Selective hydrolysis of hen egg white lysozyme at Asp-X peptide bonds promoted by oxomolybdate

Article abstract of DOI:10.1016/j.jinorgbio.2014.03.006

The activity of oxomolybdate(VI) towards hen egg white lysozyme (HEWL) was examined under physiological and slightly acidic pH conditions. Purely hydrolytic cleavage of HEWL in the presence of 10 to 100 mM of oxomolybdate(VI) after incubation at pH 5.0 and 60 °C for 2 to 7 days was observed in SDS-PAGE experiments. Four cleavage sites, which all occurred at Asp-X sequences and included the Asp18-Asn19, Asp48-Gly49, Asp52-Trp53 and Asp101-Gly102 peptide bonds, were identified with Edman degradation. The molecular interaction between [MoO₄]^{2 -} and HEWL was studied by circular dichroism (CD) and ¹H-¹⁵N heteronuclear single quantum correlation (HSQC) NMR spectroscopy. CD spectroscopy revealed a significant decrease in the α-helical content of HEWL upon addition of oxomolybdate, while ¹H-¹⁵N HSQC NMR spectroscopy identified the residues which were most affected upon interaction with [MoO₄]^{2 -}. ⁹⁵Mo NMR measurements, performed on oxomolybdate solutions containing HEWL, identified the monomeric [MoO₄]^{2 -} form as active species in the hydrolytic reaction. The hydrolysis of the Asp-Gly model peptide in the presence of oxomolybdate(VI) was studied by ¹H NMR, further supporting a hydrolytic mechanism where polarisation of the carbonyl is followed by internal nucleophilic attack on the Asp residue.

Full text of DOI:10.1016/j.jinorgbio.2014.03.006

Journal of Inorganic Biochemistry 136 (2014) 73–80
Contents lists available at ScienceDirect
Journal of Inorganic Biochemistry
journal homepage: www.elsevier.com/locate/jinorgbio
Selective hydrolysis of hen egg white lysozyme at Asp-X peptide bonds
promoted by oxomolybdate
Karen Stroobants a, Phuong Hien Ho a, Eva Moelants b, Paul Proost b, Tatjana N. Parac-Vogt a,
⁎
a
Department of Chemistry, KU Leuven, Belgium
b
Department of Microbiology and Immunology, KU Leuven, Belgium
a r t i c l e i n f o
a b s t r a c t
Article history:
The activity of oxomolybdate(VI) towards hen egg white lysozyme (HEWL) was examined under physiological
and slightly acidic pH conditions. Purely hydrolytic cleavage of HEWL in the presence of 10 to 100 mM of
oxomolybdate(VI) after incubation at pH 5.0 and 60 °C for 2 to 7 days was observed in SDS-PAGE experiments.
Four cleavage sites, which all occurred at Asp-X sequences and included the Asp18-Asn19, Asp48-Gly49,
Asp52-Trp53 and Asp101-Gly102 peptide bonds, were identified with Edman degradation. The molecular inter-
action between [MoO4]2− and HEWL was studied by circular dichroism (CD) and 1H-15N heteronuclear single
quantum correlation (HSQC) NMR spectroscopy. CD spectroscopy revealed a significant decrease in the α-
helical content of HEWL upon addition of oxomolybdate, while 1H-15N HSQC NMR spectroscopy identified the
Received 12 December 2013
Received in revised form 11 March 2014
Accepted 11 March 2014
Available online 12 April 2014
Keywords:
Oxomolybdate
Hen egg white lysozyme
Hydrolysis
Proteins
Metal complexes
Oxoanions
residues which were most affected upon interaction with [MoO4]2− 95Mo NMR measurements, performed on
.
oxomolybdate solutions containing HEWL, identified the monomeric [MoO4]2− form as active species in the hy-
drolytic reaction. The hydrolysis of the Asp-Gly model peptide in the presence of oxomolybdate(VI) was studied
by 1H NMR, further supporting a hydrolytic mechanism where polarisation of the carbonyl is followed by internal
nucleophilic attack on the Asp residue.
© 2014 Elsevier Inc. All rights reserved.
1. Introduction
ions can significantly increase the efficiency of synthetic proteases [1].
The efficient hydrolytically active agents require a strong Lewis acidity,
Selective cleavage of peptides and proteins is one of the most re-
quired and most important procedures in biochemical and bioengineer-
ing practice. Currently, either proteolytic enzymes or chemical agents
are used to achieve the peptide bond hydrolysis, however, both have
several drawbacks and often lack the required selectivity [1,2]. While
non-specific peptidases generate too small fragments for most bio-
chemical techniques, rather large peptides are obtained with costly spe-
cific peptidases. In addition, enzymatic proteins can only be used in
aqueous solution and tuning of their selectivity is not possible [1].
Chemical agents, which in some cases are used as alternative, also suffer
from several shortcomings. The best known agent, CNBr, requires harsh
conditions, results in low yields and causes an irreversible alteration of
the methionine residue where cleavage takes place [3]. Furthermore,
hydrolysis leads to fragments larger than the optimal size for most bio-
medical practices. Therefore, the development of reagents that selec-
tively cleave proteins under mild conditions is still considered as a big
challenge [1]. Since pure organic compounds cannot provide strong
attacking groups required for the hydrolysis at neutral pH in aqueous
solutions, it has been recognized long ago that the presence of metal
oxophilicity, a high coordination number and fast ligand exchange ki-
netics. Since many transition metals meet these requirements, the de-
sign and synthesis of transition metal complexes that hydrolyse
peptide bonds has become an area of intensive study in recent years.
Salts and complexes of Co(III) [4,5], Cu(II) [6,7], Ni(II) [8], Mo(IV) [9],
Pd(II) [10], Pt(II) [11], Zn(II) [12] and Zr(IV) [S.J. Shealy, K.B. Grant, Ab-
stracts of Papers of the American Chemical Society, vol. 231, 2006, pp.
