. Tryptophan (Trp) fluores-
cence quenching was applied to study the interactions be-
tween the sodium salts and HSA and BSA at pH 5.5 and 7.4.
The quenching constants were extracted by using Stern–
Volmer analysis, which suggested the formation of a 1:1
POM–protein complex in all samples. It is suggested that the
aromatic hybrid POM approaches subdomain IIA of HSA and
exhibits hydrophobic interactions with its hydrophobic tails,
whereas the Anderson core is stabilized through electrostat-
ic interactions with polar amino acid side chains from, for ex-
ample, subdomain IB.
1
FTIR, ESI-MS, H NMR spectroscopy, and elemental analysis.
Structural analysis revealed the formation of (TBA)3[Fe-
Mo6O18{(OCH2)3CNHCOC6H5}2]·3.5ACN
TBA=tetrabutylammonium, ACN=acetonitrile, bzn=TRIS-
benzoic acid alkanolamide,
(TBA-FeMo6-bzn;
TRISÀR=(HOCH2)3CÀR)),
(TBA)3[FeMo6O18{(OCH2)3CNHCOC8H7}2]·2.5ACN (TBA-FeMo6-
cin; cin=TRIS-cinnamic acid alkanolamide), (TBA)3[Mn-
Mo6O18{(OCH2)3CNHCOC6H5}2]·3.5ACN (TBA-MnMo6-bzn), and
(TBA)3[MnMo6O18{(OCH2)3CNHCOC8H7}2]·2.5ACN
(TBA-
MnMo6-cin). To make these four compounds applicable in
biological systems, an ion exchange was performed that
gave the water-soluble (up to 80 mm) sodium salts Na3[Fe-
Mo6O18{(OCH2)3CNHCOC6H5}2] (Na-FeMo6-bzn), Na3[FeMo6O18-
{(OCH2)3CNHCOC8H7}2]
(Na-FeMo6-cin),
Na3[MnMo6O18-
kinds of applications (e.g., catalysis,[2–4] material science[5–7] and
medicine,[8–11] bio- and nanotechnology,[12–15] and macromolec-
ular crystallography).[16–23] The Anderson polyoxoanion is com-
posed of six edge-sharing MO6 (M=W or Mo) octahedra that
surround a central, edge-sharing heteroatom octahedron (XO6).
The general structure can be subdivided into two categories:
the nonprotonated A-type with central heteroatoms with high
oxidation states and the general formula [Xn+M6O24](12Àn)À (M=
Mo6+, W6+; X=heteroatom, e.g., Te6+, I7+) and the protonated
Introduction
Polyoxometalates (POMs) are polyanions made up of early
transition metals in their highest oxidation states (d1 and d0)
that are bridged by oxygen atoms.[1] Changing their size,
shape, or composition enables the tuning of POMs for different
[a] E. Al-Sayed, A. Blazevic, Prof. Dr. A. Rompel
Institut für Biophysikalische Chemie, Fakultät für Chemie
Universität Wien
B-type with the general formula [Xn+M6O18](6Àn)À (M=Mo6+
,
Althanstraße 14
1090 Wien (Austria)
E-mail: annette.rompel@univie.ac.at
Homepage: http://www.bpc.univie.ac.at
W6+; X=heteroatom, e.g., Ga3+, Cr3+, Fe3+) with heteroatoms
in low oxidation states.[1]
Hybrid organic–inorganic POMs have been known for a long
time, and make it possible to combine the inorganic POM with
specific organic functionalities to give new materials with inter-
esting properties.[24–26] Hybrids based on the Anderson struc-
ture can be synthesized by attaching one or two tris(hydroxy-
methyl)methane derivatives (TRISÀR; (HOCH2)3CÀR) to the
planar metal oxide framework. The first report by Hasenknopf
et al.[27] described the grafting of RÀC(CH2OH)3 (R=CH3, NO2,
CH2OH) onto polyoxomolybdates with Ni2+, Zn2+, Fe3+, and
Mn3+ as central heteroatoms. Since then, numerous reports on
[b] A. Roller
Institut für Anorganische Chemie, Fakultät für Chemie
Universität Wien
Währinger Straße 42
1090 Wien (Austria)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201502458.
ꢀ 2015 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA.
This is an open access article under the terms of the Creative Commons At-
tribution License, which permits use, distribution and reproduction in any
medium, provided the original work is properly cited.
Chem. Eur. J. 2015, 21, 17800 – 17807
17800 ꢀ 2015 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
the further derivatization of the functional groups on the tripo-
dal alcohols through imine and peptide bond formation have
been reported, to give POMs with various properties.[28–32] The
classical method for obtaining a symmetric Anderson hybrid
POM is a one-pot reaction in a polar organic solvent in the
presence of octamolybdate, a salt of the central heteroatom
and the organic ligand.[27] The TRIS ligands were pre-functional-
ized before being introduced into the reaction mixture, that is,
the organic ligand is formed first then incorporated during the
constitution of the hybrid cluster. Further success of functional-
izing the organic ligands in Anderson polyoxometalates by
pre- or post-functionalization has recently been reviewed.[33]
The versatile applications of POMs in macromolecular X-ray
crystallography has been reviewed.[19] Several aspects make
POMs ideal candidates for use as crystallization additives,[20–23]
especially the Anderson POM.[19] It can be used as a phasing
tool and only a few binding sites are necessary, whereas
mononuclear heavy atoms must bind to multiple sites to pro-
vide useful phases, especially for large proteins.[34–36] Further-
more, most POMs have high negative charges that make it
possible to crosslink positive regions of several monomers
through electrostatic interactions. This leads to the formation
of new contacts and increases the chance of a long-range-
order formation.[18] In one instance, protein crystals have only
been obtained in the presence of POMs, in the case of mush-
room tyrosinase (abPPO4, ab=Agaricus bisporus, PPO=poly-
phenol oxidase) for which protein crystals were only obtained
in the presence of the [TeW6O24]6À anion.[20,21] The interaction
between POMs and proteins seems to be predominantly elec-
trostatic, but hydrogen bonds, covalent bonds, p–p, and van
der Waals interactions have also been observed. Encouraged
by the successful use of the Anderson POM in protein crystal-
lography, we here functionalized the archetype with aromatic
ligands. Tryptophan fluorescence quenching was applied to in-
vestigate possible interactions between the aromatic ring
grafted onto the Anderson POMs and the aromatic amino
acids on the protein surface, with the aim of enabling different
POM–protein interactions and thus using the aromatically
TRIS-functionalized hybrid Anderson–Evans POMs reported
herein as future additives in macromolecular crystallography.