133] were shown to hydrolyze peptide bonds in dipeptides,
oligopeptides and proteins. The highest selectivity was achieved for
Pd(II) and Pt(II) assisted cleavage at X-Ser/Thr/His and Met-X bonds, re-
spectively, however very low pH values were required for these reac-
tions [10,11]. Hydrolysis at physiological pH was observed in a few
cases including the cleavage of hen egg white lysozyme (HEWL) by
Co(III) salts [4], the hydrolysis of bovine serum albumin and HEWL
upon incubation with a Cu(II) complex [7] and the fragmentation of
tripeptides in the presence of Cp2MoCl2 [9]. While for the former two
cases no clear specificity was reported, anchoring of Cp2MoCl2 to the
thiolate of Cys and consequent promotion of an Mo–OH nucleophilic at-
tack towards the amide carbonyl was proposed in case of the
molybdocene induced hydrolysis in short peptides [9]. However, no fur-
ther studies on proteins were reported and so it is not clear whether a
similar mechanism would occur in larger molecules. In general, efficient
and selective hydrolysis at physiological pH has rarely been achieved
⁎
Corresponding author.
E-mail address: Tatjana.Vogt@chem.kuleuven.be (T.N. Parac-Vogt).
http://dx.doi.org/10.1016/j.jinorgbio.2014.03.006
0162-0134/© 2014 Elsevier Inc. All rights reserved.

74
K. Stroobants et al. / Journal of Inorganic Biochemistry 136 (2014) 73–80
and remains a challenging objective in transition metal based artificial
protease design.
was combined with an EV243 power supply (both produced by Con-
sort) at 200 V for 1.5 h. SDS-PAGE gels were stained with silver and an
image of each gel was taken with a GelDoc EZ Imager (Bio-Rad).
SDS-PAGE gels were blotted onto a polyvinylidene fluoride (PVDF)
membrane, the bands were cut after the membrane was stained,
destained and rinsed with water. The bands were subjected to automat-
ed NH2-terminal amino acid sequence analysis [33] (Procise 491 cLC
protein sequencer, Applied Biosystems, Foster City, CA) based on the
Edman degradation reaction as described in the publication cited above.
1H NMR spectra were recorded on a Bruker Avance 400 (400.13
MHz) spectrometer at 293 K. Samples of 500 μL containing 2 mM of
aspartyl-glycine (Asp-Gly) in the absence or presence of 120 mM of
oxomolybdate(VI) were incubated at pD 5.0 or 7.0. The pD value was
adjusted with DCl and NaOD in D2O. The pH-meter reading was
corrected by the equation: pD = pH + 0.41 [34]. As an internal refer-
ence, 0.5 mM of 3-trimethylsilyl-1 propionic acid was used. The samples
were measured directly after mixing and at different reaction times, and
kept at 60 °C between measurements.
Inspired by several reports on the biological role of oxovanadates
and oxomolybdates, our group has initiated the study of these com-
plexes as potential hydrolytic agents with reactivity towards relevant
biomolecules such as RNA, DNA and proteins. Vanadates were previous-
ly shown to inhibit several proteolytic enzymes [13]. Furthermore, the
interaction with and photooxidative cleavage of different proteins
were reported in the presence of vanadate upon UV irradiation
[14–18]. Binding of molybdates to nucleotides [19] and ATP [20] result-
ed in the hydrolytic cleavage of carbohydride and phosphoanhydride
bonds, respectively. Interestingly, antitumor activity [21,22] was report-
ed for both oxovanadate and oxomolybdate complexes. Initial studies of
our group have investigated the reactivity of V(V) and Mo(VI)
oxoanions towards DNA and RNA model systems such as 4-
nitrophenylphosphate (NPP), bis-4-nitrophenylphosphate (BNPP) and
2-hydroxypropyl-4-nitrophenylphosphate (HPNP). Hydrolysis of NPP
and BNPP was achieved in the presence of both oxoanions, and the ac-
tive species were identified to be the polynuclear [V10O28
[Mo7O24
6− [24,25] clusters. The reaction mechanism was studied in
great detail for the latter and incorporation of the substrate into the flex-
ible [Mo7O24
6− structure was shown to lead to ester bond polarisation
]
6− [23] and
95Mo NMR spectra were recorded on a Bruker Avance 600
(39.110 MHz) spectrometer at 293 K. Samples of 3 mL containing
30 mM of oxomolybdate(VI) in the absence or presence of 1.5 mM
HEWL were measured after mixing and adjustment of the pD to a
value of 5.0. The pD value was adjusted with DCl and NaOD in D2O. The
pH-meter reading was corrected by the equation: pD = pH + 0.41 [34].
CD (circular dichroism) spectroscopy experiments were performed
with a JACSO J-810 spectropolarimeter. Far UV spectra were recorded be-
tween 200 and 260 nm at 20 °C by using 1 mm quartz cells. Samples of
3 mL containing 5 μM of HEWL and 0 to 200 μM of oxomolybdate were
prepared in acetate buffer (100 mM) at pH 5.0 and measured after mixing.
15 N-1H HSQC (heteronuclear single quantum correlation) spectra
were recorded on a Bruker Avance 600 (600.13 MHz) spectrometer at
293 K. For the acquisition of the HSQC spectra, 768 scans were collected
for each FID, using the ‘hsqcetf3gf’ pulse programme with 512 data
points in the F1 dimension and 3072 data points in the F2 dimension
with spectral width of 2128 Hz (15N) and 9615 Hz (1H). Samples of
500 μL containing 1 mM of HEWL in the absence and presence of
4 mM of oxomolybdate(VI) and a pH of 5.2 in 10% D2O were measured
after mixing. The pH value was adjusted with HCl and NaOH.