Scheme 1. a)–c) Schematic illustration of the preparation steps of the organ-
ic ligands and d) the grafting onto the Anderson POM to give hybrid POMs
with aromatic ligands. Legend: MoO6: grey octahedra; O: grey spheres. Hy-
drogen atoms are omitted for clarity.
Fe3+ and Mn3+ were chosen as templating heteroatoms be-
cause both provide POM hybrids in good yields (86% based
on Mo).[38] Compounds TBA-FeMo6-bzn, TBA-FeMo6-cin, TBA-
MnMo6-bzn, and TBA-MnMo6-cin were obtained by heating
Fe(acac)3 or Mn(OAc)3, (TBA)4[a-Mo8O26], and bzn or cin at
reflux in acetonitrile for 18 h (Scheme 1d). All four compounds
were isolated as TBA salts with poor water solubility (<3 mm).
The solubility was increased by cation exchange with Na+ by
mixing the TBA salts with NaClO4 for 30 min in ACN to form
a precipitate of the Na salt.[28] This led to the isolation of Na3-
[FeMo6O18{(OCH2)3CNHCOC6H5}2]
(Na-FeMo6-bzn),
Na3[Fe-
Mo6O18{(OCH2)3CNHCOC8H7}2] (Na-FeMo6-cin), Na3[MnMo6O18-
{(OCH2)3CNHCOC6H5}2] (Na-MnMo6-bzn), and Na3[MnMo6O18-
{(OCH2)3CNHCOC8H7}2] (Na-MnMo6-cin; Figures S2 and S5 in
the Supporting Information). The final water solubility was
80 mm for Na-FeMo6-bzn and Na-MnMo6-bzn and 40 mm for
Na-FeMo6-cin and Na-MnMo6-cin with the larger hydrophobic
cinnamic acid ligand.
X-ray structural characterization
X-ray crystallographic analysis shows that the asymmetric units
in TBA-FeMo6-bzn, TBA-FeMo6-cin, TBA-MnMo6-bzn, and
TBA-MnMo6-cin consist of the hybrid Anderson POM
(Figure 1), three TBA counterions, and ACN solvent molecules
(2.5 to 3.5). A summary of crystal parameters and refinement
details are shown in Table 1. The structural analysis revealed
that all four compounds crystallize in the monoclinic crystal
system, space group C2/c for TBA-FeMo6-bzn and TBA-
MnMo6-bzn whereas TBA-FeMo6-cin and TBA-MnMo6-cin crys-
tallize in space group P21/n.
Results and Discussion
Synthesis
The pre-functionalization of the organic ligands was achieved
by using an established procedure.[37] The organic acid (benzo-
ic acid or trans-cinnamic acid) a (Scheme 1) was added to ethyl
chloroformate in the presence of N-methylmorpholin (NMM) in
THF to generate mixed anhydride b, which subsequently react-
ed with TRIS-NH2 in dimethylformamide (DMF) and triethyl-
amine (TEA) to form alkanolamides (bzn and cin) c (the
1H NMR spectroscopic characterization of bzn and cin is given
in Figures S3 and S4 in the Supporting Information). The bzn
and cin ligands differ in their carbon chain length, which leads
to more flexibility in the POMs that contain cin ligands upon
interaction with proteins.
All four compounds show the characteristic Anderson-type
structure with a central XO6 (X=Mn3+, Fe3+) octahedron sur-
rounded by six edge-shared MoO6 octahedra that form
a planar array of distorted octahedra that originates from the
outwards expansion of the Mo atoms, all in agreement with re-
ported Anderson structures.[39] The central octahedron is also
slightly flattened as indicated by the summarized bond lengths
in Table 2. Three different coordination modes of oxygen
atoms are found in the structure; six triple-bridged oxygen
atoms connect the heteroatom and two Mo atoms, six double-
Chem. Eur. J. 2015, 21, 17800 – 17807
www.chemeurj.org
17801 ꢀ 2015 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Compounds TBA-FeMo6-cin and TBA-MnMo6-cin, which
contain the larger cinnamic acid, showed distorted aromatic
units and after appropriate consideration (see the Experimental
Section in the Supporting Information for details), possible p–
p interactions in the crystal structure based on geometry and
separation (Figure 2) were found. TBA-FeMo6-cin displays two
different modes of p–p interaction between conjugated sys-
tems (Figure 2A), either between two aromatic rings or be-
tween the aromatic ring and the p* orbital in C=C. The crystal-
lographic refinement results for TBA-MnMo6-cin suggest only
p–p interactions between the aromatic ring and C=C (Fig-
ure 2B) based on geometry and separation. The separations
vary between 3.41 and 3.88 Š for all p–p interactions between
the aromatic ring and C=C in both compounds, which com-
pares well with previous reports.[40] The interactions arrange in
a parallel offset face-to-face-type p–p stacking. The exact type
of stacking between the two aromatic rings is difficult to iden-
tify due to the distortion, but is predominantly of a T-shaped
character with a separation of
Figure 1. Combined skeletal/polyhedral representation of A) [XMo6O18-
{(OCH2)3CNHCOC6H5}2]3À (X=Fe3+ (TBA-FeMo6-bzn), Mn3+ (TBA-MnMo6-
bzn)) and B) [XMo6O18{(OCH2)3CNHCOC8H7}2]3À (X=Fe3+ (TBA-FeMo6-cin),
Mn3+ (TBA-MnMo6-cin)). Legend: MoO6: grey octahedra; O: grey spheres.
Hydrogen atoms are omitted for clarity.
3.47 Š.
Table 1. Crystallographic data for TBA-FeMo6-bzn, TBA-FeMo6-cin, TBA-MnMo6-bzn, and TBA-MnMo6-cin.
Interestingly, TBA-FeMo6-bzn
and TBA-MnMo6-bzn, which
contain benzoic acid, do not dis-
play any p–p interaction be-
tween the aromatic units, possi-
bly due to the shorter carbon
chain compared with com-
pounds that contain the cin
ligand. Thus, p–p interactions
would require two POM units at
a close separation, which is elec-
trostatically unfavorable.
TBA-FeMo6-bzn
TBA-FeMo6-cin
TBA-MnMo6-bzn
TBA-MnMo6-cin
formula
C77H142.5FeMo6N8.5O26 C79H143.50FeMo6N7.50O26 C77H142.5MnMo6N8.5O26 C79H143.5MnMo6N7.5O26
Mr [gcmÀ3
]
2234.98
C2/c
2245.99
P21/n
2234.07
C2/c
2243.07
P21/n
space group
crystal system
a [Š]
b [Š]
c [Š]
monoclinic
27.0330(16)
26.8760(16)
27.6294(18)
90
monoclinic
14.8218(9)
27.4955(18)
23.9333(14)
90
monoclinic
27.0398(19)
26.9078(19)
27.6186(19)
90
monoclinic
14.8671(10)
27.4468(19)
23.8225(16)
90
a [8]
b [8]
105.084(3)
90
19382(2)
8
92.3575(14)
90
9745.3(10)
4
104.729(3)
90
19434(2)
8
92.404(2)
90
9712.3(11)
4
g [8]
V [Š3]
Z
Compounds TBA-FeMo6-bzn
and TBA-MnMo6-bzn show simi-
lar crystal packing due to the
common organic ligand, which
consists of alternate layers of
TBA counterions and ACN sol-
vent molecules and alternate
layers of the hybrid POM. The al-
ternate layers are repeated
along the b plane. Compounds
m [mmÀ1
]
0.969
0.964
0.947
0.948
reflns. collected 221627
126892
28975
0.0699
1.009
0.0484
0.1185
378613
17792
0.0718
1.085
0.0317
0.0850
268583
28596
0.0687
1.031
0.0458
0.1114
indep. reflns.
Rint
GOF on F2
R1 [I>2s(I)]
wR2 (all data)
17745
0.0714
1.123
0.0517
0.1337
Table 2. Selected bond lengths [Š] for the anions in TBA-FeMo6-bzn, TBA-FeMo6-cin, TBA-MnMo6-bzn, and
TBA-MnMo6-cin; Ot =terminal oxygen atoms.
TBA-FeMo6-cin
and
TBA-
MnMo6-cin also show similar
crystal packing and consist of al-
ternate layers. The first layer is
comprised of the inorganic POM
and TBA cations and the second
layer are composed of the or-
ganic ligand grafted onto the
TBA-FeMo6-bzn
TBA-FeMo6-cin
TBA-MnMo6-bzn
TBA-MnMo6-cin
Mn/FeÀO
Mn/FeÀOÀMo
MoÀOÀMo
MoÀOt
1.980(12)–1.990(11)
2.335(14)–2.410(12)
1.910(9)–1.928(13)
1.699(8)–1.706(12)
1.981(11)–1.999(4)
2.350(8)–2.396(12)
1.914(9)–1.927(11)
1.696(11)–1.709(13)
1.957(4)–2.016(9)
2.331(13)–2.378(9)
1.911(12)–1.928(12)
1.697(15)–1.707(13)
1.911(11)–2.054(12)
2.314(12)–2.405(9)
1.909(15)–1.929(11)
1.700(9)–1.710(10)
bridged oxygen atoms connect two Mo atoms, and two termi-
nal oxygen atoms are connected to each of the six Mo atoms.
The bond lengths of the three different binding modes are
summarized in Table 2 and are in good agreement with other
TRIS-functionalized Anderson POMs.[28,38] The organic ligands
are grafted directly onto the oxygen atoms that surround the
heteroatom (Figure 1).
POM and TBA cations. This build up is found along both the
a and c planes.
Chem. Eur. J. 2015, 21, 17800 – 17807
www.chemeurj.org
17802 ꢀ 2015 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Samples of Na-FeMo6-cin and Na-MnMo6-bzn were dis-
solved in water that contained 10 mm buffers at pH 4 (ammo-
nium acetate), 7 (ammonium bicarbonate), and 9 (ammonium
carbonate) and ESI-MS spectra were recorded after 24 h.