]
]
and consequent nucleophilic attack by water. Conversion of the active
Mo(VI) species to [P2Mo5O23]6− upon reaction prevented catalytic ac-
tivity [25]. However, catalytic cleavage of HPNP was achieved with the
same [Mo7O24]
6− reagent [26] and this substrate was also hydrolyzed
in the presence of oxovanadate [27]. These promising results have en-
couraged us to further investigate the potential of oxovanadate and
oxomolybdate as reactive species towards biological model systems
and recently we have reported the hydrolysis of serine-containing pep-
tides promoted in the presence of oxovanadate and oxomolybdate an-
ions [28,29]. It was shown that polarisation of the carbonyl group of
the residue adjacent to Ser upon binding to the V(V) or Mo(VI) transi-
tion metal, respectively, led to peptide bond activation, facilitating an in-
ternal attack of the hydroxyl group of the Ser residue. The reaction was
proven to be purely hydrolytic in nature, demonstrating the first exam-
ple of hydrolytic peptide bond cleavage promoted by negatively charged
oxoanions. While previously reported hydrolytically active metal-
substituted polyoxotungstate reactivities towards proteins were relying
on the Lewis acidity of an incorporated metal ion [30,31], the demon-
strated reactivity in the current paper originates from the oxomolybdate
structure itself.
3. Results and discussion
3.1. Hydrolysis of HEWL promoted by molybdate(VI)
In this paper we further study the reactivity of oxomolybdate(VI) to-
wards HEWL in order to explore the ability of oxoanions to hydrolyze
larger biomolecules. Hydrolysis of HEWL, a protein consisting of 129
amino acids was examined under neutral and slightly acidic pH condi-
tions and the molecular interaction between oxomolybdate(VI) and
HEWL was studied by several complementary techniques.
Solutions of 1.0 μg/μL (0.07 mM) of HEWL were incubated with so-
dium molybdate at pH 5.0 and 60 °C during 2, 4, and 7 days. Since the
melting of HEWL was shown to occur only at about 75 °C, the protein
structure is expected to be stable at 60 °C [35]. However, to ensure the
stability of HEWL at 60 °C and after incubation, 1H NMR spectra of the
protein were measured both at room temperature and 60 °C after dis-
solving at pD 5 as well as incubation at pD 5 and 60 °C during 3 days
(Figure S1). Although some temperature effects occurred, the character-
istic peaks below 0 ppm which are indicative of protein folding [30],
were observed under all mentioned conditions. Therefore, it can be as-
sumed that HEWL is stable at 60 °C, also after incubation during longer
periods. The concentrations of oxomolybdate(VI) in the reaction mix-
tures were ranging from 10 to 100 mM. The SDS-PAGE experiments
were carried out after 2, 4, and 7 days and the results are shown in
Fig. 1. Two fragments with a molecular mass of ~12 and 9 kDa were ob-
served for all samples containing different concentrations of
oxomolybdate(VI). The intensity of these two bands increased with in-
creasing incubation time. The sum of the molecular mass of the two
bands is significantly larger than the molecular mass of HEWL
(14.3 kDa), suggesting that they result from the cleavage at two specific
sites of HEWL. The smaller fragments produced by the cleavage at these
two sites should have masses of ca. 2 and 5 kDa, which are too low to
detect on SDS-PAGE since they migrate together with the dye front.
2. Experimental section
HEWL and sodium molybdate were purchased from Acros and used
without further purification. The primary structure of HEWL is a single
polypeptide chain of 129 amino acids with a molecular mass of approx-
imately 14.3 kDa [32].
1H NMR spectra were recorded on
a Bruker Avance 400
(400.13 MHz) spectrometer. Solutions containing HEWL (1 mM) were
prepared in D2O and the pD of the solution was adjusted with DCl/
NaOD to achieve a pD value of 5 (based on pD = pH value + 0.41).
The sample was measured both at room T and 60 °C after dissolving,
heated for 3 days at 60 °C and measured again both at room T and 60 °C.
SDS-PAGE was carried out to visualize the hydrolytic activity. Sam-
ples of 1.5 mL containing 1 μg/μL (0.07 mM) of HEWL and 0, 10, 20, 50
and 100 mM of oxomolybdate(VI) were incubated at pH 5.0 or 7.0 and
60 °C in the presence or absence of light. Sample aliquots of 100 μL
were taken after 2, 4 and/or 7 days. An OmniPAGE electrophoretic cell

K. Stroobants et al. / Journal of Inorganic Biochemistry 136 (2014) 73–80
75
Fig. 1. SDS-PAGE of HEWL. Hydrolysis of 1.0 μg/μL (0.07 mM) of HEWL in the presence of oxomolybdate(VI) (10, 20, 50, and 100 mM) after 2, 5, and 7 days at pH 5.0 and 60 °C.
Interestingly, although the aqueous speciation of oxoanions is highly
dependent on the starting concentration of molybdate, the identical
cleavage sites observed in all experiments indicate that the same type
of oxoanion is the hydrolytically active species in all cases. Hydrolysis
of HEWL was also observed in the presence of oxovanadate, however,
the reactivity was considerably lower in this case.
The brown bands that appeared at the bottom of the gels were only
detected in the samples where oxomolybdate(VI) is present. To test the
origin of these bands, a control sample containing only the sodium mo-
lybdate solution and no HEWL was used in a similar SDS-PAGE experi-
ment (data not shown). As expected, the brown band also appeared,
confirming that these bands do not originate from protein fragments
but are due to the presence of molybdate. These are most likely prod-
ucts formed in the reaction between oxomolybdate(VI) and dithiothre-
itol (DTT) which is present in the sample buffer used for the SDS-PAGE
experiments [36].
Solutions of 1.0 μg/μL (0.07 mM) of HEWL were also incubated at
pH 7.0 in the presence of oxomolybdate(VI) with concentrations rang-
ing from 10 to 100 mM at 60 °C. SDS-PAGE was carried out after 2 and
7 days as shown in Fig. 2, and interestingly the same fragment with a
molecular mass of about 12 kDa was observed. Although a fragment of
similar size was observed as in the reaction performed at pH 5.0, the re-
action at higher pH appeared to be much slower.