Figure 3 shows peak envelopes of the intact clusters
Na2[FeMo6O18{(OCH3CNHCOC8H7}2]À (found: 1461.4; calcd:
1461.4) and Na2[MnMo6O18{(OCH2)3CNHCOC6H5}2]À (found:
1408.4; calcd: 1408.4) at the three different pH values. This
qualitatively confirms the presence of the intact clusters, with
a satisfactory overlap of the superimposed simulated pattern.
The full spectra are highly similar in terms of fragmentation
pattern and the relative intensity of the main peaks in all sam-
ples in the pH study (Figure 3). The spectra are also similar to
TBA-FeMo6-cin and TBA-MnMo6-bzn recorded in the absence
of any buffer (Figure S6 in the Supporting Information). This in-
dicates a comparable POM concentration in all samples and
suggests solution stability over a pH range from 4 to 9.
Figure 2. p–p interactions found in A) TBA-FeMo6-cin and B) TBA-MnMo6-
cin and their observed separations in the crystal structures. Legend: MoO6:
red octahedra; FeO6: orange octahedra; MnO6: blue octahedra; C: black
spheres; N: blue spheres; O: red spheres. Hydrogen atoms are omitted for
clarity.
SDS-PAGE study
SDS-PAGE was applied to show that Na-FeMo6-bzn, Na-FeMo6-
cin, Na-MnMo6-bzn, and Na-MnMo6-cin are hydrolytically inac-
tive towards proteins. Hydrolytic activity has been reported for
POM archetypes in which the heteroatoms are more ex-
posed.[15] Human and bovine serum albumin (HSA and BSA)
proteins were chosen because their positive surface charge
has been extensively investigated under the typical physical
conditions applied in protein crystallization. The aqueous
buffer systems used are the same as in the hydrolytic stability
study performed by using ESI-MS. Compounds Na-FeMo6-bzn,
Na-FeMo6-cin, Na-MnMo6-bzn, and Na-MnMo6-cin were
added in 10- and 100-fold excess of HSA/BSA and were ana-
lyzed after 4 d at 208C. Controls with starting reagents and
nondecorated Anderson POMs are included in the experiment.
The SDS-PAGE (14 % polyacrylamide gel) was stopped just
before the loading buffer finished traveling across the gel to
insure detection of small protein fractions.
FTIR spectroscopy
The IR transmission spectra of TBA-FeMo6-bzn, TBA-FeMo6-
cin, TBA-MnMo6-bzn, and TBA-MnMo6-cin are presented and
discussed in the Supporting Information (Figure S1).
ESI-MS characterization and hydrolytic stability study
Electrospray ionization mass spectrometry (ESI-MS) was used
to characterize TBA-FeMo6-bzn, TBA-FeMo6-cin, TBA-MnMo6-
bzn, and TBA-MnMo6-cin. The most relevant peak envelopes
of (TBA)[FeMo6O18(C22H24N2O8)]2À (calcd: 803.4; found: 803.4),
(TBA)[FeMo6O18(C26H28N2O8)]2À (calcd: 829.4; found: 829.4),
(TBA)[MnMo6O18(C22H24N2O8)]2À (calcd: 802.9; found: 802.9), and
(TBA)[MnMo6O18(C26H28N2O8)]2À (calcd: 828.9; found: 828.9),
which confirmed the presence of the intact clusters in the
compounds, are shown in Figure S6 in the Supporting Informa-
tion. The rest of the spectra display a quite complex fragmen-
tation pattern but are similar for all four compounds. They
form mainly oxo-molybdo fragments with Mo in different oxi-
dation states, in accordance with previous reports.[41]
Both gels show (Figure 4) the intact serum albumin protein
at 66 (BSA) or 66.5 kDa (HSA) with no lower mass fractions de-
tectable, even at a 100-fold excess of Na-FeMo6-bzn, Na-
FeMo6-cin, Na-MnMo6-bzn, and Na-MnMo6-cin, in agreement
with previously reported TRIS-functionalized Anderson
POMs.[24] The only sample that showed lower mass fractions
was that with FeCl3, which indicates nonspecific cleavage with
both proteins due to the Lewis acid properties of Fe3+ [45]
.
The hydrolytic stability of compounds Na-FeMo6-cin and
Na-MnMo6-bzn were subject to investigation with ESI-MS after
24 h in aqueous buffer solutions in the pH range of 4 to 9. Re-
ports on hydrolytic stability are rather scare due to the chal-
lenge of maintaining the correct isomer in solution or prevent-
ing conversion into different structures.[42] POMs that have
been organically modified and feature both covalent and non-
covalent attachment of organic ligands and biomolecules have
all shown increased hydrolytic stability under physiological
conditions.[43,44] Recently, the single-side grafted [GaM-
o6O18(OH)3{(OCH2)3CCH2OH}]3À anion was also confirmed to be
stable in the pH range of 4 to 9 for up to 24 h.[24]
Fluorescence quenching measurements to investigate the
POM–HSA interaction
Fluorescence quenching is a well-established experiment for
the investigation of ligand–protein interactions. Therefore, it
has been previously used to analyze the interactions between
HSA/BSA and different POMs, including Keggin, Wells–Dawson,
Lindqvist, and wheel-shape-structured POMs.[46–50] However, all
of these investigations were performed with solely inorganic
POMs. Herein, fluorescence quenching was used to gain more
insights into the interaction between aromatic hybrid POMs
Chem. Eur. J. 2015, 21, 17800 – 17807
www.chemeurj.org
17803 ꢀ 2015 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 3. Peak envelopes of Na-FeMo6-cin (top) and Na-MnMo6-bzn (bottom) at pH 4 (left), 7 (center), and 9 (right) after 24 h; experimental pattern is in
black and simulated pattern is overlaid in red.
and the fluorophore tryptophan of HSA. HSA contains one
tryptophan residue at position 214, whereas BSA has two that
are located at positions 134 and 213 in the amino acid se-
quence.[51] HSA and BSA protein were investigated at pH 5.5
and 7.4 in solutions with different concentrations of Na-
FeMo6-bzn, Na-FeMo6-cin, Na-MnMo6-bzn, and Na-MnMo6-
cin. The concentration of the proteins was kept constant
(1 mgmLÀ1), whereas the POM concentrations were increased
up to 0.4-fold (0.006, 0.012, 0.025, 0.05, 0.1, 0.2, and 0.4-fold).
Table 3 shows the calculated quenching constants for HSA
(see Table S1 in the Supporting Information for the BSA data)
and the number of bound molecules. Figure 5A,B shows the
emission spectra and corresponding derived Stern–Volmer plot
of Na-FeMo6-cin (the remaining emission and Stern–Volmer
plots for Na-FeMo6-cin and BSA, Na-FeMo6-bzn, Na-MnMo6-
bzn, and Na-MnMo6-cin with BSA and HAS are given in Fig-
ures S7–S10 in the Supporting Information).
The emission spectra show a maximum at l=312 nm, which
decreases with increasing POM concentrations due to binding
of POM to HSA. In addition, there is a noticeable shift towards
lower wavelengths (with increasing POM concentration), which
suggests a decrease in polarity within the immediate environ-
ment of the tryptophan. The quenching constants are higher
at lower pH (5.5) due to the higher overall surface charge of
the protein, which is in accordance with previous results.[50] It
is also worth noting that all POM hybrids form a 1:1 complex
with the protein, which is also in agreement with previous re-
ports that involve different inorganic POM archetypes.[46–50]
This indicates not only successful POM binding but that the in-
teraction can only take place at a conformational strongly de-
fined site, which was suggested to be subdomain IIA. It is well
established that compounds that bind to subdomain IIA are
likely to enhance fluorescence quenching of HSA because
compounds bound at other cavities (e.g., subdomain IIIA)
would not exhibit any fluorescence quenching due to their
greater distance from tryptophan 214. Therefore, it is suggest-
ed that the aromatic hybrid POMs reported herein interact
with subdomain IIA.
Table 3. Quenching constants and number of binding molecules for the
investigated albumin proteins and pH values.
POM
Protein
Kq [mÀ1
]
n
pH
Na-FeMo6-bzn
Na-FeMo6-cin
Na-MnMo6-bzn
Na-MnMo6-cin
Na-FeMo6-bzn
Na-FeMo6-cin
Na-MnMo6-bzn
Na-MnMo6-cin
Na-MnMo6-cin
HSA
HSA
HSA
HSA
HSA
HSA
HSA
HSA
1.3”105
1.4”105
4.4”105
1.2”105
5.0”104
9.3”104
6.0”104
9.8”104
2.5”106
1.3
1.2
1.3
1.3
1.1
1.2
1.1
1.2
0.9
5.5
5.5
5.5
5.5
7.4
7.4
7.4
7.4
5.5
To confirm this binding site, further fluorescence quenching
experiments were performed with a HSA–indometacin (HSA-
IMN) complex. IMN is known to bind to subdomain IIA of HSA,
which was proven by X-ray crystallography.[52] Moreover, it has
HSA+IMN
Chem. Eur. J. 2015, 21, 17800 – 17807
www.chemeurj.org
17804 ꢀ 2015 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 4. SDS-PAGE screening of BSA/HSA solutions that contain Na-FeMo6-
bzn, Na-FeMo6-cin, Na-MnMo6-bzn, and Na-MnMo6-cin in 10 and 100-fold
excess. Reagents used to synthesize the POMs are also included in 100-fold
excess. MnMo6O24 =Na3[Mn(OH)6Mo6O18]; FeMo6O24 =Na3[Fe(OH)6Mo6O18];
MnMo6O24(TRIS)2 =TBA3[FeMo6O18{(OCH2)3CNH2}2];[27] M=Marker. Top: SDS-
PAGE samples with BSA; bottom: SDS-PAGE samples with HSA.
been shown that IMN is able to bind to this site in the pres-
ence of other compounds, such as cinnamic acid. Therefore,
several emission fluorescence spectra of HSA-IMN-Na-MnMo6-
cin were recorded (Figure 5C). The calculated binding constant
of HSA-IMN-Na-MnMo6-cin was significantly larger than that of
HSA-Na-MnMo6-cin without the drug (Table 3), which indicates
that both the drug and the hybrid POM were simultaneously
bound to subdomain IIA of HSA.
Figure 5. Emission fluorescence spectra of Na-FeMo6-cin with HSA
([HSA]=10À5 mÀ1) in 10 mm NaOAc buffer at pH 5.5 (A) and 7.4 (B). The top
line in each spectrum was recorded in the absence of Na-FeMo6-cin fol-
lowed by stepwise increase (0.006, 0.012, 0.025, 0.05, 0.1, 0.2, and 0.4-fold)
of Na-FeMo6-cin. The emission fluorescence spectrum in C shows the mea-
surement with indometacin present (0.4-fold) and stepwise Na-FeMo6-cin in-
crease (0, 0.1, 0.2, 0.4, 0.8, and 1-fold). Insets: The plot of the derived Stern–
Volmer equation (with R2 =0.99).