As previous studies have shown that photoradiation of molybdate so-
lutions can result in reduction of oxomolybdate anions and can lead to
the formation of reactive species, the effect of visible light radiation was
also examined [37–41]. Therefore, solutions of 1.0 μg/μL (0.07 mM) of
HEWL were incubated with 20 mM of oxomolybdate(VI) at pH 5.0 and
60 °C and either left in the daylight or kept in the dark. After 2 days,
SDS-PAGE was carried out and the results are shown in Fig. 3. As can be
seen from the gels, there was no difference between these samples, sug-
gesting that the cleavage of HEWL promoted by oxomolybdate(VI) is not
photo-oxidative in nature. Furthermore, the Edman degradation experi-
ments (vide supra) unambiguously proved that the cleavage was hydro-
lytic in nature. These hydrolytic reactions occur at slower rates as
compared to the photo-oxidative cleavage of different proteins in
the presence of vanadate, where half lives of only few minutes were ob-
tained for comparable or up to 50 times lower oxometalate/protein ratios
[16,42]. However, the cleaved polypeptide fragments resulting from oxi-
dative cleavage are difficult to identify by Edman degradation [16] and in
general less useful in biochemical and biomedical procedures which re-
quire protein fragmentation.
3.2. Identification of the hydrolysis sites
The peptide fragments resulting from the HEWL hydrolysis in
the presence of oxomolybdate(VI), were identified by western
blot and subsequent NH2-terminal Edman degradation analysis of
a minimum of 5 residues. Fragments with the LysValPheGlyArg
NH2-terminus of HEWL (at ~12 kDa and b5 kDa) or one of the fol-
lowing NH2-terminal sequences were obtained: AsnTyrArgGlyTyr
(at ~ 12 kDa), GlySerThrAspTrp (at ~ 9 kDa), TrpGlyIleLeuGln (at
~ 9 kDa) and GlyAsnGlyMetAsn (at b 5 kDa). From these results
four cleavage sites at the Asp18-Asn19, Asp48-Gly49, Asp52-
Trp53 and Asp101-Gly102 peptide bonds were identified. Hydroly-
sis at these positions results in four pairs of fragments with molec-
ular masses of about 2.0/12.3, 5.4/8.9, 5.8/8.5 and 11.1/3.2 kDa,
respectively. These results are in good agreement with the esti-
mated masses of 12 and 9 kDa on the low resolution gels in
Fig. 1, as both bands seem to contain two fragments of around
12.3/11.1 and 8.9/8.5 kDa, respectively.
Strikingly, all four cleavage sites in HEWL occurred at an Asp-X amide
bond. The previous studies on oxomolybdate(VI) promoted hydrolysis of
dipeptides have indicated that the oxoanion has high affinity towards hy-
drolysis of X-Ser bonds, and interestingly both mechanisms of hydrolysis
are characterized by the requirement of side chain assistance in the hy-
drolysis of the adjacent peptide bond. In the hydrolysis of X-Ser dipep-
tides, higher reaction rates were observed due to an N → O acyl
rearrangement and internal attack on the Mo(VI) polarized carbonyl.
Similar internal nucleophilic attack is known in Asp residues, leading to
Asp-X peptide bond hydrolysis [43]. Polarisation of the carbonyl in the
presence of oxomolybdate(VI) and consequent internal nucleophilic at-
tack can lead to the observed Asp directed hydrolysis as shown in Fig. 4.
Polarisation of the carbonyl of the Asp residue might be in competition
with coordination to the carboxyl of the Asp side chain in accordance
with a proposed interaction between vanadate and phosphoenolpyr-
uvate [44]. However, the latter interaction will not lead to hydrolysis of
the peptide bond. Previous studies have shown that the carbonyl coordi-
nation hydrolysis mechanism is favoured at pH values below 6.0 and it is
in part responsible for the known labile properties of Asp containing

76
K. Stroobants et al. / Journal of Inorganic Biochemistry 136 (2014) 73–80
Fig. 2. SDS-PAGE of HEWL. Hydrolysis of 1.0 μg/μL (0.07 mM) of HEWL in the presence of oxomolybdate(VI) (10, 20, 50, 100, and 150 mM) after 2 and 7 days at pH 7.0 and 60 °C.
peptide bonds in acidic media [43]. While peptide bond hydrolysis was
shown to be dominant over succinimidyl peptide formation, which
also occurs at pH values below 6.0, formation of iso-Asp, D-Asp and D-
iso-Asp species are more favourable reactions above pH 6.0 as compared
to the peptide bond hydrolysis pathway. The pH dependence of the nu-
cleophilic attack of the Asp side chain on Asp-X peptide bonds can explain
the much faster rate of HEWL hydrolysis observed at pH 5.0 as compared
to pH 7.0, as well as the low reaction rate which was previously reported
for Asp-Ala hydrolysis in the presence of oxomolybdate(VI) at pH 7.0
[28]. Furthermore, the different pH conditions as well as changes in the
pKa of side chain residues in proteins might explain the distinct selectiv-
ities in dipeptide versus protein targets.
In order to gain more insight into the ability of oxomolybdate(VI) to
cleave Asp-X peptide bonds, the hydrolysis of the dipeptide Asp-Gly was
investigated at 60 °C and pD 5.0 and 7.0, both in the absence and
presence of oxomolybdate(VI). The reactions were conveniently follow-
ed by 1H NMR spectroscopy since the peaks corresponding to the CH of
the Asp residue in Asp-Gly (~4.25 ppm) as well as the peak correspond-
ing to the free Gly, which is the reaction product (~3.98 ppm), were well
separated throughout the reaction. Integration of the proton resonances
allowed for calculation of the hydrolysis rate constants at the different
conditions, which are presented in Table 1. The labile character of the
Asp-Gly bond is reflected by the high degree of background cleavage
both at pD 5.0 and 7.0. However, the rate of Asp-Gly hydrolysis at pD
5.0 is increased by 40% upon addition of oxomolybdate(VI), while no en-
hancement is observed at pD 7.0, which is in agreement with the earlier
proposed mechanism. While only 3% hydrolysis of Asp-Ala (2 mM) was
reported in the presence of 120 mM of oxomolybdate(VI) at 60 °C and
pD 7.0 after 60 h in a previous study [28], 10% conversion of Asp-Gly
was observed under the same conditions in the present study. It should
be noted in this regard that a considerable influence of the adjacent res-
idue was reported for the rate of Asp-X hydrolysis. The highest conver-
sion rates were observed for Asp-Gly hydrolysis, followed by the
peptides Asp-Pro, Asp-His and Asp-Ser [43]. In general, Asp-X bonds
are stabilized in the protein structure, as reflected by the fact that protein
fragmentation was not detected in the absence of oxomolybdate(VI)
while considerable background cleavage was observed for the hydrolysis
of Asp-Gly in the absence of oxomolybdate(VI).