Taking into account the size of the hybrid POM Na-MnMo6-
cin (23.8”8.8”2.4 Š) and the sizes of all HSA cavities, there is
only one cavity in which the aromatic hybrid POMs are sterical-
ly able to penetrate the core of HSA, namely the cavity that
leads to subdomain IIA (Figure 6).
Na3[Fe(OH)6Mo6O18] up to two molar equivalents showed no
quenching of the tryptophan signal at either pH value (5.5 and
7.4), which revealed no binding in the vicinity of tryptophan
214 and thus indicated the importance of the aromatic moiet-
ies in the binding of the hybrid POMs reported herein. It is
With regard to the types of interaction between HSA and
the aromatic hybrid POMs, electrostatic and especially hydro-
phobic interactions are suggested. A control experiment with
Chem. Eur. J. 2015, 21, 17800 – 17807
www.chemeurj.org
17805 ꢀ 2015 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
been accessible so far for POMs, which could result in the for-
mation of new protein crystals promoted by POMs.
Experimental Section
Full experimental data and synthesis procedures can be found in
the Supporting Information. All reagents and chemicals were of an-
alytical grade and used without further purification. All reagents
and chemicals were supplied by Sigma–Aldrich Chemical Company
and solvents were supplied by Merck Chemicals. Single-crystal X-
ray diffraction data were collected at 100 K by using a Bruker D8
Venture diffractometer equipped with a multilayer monochroma-
tor, a MoKa INCOATEC microfocus sealed tube (l=0.71073 Š), and
a
CMOS Photon Detector. CCDC 1406885 (TBA-FeMo6-bzn,),
1406886 (TBA-MnMo6-bzn), 1406887 (TBA-FeMo6-cin), and
1406888 (TBA-MnMo6-cin) contain the supplementary crystallo-
graphic data for this paper. These data are provided free of charge
by The Cambridge Crystallographic Data Centre.
Figure 6. Hypothetical binding of aromatic hybrid POM to HSA. HSA is
shown as a ribbon structure with all subdomains represented in different
colors. Right: The aromatic hybrid POM is depicted with its dimensions. The
red arrow indicates the big cavity that leads to subdomain IIA (in red).
Inset: A closer view of the hypothetical position of the POM together with
IMN, which is located next to fluorophore Trp214. The POM is illustrated as
a combination of polyhedra and sticks, whereas IMN and Trp214 are illustrat-
ed sticks (color code: IMN: carbon=cyan, blue=nitrogen, red=oxygen,
light green=chloride; Trp214: red=carbon, blue=nitrogen).
Synthesis of TBA-FeMo6-bzn,
(TBA)3[FeMo6O18{(OCH2)3CNHCOC6H5}2]·3.5ACN
The synthesis was carried out according to a published proce-
dure.[36] Tetrabutylammonium octamolybdate was dissolved in ace-
tonitrile and heated at reflux with Fe(acac)3 and the ligand
(HOCH2)3CNHCOC6H5 for 18 h. After cooling to RT, the red mixture
was centrifuged to remove the precipitate and give a dark red so-
lution. Crystals suitable for X-ray crystallographic analysis were ob-
tained through ether diffusion after a few days. FTIR: n˜ =2960 (v
CH3, s), 2934 (v CH3, s), 2873 (v CH3, s), 1674 (v C=O, s), 1599 (v Ar,
w) 1578 (v Ar, w), 1517 (v Ar, m), 1482 (d CH2, s), 1380 (d CH3, m),
1319 (m), 1268 (m), 1102 (m), 1031 (v CÀO, m), 939 (s), 918 (s), 902
(v Mo=O, s), 808 (w), 647 (v Mo-O-Mo, s), 559 (m) 406 cmÀ1 (m). El-
emental analysis calcd (%) for FeMo6O26C70H132N5 (2091.3 gmolÀ1):
C 40.20, H 6.31, O 19.41, N 3.26, Fe 2.67, Mo 28.01; found: C 40.19,
H 6.28, O 19.38, N 3.24, Fe 2.61, Mo 27.94.
suggested that the aromatic hybrid POM approaches subdo-
main IIA through the above-mentioned cavity and then exhibit
hydrophobic interactions with its hydrophobic tails, whereas
the Anderson core is stabilized through electrostatic interac-
tions with polar amino acid side chains from, for example, sub-
domain IB (Figure 6). This is not surprising because the organic
moiety of the hybrid POMs are structurally very similar to cin-
namic acid, which has been shown to bind to HSA sub-
domain IIA (together with IMN) by directly interacting with
Trp214.[52] Thus, the aromatic hybrid POMs reported herein
might interact similarly with HSA through their organic groups.
Synthetic procedures for TBA-FeMo6-bzn, TBA-FeMo6-cin, TBA-
MnMo6-bzn, TBA-MnMo6-cin, Na-FeMo6-bzn, Na-FeMo6-cin, Na-
MnMo6-bzn, Na-MnMo6-cin, bzn, and cin, and the full experimen-
tal information are given in the Supporting Information.
Conclusion
Acknowledgements
Four different hybrid organic–inorganic Anderson POMs were
synthesized in an organic solvent and introduced to aqueous
environments through a cation exchange step. They show
robust hydrolytic stability in a pH range of 4 to 9 for up to
24 h and are hydrolytically inactive in aqueous buffer solutions
in the presence of BSA and HSA proteins. Instead, they interact
through electrostatic, hydrophobic, or p–p interactions, or
a combination of these. This introduces the possibility for an-
other mode of POM–protein interaction in addition to the
demonstrated electrostatic interaction. The terminal oxygen
atoms on the Anderson POM can interact electrostatically with
positively charged amino acids. The double-sided grafting of
the aromatic ligands may allow for p–p interactions or hydro-
phobic interactions at two different sites. This makes the
hybrid Anderson POMs reported herein potentially superior to
pure inorganic structures that have been successfully applied
so far as additives in macromolecular crystallography. Thus,
they may in theory stabilize new protein regions that have not
The research was funded by the Austrian Science Fund (FWF):
P27534. The EU COST action PoCheMoN (CM1203) is gratefully
acknowledged. We are grateful to Ing. Peter Unteregger for his
support with the ESI-MS measurements at the Mass Spectrom-
etry Centre, University of Vienna. We also thank AoUniv.-Prof.
Eugen Libowitzky for access to ATR-IR measurements and
AoUniv.-Prof. Dr. Markus Galanski for support with NMR spec-
troscopy measurements. We acknowledge AssUniv.-Prof.
Mag. Dr. Wilfried Kçrner for help with ICP-OES measurements,
Department of Environmental Geosciences, University of
Vienna. Last, the authors wish to thank Dipl.-Ing. Matthias Pret-
zler, Aleksandar Bijelic, M.Sc., and Ioannis Kampatsikas, M.Sc.,
for valuable discussions concerning this work.
Keywords: characterization
·
hydrophobic
effect
·
polyoxometalates · synthesis · X-ray crystallography
Chem. Eur. J. 2015, 21, 17800 – 17807
www.chemeurj.org
17806 ꢀ 2015 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[1] M. T. Pope, Springer, Heteropoly and Isopoly Oxometalates, Springer, Hei-
delberg, 1983.
[28] P. R. Marcoux, B. Hasenknopf, J. Vaissermann, P. Gouzerh, Eur. J. Inorg.
Chem. 2003, 2406–2412.
[2] I. V. Kozhevnikov, Catalysis by Polyoxometalates, Wiley, Chichester, 2002.
[3] A. C. Estrada, I. C. M. S. Santos, M. M. Q. Simðes, M. G. P. M. S. Neves,
J. A. S. Cavaleiro, A. M. V. Cavaleiro, Appl. Catal. A 2011, 392, 28–35.
[4] N. Mizuno, K. Kamata, K. Yamaguchi, Top. Catal. 2010, 53, 876–893.
[5] E. C. J. J. Borra’s-Almenar, A. Müller, T. Pope, Polyoxometalate Molecular
Science, Springer, Dordrecht, 2003.
[6] E. Coronado, C. J. Gómez-García, Chem. Rev. 1998, 98, 273–296.
[7] U. Kortz, A. Müller, J. van Slageren, J. Schnack, N. S. Dalal, M. Dressel,
Coord. Chem. Rev. 2009, 253, 2315–2327.
[8] J. T. Rhule, C. L. Hill, D. A. Judd, R. F. Schinazi, Chem. Rev. 1998, 98, 327–
357.
[9] H. Stephan, M. Kubeil, F. Emmerling, C. E. Müller, Eur. J. Inorg. Chem.
2013, 1585–1594.
[29] S. Favette, B. Hasenknopf, J. Vaissermann, P. Gouzerh, C. Roux, Chem.
Commun. 2003, 2664–2665.
[30] Y.-F. Song, D.-L. Long, L. Cronin, Angew. Chem. Int. Ed. 2007, 46, 3900–
3904; Angew. Chem. 2007, 119, 3974–3978.
[31] M. H. Rosnes, C. Musumeci, C. P. Pradeep, J. S. Mathieson, D.-L. Long, Y.-
F. Song, B. Pignataro, R. Cogdell, L. Cronin, J. Am. Chem. Soc. 2010, 132,
15490–15492.
[32] C. Yvon, A. J. Surman, M. Hutin, J. Alex, B. O. Smith, D.-L. Long, L.
Cronin, Angew. Chem. Int. Ed. 2014, 53, 3336–3341; Angew. Chem. 2014,
126, 3404–3409.
[33] A. Blazevic, A. Rompel, Coord. Chem. Rev. 2016, 307, 42–64.
[34] G. Rudenko, L. Henry, C. Vonrhein, G. Bricogne, J. Deisenhofer, Acta Crys-
tallogr. Sect. D 2003, 59, 1978–1986.
[10] T. Yamase, J. Mater. Chem. 2005, 15, 4773–4782.
[11] H. U. V. Gerth, A. Rompel, B. Krebs, J. Boos, C. Lanvers-Kaminsky, Anti-
Cancer Drugs 2005, 16, 101–106.
[12] G. Absillis, R. Van Deun, T. N. Parac-Vogt, Inorg. Chem. 2011, 50, 11552–
11560.
[13] M. Gluehmann, R. Zarivach, A. Bashan, J. Harms, F. Schluenzen, H. Bar-
tels, I. Agmon, G. Rosenblum, M. Pioletti, T. Auerbach, H. Avila, H. A. S.
Hansen, F. Franceschi, A. Yonath, Methods 2001, 25, 292–302.
[14] S. G. Sarafianos, U. Kortz, M. T. Pope, M. J. Modak, Biochem. J. 1996, 319,
619–626.
[35] W. A. Hendrickson, C. M. Ogata in Methods Enzymol., Vol. 276 (Ed.: W.