Monomeric oxomolybdate was previously identified as the active
species in dipeptide bond hydrolysis by detailed kinetic experiments
[28]. The monomeric [MoO4]2− was also shown to coordinate to the
tripeptide glycylglycylglycine [45]. In contrary, no studies reporting on
peptide or protein interactions with [Mo7O24
periments also do not give evidence for this interaction. The [Mo7O24
]
6− are known, and our ex-
6−
]
has been previously reported to promote phosphodiester bond hydroly-
sis, a process which occurs via partial detachment of an MoO4 unit and
the incorporation of the structurally analogous phosphoester substrate
into the polyoxometalate structure. Due to the lack of any structural re-
semblance with MoO4, this process is unlikely to occur with peptides
and proteins [28]. However, as the speciation of oxomolybdate anions
is highly dependent on temperature, pH and concentration, it is inter-
esting to consider the species distribution under reaction conditions of
pH 5.0 and 7.0. In general, high pH and low concentrations of molybdate
favour [MoO4]2− in solution, while low pH and high concentrations fa-
vour the formation of polyoxoanions. We have previously developed a
thermodynamic model, which allowed us to calculate the percentage
Fig. 3. SDS-PAGE of HEWL. Hydrolysis of 1.0 μg/μL (0.07 mM) of HEWL in the presence
of 20 mM of oxomolybdate(VI) after 2 days at pH 5.0 and 60 °C in the absence (lane 1
and 2 — replicas) and presence (lane 3 and 4 — replicas) of light.

K. Stroobants et al. / Journal of Inorganic Biochemistry 136 (2014) 73–80
77
Fig. 4. Schematic representation of the Asp assisted peptide bond hydrolysis in the presence of molybdate(VI).
of different oxomolybdate species, depending on the conditions [24].
The formation constants of the different oxomolybdate species at
50 °C were used to estimate the [MoO4]2− percentage in 10, 30 and
100 mM solutions at pH 5.0 and 7.0, respectively (Table 2). The calcula-
tions reveal that under all conditions the [MoO4]2− anion is present in
solution, although its concentration ranges from ca. 9.8 to 23 mM in
10 to 100 mM solutions, respectively. However, since the HEWL concen-
trations in the hydrolysis experiments were in the order of 1 μg/μL
(0.07 mM), the active [MoO4]2− species is always in large excess with
single minimum at 215 nm. CD spectroscopy was previously applied
as a method to study POM-protein interactions [46–49].
HEWL mainly consists of α-helices and only contains a small β-plate
region, which is reflected in a CD spectrum with minima at 208 and
222 nm, characteristic for α-helical structure content. Solutions contain-
ing 5.0 μM of HEWL and varying concentrations of oxomolybdate(VI)
(ranging from 0 to 200.0 μM) were prepared in acetate buffer
(100 mM) at pH 5.0. As can be seen in Fig. 6, the CD signal intensity
gradually decreases upon the addition of increasing amounts of
oxomolybdate(VI). The changes in CD spectra suggest that interaction be-
tween oxomolybdate(VI) and HEWL takes place in solution which leads
to the loss of α-helical structure content of HEWL. In a previous study
on the hydrolysis of HEWL in the presence of the heteropolyoxometalate
Ce(IV)-Keggin, a similar decrease in α-helical structure content was ob-
served upon addition of 1 equivalent of the polyoxometalate complex.
However, as can be seen in Fig. 6, an increase in the oxomolybdate(VI)
concentration causes a gradual decrease in the α-helix content indicating
multiple electrostatic interactions between oxomolybdate(VI) and the
protein. Furthermore, the deformation of the peaks at the largest
oxomolybdate(VI) concentrations is indicative of denaturation of the
protein.
respect to HEWL. Furthermore,
a
95Mo NMR measurement of
molybdate(VI) in the presence of HEWL indicates slight shifting and se-
vere broadening of the characteristic [MoO4]2− peak as shown in Fig. 5
which presumably are caused by interactions with the protein. Similar
effects were previously observed for interaction between vanadate
and protein molecules. Different vanadate species were observed in
51V NMR and peak shifting, intensity decrease and broadening were ob-
served for two peaks, in accordance with those vanadate species
interacting with tubulin [39]. Interestingly, the peaks of the polyoxo
forms could not be detected in the 95Mo NMR spectrum of a 30 mM so-
lution either due to the excessive broadening in the presence of HEWL
or because the equilibrium between [MoO4]2− and [Mo7O24 6− is fully
]
shifted to the [MoO4]2− form upon addition of HEWL.
3.3.2. 1H-15N HSQC
The interaction of oxomolybdate(VI) with HEWL was further exam-
3.3. Interactions of oxomolybdate(VI) with HEWL
3.3.1. CD spectroscopy
CD spectroscopy was further used to gain information about the sec-
ondary structure of HEWL in solution. The secondary structure content
can be determined by CD spectroscopy in the far-UV spectral region as
α-helices have two minima at 208 and 222 nm, while β-sheets have a
ined by 1H-15N HSQC NMR spectroscopy. Measurement of the 1H-15
N
HSQC spectrum of HEWL in the absence and presence of the
oxometalate provides information on the shifting of amino acid resi-
dues and thus the positions where interaction with oxomolybdate(VI)
takes place. 1H-15N HSQC spectra were recorded on a 1.0 mM HEWL
sample in the absence and in the presence of 3.0 mM of Na2MoO4. The
Table 2
Table 1
Molybdate(VI) speciation in 10, 30, and 100 mM solutions at pH 5.0 and 7.0.
Asp-Gly hydrolysis rate constants in the absence and presence of oxomolybdate(VI) at pD
5.0 and 7.0 and 60 °C.
[MoO4]2− (initial)
[MoO4]2− (equilibrium)
% [MoO4]2−
pH 5
pH 7
10 mM
30 mM
100 mM
10 mM
30 mM
100 mM
9.8 mM
17 mM
23 mM
10 mM
30 mM
98%
59%
23%
100%
100%
100%
Asp-Gly+
pD
kobs (s−1
)
t1/2 (h)
/
5.0
5.0
7.0
7.0
1.68 × 10−6
2.34 × 10−6
0.87 × 10−6
0.84 × 10−6
114.5
82.3
220.4
228.8
molybdate(VI)
/
molybdate(VI)
100 mM

78
K. Stroobants et al. / Journal of Inorganic Biochemistry 136 (2014) 73–80
Fig. 5. 95Mo NMR spectrum of 30 mM of oxomolybdate(VI) in the presence of 1.5 mM of HEWL at pD 5.0 (bottom) as compared to a reference spectrum of sodium molybdate at high
concentration and pD 10 where only the [MoO4]2− is expected to be present (top). The shifting and severe broadening of the peak occur due to interactions with the protein.
chemical shift perturbations (CSP) of the measured peaks were calcu-
lated by the following equation:
cleaved peptide bonds. No shifting was seen for residues Asp18, Asn19,
Asp48, Gly49 and Asp101, however, several side chains surrounding
these positions are affected by the addition of oxomolybdate(VI). From
these results, higher preference for interaction at the Asp52-Trp53 and
Asp101-Gly102 peptide bonds might be expected. In addition to the
shifts observed in the vicinity of the cleavage sites, several other shifts
were also observed suggesting that only certain interactions can cause
peptide bond hydrolysis. Medium to large CSPs were observed for the
side chains of most solvent accessible Arg residues. These interactions
are most likely electrostatic in nature as they occur between the side
chain of Arg residues, containing a positively charged guanidinium
group, and the negatively charged molybdate. Although a shift was
seen for Ser36, cleavage did not occur at neighbouring peptide bonds
as this residue is distant from any of the hydrolysis positions in the 3D
HEWL structure. None of the nine remaining Ser residues were shifted.
Since two Ser residues, Ser50 and Ser100, are located near the Asp48-
Gly49 and Asp52-Trp53 hydrolysis sites, assistance of these residues
could be anticipated in accordance with the previously described Ser
side chain N → O acyl rearrangement mechanism [28]. However, since
no shifting was observed for either of both residues, the proposed Asp
side chain assistance mechanism is the most likely to occur also at
these positions. Modelling and docking studies are required to gain
more detailed understanding of differences in interaction and reaction
with dipeptide versus protein targets.
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2
ðΔδNÞ
2
Δδ ¼ ðΔδHÞ þ
25
where ΔδH is the difference in the 1H amide resonance before and after
ligand addition, while ΔδN is the 15N difference of the amide resonance
before and after ligand addition. Δδ is the chemical shift perturbation
(CSP) upon binding of the ligand, oxomolybdate(VI) in this case. The
CSP value allows the determination of the relative strength of interac-
tion between an amino acid residue and the ligand (Table 3).
By overlapping the 1H-15N HSQC NMR spectrum of HEWL in the ab-
sence of oxomolybdate(VI) with the one recorded in the presence of
oxomolybdate(VI), all the shifted peaks could be assigned. Upon visual-
isation of the observed shifts as well as the hydrolysis sites as shown in
Fig. 7, it is obvious that shifts occur at distinct positions as well as in the
surroundings of these sites. Small shifts are observed for residues Asp52,
Trp53 and Gly102, which are located in the immediate vicinity of the
3.4. Selectivity of the hydrolysis
The SDS-PAGE and Edman degradation experiments clearly indicated
that oxomolybdate selectively hydrolyzes HEWL at four distinct
Table 3
Relationship between CSP (Δδ) and interactions [50].
Δδ
Interaction
b0.015
No
0.015–0.025
0.025–0.05
N0.05
Small
Medium
Large
Fig. 6. CD spectra of 5.0 μM of HEWL in the presence of different concentrations of
oxomolybdate(VI) (0 → 200.0 μM:bottom→top) in 100 mM acetate buffer at pH 5.0.

K. Stroobants et al. / Journal of Inorganic Biochemistry 136 (2014) 73–80
79
Fig. 7. 1H, 15 N HSQC chemical shift perturbations (CSP) for HEWL resonances in the presence of 3 equivalents of oxomolybdate(VI). Shifted residues are schematically represented on the
cartoon visualisation of HEWL in the inset (residues with medium shifts as yellow spheres and residues with small shifts as green spheres). The hydrolysis positions are presented in red,
small shifts are seen for Asp52, Trp53 and Gly102, these are shown as red spheres.
positions having an Asp-X sequence. Visualisation of the cleavage sites
occurring at the Asp18-Asn19, Asp48-Gly49, Asp52-Trp53 and Asp101-
Gly102 peptide bonds on the three-dimensional structure of HEWL
shows that all four positions are located in solvent accessible loops. The
loops containing the Asp18-Asn19 and Asp101-Gly102 peptide bonds
are found in the α-helical part of the structure, while the Asp48-Gly49
and Asp52-Trp53 peptide bonds are located near the small β-sheet re-
gion of HEWL. Interaction in the α-helical region is clearly evidenced in
the CD spectra through the decrease in α-helical structure content
upon addition of increasing amounts of oxomolybdate(VI) to the
HEWL solution. Furthermore, several shifts in the 1H, 15N HSQC NMR
spectra were observed in the α-helical region of HEWL. Although bind-
ing effects in the β-sheet part of the structure could not be deduced
from the CD spectroscopy data direct shifts of Asp52 and Trp53 were
seen in the 1H, 15N HSQC NMR experiments.
While in previous studies with dipeptides, which were performed at
neutral pH, Gly-Ser was identified as the most favourable bond to be hy-
drolyzed in the presence of oxomolybdate(VI), only Asp-X bonds were
efficiently hydrolyzed in the reaction with HEWL. In these studies,
[MoO4]2− was identified as the active oxomolybdate(VI) species and al-
though it is not the most abundant species at pH 5.0 at all studied con-
centrations, it is likely to be the active species also in these protein
hydrolysis reactions. The preferential hydrolysis of Asp-X bonds at
pH 5.0 presumably is a consequence of the higher labile character of
these bonds in acidic conditions. A 100-fold faster cleavage of the Asp-
Gly bond compared to other peptide bonds was reported in dilute
acids [43]. The interaction of oxomolybdate with the carbonyl oxygen
leads to polarisation of the peptide bond thus facilitating the subsequent
nucleophilic attack by a hydroxyl group [28]. The importance of
oxomolybdate in the hydrolysis mechanism has also been experimen-
tally proven, as for control experiments no hydrolysis was observed ei-
ther in peptides or in HEWL in the absence of oxomolybdate.
oxomolybdate(VI) has high affinity towards the hydrolysis of X-Ser
bonds. For both mechanisms side chain assistance and polarisation of
the carbonyl bond by oxomolybdate are essential. However, while in
the hydrolysis of X-Ser dipeptides, an N → O acyl rearrangement and
an internal attack of the Ser side chain hydroxide on the carbon atom
of the peptide bond occur, in Asp-X peptide bond hydrolysis an internal
attack of the Asp side chain hydroxide on the peptide bond causes for-
mation of a cyclic anhydride with release of the C-terminal peptide
and followed by hydrolysis of the anhydride to the N-terminal peptide.
The interaction between oxomolybdate and HEWL leads to significant
decrease in the α-helical structure content of the protein, consistent
with the interaction near the cleavage sites. 1H,15N HSQC NMR spectros-
copy measurements revealed that not only residues in the immediate
vicinity of the four cleavage sites were affected upon addition of
oxomolybdate, but that Arg side chains were also implicated in the
binding. The latter interactions are presumably electrostatic in nature
and occur between the negatively charged oxomolybdate anion and
the positively charged guanidinium group of the Arg side chains. Fur-
ther experiments, including docking studies, will be needed in order
to fully elucidate the mechanism of this novel reaction.
Acknowledgements
K.S. thanks the FWO Flanders (Belgium) for the doctoral fellowship.
P. H. H. thanks the Vietnamese Government and KU Leuven for a doctoral
fellowship. T. N. P. V. thanks KU Leuven (OT/13/060) and FWO Flanders
(G.0864.14 N) for the financial support. P.P. thanks the Concerted Re-
search Actions (G.O.A./2013/014) of the KU Leuven for the support.
Appendix A. Supplementary data
Supplementary data to this article can be found online at http://dx.
doi.org/10.1016/j.jinorgbio.2014.03.006.
4. Conclusions
In this study we report the first example of purely hydrolytic protein
cleavage promoted by an oxomolybdate ion. The cleavage of HEWL was
shown to be remarkably selective for Asp-X sequences, and occurred at
the Asp18-Asn19, Asp48-Gly49, Asp52-Trp53 and Asp101-Gly102
peptide bonds. Previous studies with dipeptides have shown that
References
[1] K.B. Grant, M. Kassai, Curr. Org. Chem. 10 (2006) 1035–1049.
[2] J.P. Whitelegge, S.M. Gomez, K.F. Faull (Eds.), Proteomics of Membrane Proteins, Ac-
ademic Press, 2003.
[3] B.J. Smith (Ed.), Protein Sequencing Protocols, Humana Press, Totowa, 1997.

80
K. Stroobants et al. / Journal of Inorganic Biochemistry 136 (2014) 73–80
[4] C.V. Kumar, A. Buranaprapuk, A. Cho, A. Chaudhari, Chem. Commun. (2000)
[31] K. Stroobants, G. Absillis, E. Moelants, P. Proost, T.N. Parac-Vogt, Chem. Eur. J. (2013),
http://dx.doi.org/10.1002/chem.201303622.
597–598.
[5] J.W. Jeon, S.J. Son, C.E. Yoo, I.S. Hong, J. Suh, Bioorg. Med. Chem. 11 (2003)
2901–2910.
[32] C.C.F. Blake, D.F. Koenig, G.A. Mair, A.C.T. North, D.C. Phillips, V.R. Sarma, Nature 206
(1965) 757–761.
[6] M.S. Kim, J. Suh, Bull. Korean Chem. Soc. 26 (2005) 1911–1920.
[7] V. Rajendiran, M. Palaniandavar, P. Swaminathan, L. Uma, Inorg. Chem. 46 (2007)
10446–10448.
[8] W. Bal, R.T. Liang, J. Lukszo, S.H. Lee, M. Dizdaroglu, K.S. Kasprzak, Chem. Res.
Toxicol. 13 (2000) 616–624.
[33] T. Loos, A. Mortier, P. Proost, in: M.H. Tracy, J.H. Damon (Eds.), Methods in Enzymol-
ogy, vol. 461, Academic Press, 2009, pp. 3–29.
[34] Covingto Ak, M. Paabo, R.A. Robinson, R.G. Bates, Anal. Chem. 40 (1968) 700.
[35] T. Knubovets, J.J. Osterhout, P.J. Connolly, A.M. Klibanov, Proc. Natl. Acad. Sci. U. S. A.
96 (1999) 1262–1267.
[9] A. Erxleben, Inorg. Chem. 44 (2005) 1082–1094.
[36] S.J.N. Burgmayer, E.I. Stiefel, Inorg. Chem. 27 (1988) 2518–2521 (American
Chemical Society).
[37] C.V. Kumar, A. Buranaprapuk, G.J. Opiteck, M.B. Moyer, S. Jockusch, N.J. Turro, Proc.
Natl. Acad. Sci. U. S. A. 95 (1998) 10361–10366.
[38] T. Jyotsna, K. Bandara, C.V. Kumar, Photochem. Photobiol. Sci. 7 (2008) 1531–1539
(The Royal Society of Chemistry).
[39] J.J. Correia, L.D. Lipscomb, J.C. Dabrowiak, N. Isern, J. Zubieta, Arch. Biochem. Biophys.
309 (1994) 94–104.
[10] L. Zhu, N.M. Kostic, Inorg. Chim. Acta 339 (2002) 104–110.
[11] N.M. Milovic, L.M. Dutca, N.M. Kostic, Chem. Eur. J. 9 (2003) 5097–5106.
[12] M. Yashiro, Y. Kawakami, J.I. Taya, S. Arai, Y. Fujii, Chem. Commun. (2009)
1544–1546.
[13] D.C. Crans, Pure Appl. Chem. 77 (2005) 1497–1527.
[14] C.R. Cremo, G.T. Long, J.C. Grammer, Biochemistry 29 (1990) 7982–7990.
[15] C.R. Cremo, J.A. Loo, C.G. Edmonds, K.M. Hatlelid, Biochemistry 31 (1992) 491–497.
[16] J.C. Grammer, J.A. Loo, C.G. Edmonds, C.R. Cremo, R.G. Yount, Biochemistry 35
(1996) 15582–15592.
[40] J.C. Grammer, J.A. Loo, C.G. Edmonds, C.R. Cremo, R.G. Yount, Biochemistry 35
(1996) 15582–15592 (American Chemical Society).
[17] S.M. Hua, G. Inesi, J. Biol. Chem. 275 (2000) 30546–30550.
[18] E.E. Fetsch, A.L. Davidson, Proc. Natl. Acad. Sci. U. S. A. 99 (2002) 9685–9690.
[19] H. Weil-Malherbe, R.H. Green, Biochem. J. 49 (1951) 286–292.
[20] E. Ishikawa, T. Yamase, J. Inorg. Biochem. 100 (2006) 344–350.
[21] A.M. Evangelou, Crit. Rev. Oncol. Hematol. 42 (2002) 249–265.
[22] H. Yanagie, A. Ogata, S. Mitsui, T. Hisa, T. Yamase, M. Eriguchi, Biomed.
Pharmacother. 60 (2006) 349–352.
[23] N. Steens, A.M. Ramadan, G. Absillis, T.N. Parac-Vogt, Dalton Trans. 39 (2010)
585–592.
[24] L. Van Lokeren, E. Cartuyvels, G. Absillis, R. Willem, T.N. Parac-Vogt, Chem. Commun.
(2008) 2774–2776.
[25] E. Cartuyvels, G. Absillis, T.N. Parac-Vogt, Chem. Commun. (2008) 85–87.
[26] G. Absillis, E. Cartuyvels, R. Van Deun, T.N. Parac-Vogt, J. Am. Chem. Soc. 130 (2008)
17400–17408.
[27] N. Steens, A.M. Ramadan, T.N. Parac-Vogt, Chem. Commun. (2009) 965–967.
[28] P.H. Ho, K. Stroobants, T.N. Parac-Vogt, Inorg. Chem. 50 (2011) 12025–12033
(American Chemical Society).
[41] G. Zhang, B. Keita, C.T. Craescu, S. Miron, P. de Oliveira, L. Nadjo, J. Phys. Chem. B 111
(2007) 11253–11259 (American Chemical Society).
[42] C.M. Bergamini, M. Signorini, S. Hanau, M. Rippa, P.P. Delaureto, M.A. Cremonini,
Arch. Biochem. Biophys. 321 (1995) 1–5.
[43] B. Testa, J.M. Mayer (Eds.), Hydrolysis in Drug and Prodrug Metabolism, Wiley-VCH,
Weinheim, 2003.
[44] M. Aureliano, J. Leta, V.M.C. Madeira, L. Demeis, Biochem. Biophys. Res. Commun.
201 (1994) 155–159.
[45] T. Yamase, M. Inoue, H. Naruke, K. Fukaya, Chem. Lett. (1999) 563–564.
[46] G. Zhang, B. Keita, J.-C. Brochon, P. de Oliveira, L. Nadjo, C.T. Craescu, S. Miron, J. Phys.
Chem. B 111 (2007) 1809–1814 (American Chemical Society).
[47] G. Zhang, B. Keita, C.T. Craescu, S. Miron, P. de Oliveira, L. Nadjo, Biomacromolecules
9 (2008) 812–817 (American Chemical Society).
[48] L. Zheng, Y. Ma, G. Zhang, J. Yao, B.S. Bassil, U. Kortz, B. Keita, P. de Oliveira, L.
Nadjo, C.T. Craescu, S. Miron, Eur. J. Inorg. Chem. 2009 (2009) 5189–5193
(WILEY-VCH Verlag).
[49] L. Zheng, Y. Ma, G. Zhang, J. Yao, B. Keita, L. Nadjo, Phys. Chem. Chem. Phys. 12
[29] P.H. Ho, T. Mihaylov, K. Pierloot, T.N. Parac-Vogt, Inorg. Chem. 51 (2012) 8848–8859.
[30] K. Stroobants, E. Moelants, H.G.T. Ly, P. Proost, K. Bartik, T.N. Parac-Vogt, Chem. Eur.
J. 19 (2013) 2848–2858.
(2010) 1299–1304 (The Royal Society of Chemistry).
[50] A.S. Kanai, A, Y. Fukuda, K. Okada, T. Matsuda, T. Sakamoto, Y. Muto, S. Yokoyama, G.
Kawai, M. Tomita, RNA 15 (2009) 420–431.