C. W. Carter, Jr.), Academic Press, London, 1997, pp. 494–523.
[36] G. Zhang, B. Keita, C. T. Craescu, S. Miron, P. de Oliveira, L. Nadjo, J. Phys.
Chem. B 2007, 111, 11253–11259.
[37] H. S. Rho, H. S. Baek, D. H. Kim, I. S. Chang, Notes 2006, 27, 585.
[38] M. H. Rosnes, C. Yvon, D.-L. Long, L. Cronin, Dalton Trans. 2012, 41,
10071–10079.
[39] H. T. Evans, Jr., Acta Crystallogr. Sect. B 1974, 30, 2095–2100.
[40] Q. Wu, W.-L. Chen, D. Liu, C. Liang, Y.-G. Li, S.-W. Lin, E. Wang, Dalton
Trans. 2011, 40, 56–61.
[15] K. Stroobants, E. Moelants, H. G. T. Ly, P. Proost, K. Bartik, T. N. Parac-
Vogt, Chem. Eur. J. 2013, 19, 2848–2858.
[41] E. F. Wilson, H. N. Miras, M. H. Rosnes, L. Cronin, Angew. Chem. Int. Ed.
2011, 50, 3720–3724; Angew. Chem. 2011, 123, 3804–3808.
[42] C. L. Hill, C. M. Prosser-McCartha, Coord. Chem. Rev. 1995, 143, 407–455.
[43] G. Geisberger, S. Paulus, M. Carraro, M. Bonchio, G. R. Patzke, Chem. Eur.
J. 2011, 17, 4619–4625.
[16] D. Janell, A. Tocilj, I. Kçlln, F. Schlünzen, M. Glühmann, H.A.S. Hansen, J.
Harms, A. Bashan, I. Agmon, H. Bartels, M. Kessler, S. Weinstein, F. Fran-
ceshi, A. Yonath in Polyoxometalate Chemistry From Topology via Self-As-
sembly to Applications (Eds.: M. Pope, A. Müller), Kluwer, Dordrecht,
2001, pp. 391–415.
[44] A. Flütsch, T. Schroeder, M. G. Grütter, G. R. Patzke, Bioorg. Med. Chem.
Lett. 2011, 21, 1162–1166.
[17] A. Bashan, A. Yonath, J. Mol. Struct. 2008, 890, 289–294.
[18] A. Bijelic, C. Molitor, S. G. Mauracher, R. Al-Oweini, U. Kortz, A. Rompel,
ChemBioChem 2015, 16, 233–241.
[19] A. Bijelic, A. Rompel, Coord. Chem. Rev. 2015, 299, 22–38.
[20] S. G. Mauracher, C. Molitor, R. Al-Oweini, U. Kortz, A. Rompel, Acta Crys-
tallogr. Sect. F 2014, 70, 263–266.
[45] B. Plietker, M. Beller, Iron Catalysis: Fundamentals and Applications,
Springer, Berlin, Heidelberg, 2011.
[46] V. Goovaerts, K. Stroobants, G. Absillis, T. N. Parac-Vogt, Phys. Chem.
Chem. Phys. 2013, 15, 18378–18387.
[47] G. Hungerford, F. Hussain, G. R. Patzke, M. Green, Phys. Chem. Chem.
Phys. 2010, 12, 7266–7275.
[21] S. G. Mauracher, C. Molitor, R. Al-Oweini, U. Kortz, A. Rompel, Acta Crys-
tallogr. Sect. D 2014, 70, 2301–2315.
[48] G. Zhang, B. Keita, J.-C. Brochon, P. de Oliveira, L. Nadjo, C. T. Craescu, S.
Miron, J. Phys. Chem. B 2007, 111, 1809–1814.
[22] F. Schluenzen, A. Tocilj, R. Zarivach, J. Harms, M. Gluehmann, D. Janell,
A. Bashan, H. Bartels, I. Agmon, F. Franceschi, A. Yonath, Cell 2000, 102,
615–623.
[23] C. Molitor, S. G. Mauracher, A. Rompel, Acta Crystallogr. Sect. F 2015, 71,
746–751.
[24] A. Blazevic, E. Al-Sayed, A. Roller, G. Giester, A. Rompel, Chem. Eur. J.
2015, 21, 4762–4771.
[25] A. Dolbecq, E. Dumas, C. R. Mayer, P. Mialane, Chem. Rev. 2010, 110,
6009–6048.
[49] G. Zhang, B. Keita, C. T. Craescu, S. Miron, P. de Oliveira, L. Nadjo, Bio-
macromolecules 2008, 9, 812–817.
[50] L. Zheng, Y. Ma, G. Zhang, J. Yao, B. Keita, L. Nadjo, Phys. Chem. Chem.
Phys. 2010, 12, 1299–1304.
[51] Y. Moriyama, D. Ohta, K. Hachiya, Y. Mitsui, K. Takeda, J. Protein Chem.
1996, 15, 265–272.
[52] F. Yang, J. Yue, L. Ma, Z. Ma, M. Li, X. Wu, H. Liang, Mol. Pharmaceutics
2012, 9, 3259–3265.
[26] A. Proust, B. Matt, R. Villanneau, G. Guillemot, P. Gouzerh, G. Izzet,
Chem. Soc. Rev. 2012, 41, 7605–7622.
Received: June 24, 2015
[27] B. Hasenknopf, R. Delmont, P. Herson, P. Gouzerh, Eur. J. Inorg. Chem.
2002, 1081–1087.
Published online on November 3, 2015
Chem. Eur. J. 2015, 21, 17800 – 17807
www.chemeurj.org
17807 ꢀ 2015 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim