Inhibition of human prion neuropeptide PrP106-126 aggregation by hexacoordinated ruthenium complexes

Article abstract of DOI:10.1016/j.jinorgbio.2013.07.009

Prion disease is a neurodegenerative disorder that can occur among humans and other animals. The aberrant isoform of prion protein PrP^Sc has been identified as the infectious agent. The neuropeptide PrP106-126 has been widely used as a suitable model to study the biological and physiochemical properties of PrP^Sc. PrP106-126 shares several physicochemical and biological properties with PrP^Sc, including cellular toxicity, fibrillogenesis, and membrane-binding affinity. Ruthenium complexes are commonly employed in anti-cancer studies due to their low cellular toxicity. In this study, six hexacoordinated ruthenium complexes with different molecular configurations were used to investigate their effects on PrP106-126 aggregation inhibition. Results revealed that the interaction between the complexes and the peptide included metal coordination and hydrophobic interaction mainly. Those complexes with aromatic structure displayed better inhibitory effects, although they only had a common binding affinity to PrP106-126. This study provided better understanding on the interaction of metal complexes with PrP106-126 and paved the way for potential Ru-based metallodrugs against prion diseases.

Full text of DOI:10.1016/j.jinorgbio.2013.07.009

Journal of Inorganic Biochemistry 128 (2013) 1–10
Contents lists available at SciVerse ScienceDirect
Journal of Inorganic Biochemistry
journal homepage: www.elsevier.com/locate/jinorgbio
Inhibition of human prion neuropeptide PrP106-126 aggregation
by hexacoordinated ruthenium complexes
⁎
Xuesong Wang, Bingbing Zhang, Cong Zhao, Yanli Wang, Lei He, Menghan Cui, Xiaotong Zhu, Weihong Du
Department of Chemistry, Renmin University of China, Beijing 100872, China
a r t i c l e i n f o
a b s t r a c t
Article history:
Prion disease is a neurodegenerative disorder that can occur among humans and other animals. The aberrant
isoform of prion protein PrPSc has been identified as the infectious agent. The neuropeptide PrP106-126 has
Received 8 March 2013
Received in revised form 5 July 2013
Accepted 8 July 2013
been widely used as a suitable model to study the biological and physiochemical properties of PrPSc
.
PrP106-126 shares several physicochemical and biological properties with PrPSc, including cellular toxicity,
fibrillogenesis, and membrane-binding affinity. Ruthenium complexes are commonly employed in anti-can-
cer studies due to their low cellular toxicity. In this study, six hexacoordinated ruthenium complexes with
different molecular configurations were used to investigate their effects on PrP106-126 aggregation inhibi-
tion. Results revealed that the interaction between the complexes and the peptide included metal coordina-
tion and hydrophobic interaction mainly. Those complexes with aromatic structure displayed better
inhibitory effects, although they only had a common binding affinity to PrP106-126. This study provided bet-
ter understanding on the interaction of metal complexes with PrP106-126 and paved the way for potential
Ru-based metallodrugs against prion diseases.
Available online 15 July 2013
Keywords:
Prion neuropeptide
Aggregation
Ruthenium complex
Inhibition
Ligand configuration
© 2013 Published by Elsevier Inc.
1. Introduction
different secondary structure elements. PrPSc has significantly more β
sheets and lesser α helix compared with PrPC [9]. PrPC–PrPSc conversion
Prion diseases or transmissible spongiform encephalopathies
(TSE) comprise a family of fatal neurodegenerative disorders. These dis-
eases infect humans with the Creutzfeldt–Jakob disease (CJD), the
Gerstmann–Sträussler–Scheinker syndrome (GSS), and Kuru, while
other animals are infected with bovine spongiform encephalopathy
(BSE), chronic wasting disease, and scrapie [1,2]. All these diseases
have been associated with the accumulation of misfolding protein,
which is an aberrant isoform of the prion protein (PrPC) and is named
as PrPSc [3,4]. PrPC is a common membrane protein expressed in the
entire body, predominantly in the central nervous system of mammali-
an species. However, the biological function of PrPC is still unclear.
Several experiments have suggested that PrPC can bind metal ions in
vivo, such as copper (Cu), zinc (Zn), and manganese (Mn), and is
involved in the homeostasis of trace elements, enzymatic activity, and
cellular signal transduction processes [5,6]. Earlier reports propose
that this protein may represent an important tool in the treatment
of cardiovascular diseases by angioplasty [7]. In vivo human and
chicken PrPs exhibit SOD-like activity associated with octarepeat and
hexarepeat regions. The fact that Cu (II) ion binds to hexapeptide repeat
region of avian prion via side chain imidazoles of histidine residue
reveals the highest SOD activity [8].
results in significant protein conformational change in their physico-
chemical properties, without any chemical modification.
PrPSc is resistant to protease K and shows a strong tendency toward
aggregation into insoluble fibrils that disrupt neuronal function. The
mechanism through which PrPSc catalyzes PrPC from helical protein to
β sheet-contained aggregates has not yet been elucidated, as well
as the complete structure of PrPSc. The neuropeptide PrP106-126
(106-KTNMKHMAGAAA AGAVVGGLG-126), a N-terminal fragment of
the human prion protein, has been widely used as a suitable model to
study the biological and physiochemical properties of PrPSc. PrP106-
126 is highly conserved among various species and has been suggested
as the most important region to initiate conformational change and to
lead the conversion of PrPC to PrPSc [10–12]. PrP106-126 shares several
physicochemical and biological properties with PrPSc, including cellular
toxicity, fibrillogenesis, and membrane-binding affinity [13]. Further-
more, the toxicity of PrPSc and PrP106-126 requires the expression of
PrPC to cause cell death, and they can bind with PrPC at 112–119
residues [14]. PrP106-126 is composed of two distinct regions, namely,
the hydrophilic (K106-M112) and hydrophobic (A113-G126) regions.
The toxicity of PrP106-126 has been assumed to be correlated with its
primary sequence. Different synthetic and recombinant PrP-derived
peptides (mainly PrP106-126), as far as mechanisms of aggregation
and amyloid formation are concerned have been studied to reveal
how these different spatial conformations affect neuronal death [15].
PrP106-126 shows a high tendency to aggregate into the β-sheet
structure, forming amyloid fibrils in vitro and becoming partially
The abnormal scrapie isoform PrPSc has been considered as the
infectious agent. PrPSc has an identical primary structure to PrPC, but
⁎
Corresponding author. Tel./fax: +86 10 62512660/6444.
E-mail address: whdu@chem.ruc.edu.cn (W. Du).
0162-0134/$ – see front matter © 2013 Published by Elsevier Inc.
http://dx.doi.org/10.1016/j.jinorgbio.2013.07.009

2
X. Wang et al. / Journal of Inorganic Biochemistry 128 (2013) 1–10
resistant to proteolysis K [16,17]. Recent studies reported that the
oligomerization of PrP106-126 emerged from the association of
ordered-hairpin monomers rather than of disordered monomers
[18,19]. Earlier-ordered oligomers were stacked by the interface of
hydrophobic C-terminal residues (A113-G126), which may increase
the rate of fibril growth and ultimately form the fibril structure [20].
Physical strategies for inhibiting amyloid fibril formation, including
high hydrostatic pressure, low temperature, and laser irradiation,
were critically evaluated [21]. Another work employed recombinant
yeast-derived molecular chaperon Hsp104 to inhibit the fibril
assembly of the synthetic PrP106-126 peptide. Hsp104 was able to
disassemble the mature PrP106-126 fibrils in vitro, recovering the
cytotoxicity of PrP106-126 on SK-N-SH cells [22].
In the present study, a series of hexacoordinated Ru complexes
(Scheme 1) have been synthesized, aiming to elucidate their interac-
tions with PrP106-126. In addition, changes in the physicochemical
properties of PrP106-126 induced by different Ru complexes are
observed and the relationship between the ligand property and aggre-
gation of this active peptide is clarified, for the development of novel
therapeutic agents against prion diseases.
2. Materials and methods
2.1. Materials
The relationship between the amyloid structure and its peptide
toxicity implies that PrP106-126 toxicity may be inhibited in case the
amyloid structure failed to form. A number of studies demonstrated
that divalent metal cation binding to PrP106-126 modulated its aggre-
gation and neurotoxic properties, and the histidyl residue is crucial to
bind metal cations [5,6,23–25]. Interestingly, recent studies found
potential therapeutic applications of metal compounds for the treat-
ment of Alzheimer's disease by targeting the metal binding site and by
alkylating the imidazole side chains of Aβ [24,26]. In addition, our pre-
vious investigation on the interactions of Pt-, Pd-, and Au-based metal
complexes with PrP106-126 demonstrated that tetra-coordinated com-
plexes markedly affected the conformation of PrP106-126 and inhibited
its aggregation by metal coordination, suggesting a new approach for
the design and development of metallodrugs against prion diseases
[27,28].
The well-developed synthetic chemistry of ruthenium (Ru), particu-
larly with am(m)ine and imine ligands, provides several approaches to
innovative metallopharmaceuticals [29]. Increasing knowledge on the
biological effects and lower cellular toxicity of Ru complexes had been
reported in the study of the fundamental characteristics of Ru com-
plexes on drug developments [30–32]. Ru complexes using am(m)ine
as ligands selectively bind to imine sites, such as histidyl imidazole
nitrogens on proteins and the N7 site on the imidazole ring of purine
nucleotides [29]. The role of Ru-based compounds in the interaction
with Aβ amyloid protein has also been previously reported [33,34].
Human prion protein fragment PrP106-126 was chemically synthe-
sized by the SBS Co. Ltd. (Beijing, China) and further purified and iden-
tified by HPLC and mass spectrometry (MS) with more than 95% purity.
The peptide in this study had free C and N termini, unless otherwise
specified. The Ru complexes were prepared as described in literature
[35–41]. The Ru complexes were dissolved in DMSO or d6-DMSO and
stored at −20 °C for future use. All other reagents were of analytical
grade.
2.2. Electrospray ionization mass spectrometry (ESI-MS)
The peptide concentration used in MS determination was
constant at 50 μM. Equivalent Ru complex was incubated with the
peptide. ESI-MS spectra were recorded in the positive mode via
direct introduction of the samples at 3 μL·min−1 flow rate using an
APEX IV FT-ICR high-resolution mass spectrometer (Bruker, USA),
equipped with a conventional ESI source. The working condition
included the following: end plate electrode voltage, −3500 V,
capillary entrance voltages, −4000 V, skimmer 1 voltage, 30 V, and
200 °C dry gas temperature. The flow rates of the drying and
nebulizer gases were set at 12 and 6 L·min−1, respectively. The
Data Analysis 4.0 software program (Bruker) was used for acquisi-
tion, and the deconvoluted masses were obtained using the integrat-
ed deconvolution tool.
Scheme 1. Molecular structure of hexacoordinated Ru-complexes.

X. Wang et al. / Journal of Inorganic Biochemistry 128 (2013) 1–10
3
2.3. NMR experiments
3. Results
The 1H–1H TOCSY (total correlation spectroscopy) experiments
were carried out to identify the probable binding site between
ruthenium complex and PrP106-126. Here the complexes NAMI-A
([trans-RuCl4(DMSO)(Im)][ImH]) and cis-(bpy)2RuCl2 (bpy: bipyridine)
were selected as the model. The final concentration of the peptide was
1 mM, and double amounts of complex were added to the peptide
solution. The sample for NMR study was prepared in H2O containing
10% d6-DMSO. The 2D NMR experiments were performed on a Bruker
Avance 600 MHz spectrometer at 25 °C. Suppression of the residual
water signal was achieved by the WATERGATE pulse program with
gradients.
3.1. Synthesis of the ruthenium complexes
Synthesis of cis-RuCl2(DMSO)4 was completed according to the
earlier literatures [35,36]: 78 mg RuCl3·3H2O was dissolved in 2.1 mL
DMSO and hydrogen was bubbled through at 80 °C for 10 h. The yellow
crystalline material was collected and washed with benzene and dried
under nitrogen (yield 88%). Recrystallization was performed from
benzene–DMSO solution. The complexes with DMSO ligand have only
simple and undistinguished NMR signal, so the infrared spectroscopy
(IR) was employed to confirm its identity. The IR spectral bands of
cis-RuCl2(DMSO)4, observed at 351 cm−1
,
383 cm−1
,
424 cm−1
,
,
,
482 cm−1, 677 cm−1, 717 cm−1, 937 cm−1, 960 cm−1, 993 cm−1
1024 cm−1
,
1097 cm−1
,
1114 cm−1
,
1290 cm−1
,
1309 cm−1
2.4. Circular dichroism spectroscopy
1400 cm−1, and 1419 cm−1, corresponded with the reported literature
[35].
Circular dichroism (CD) spectra were obtained through a Jasco J-810
spectropolarimeter (Japan Spectroscopy Co., Japan). The sample was
prepared in 5 mM phosphate buffer at pH 7.2. The final peptide concen-
tration was 0.1 mM. A 1 mm quartz cell was used for all CD spectra. The
spectra were recorded between 190 and 250 nm with 0.5 nm spectral
step and 2 nm bandwidth. A scan rate of 100 nm min−1 with 1 s re-
sponse time was employed. The background spectrum of corresponding
Ru-complex was removed using the same buffer. The final spectrum
for each sample was derived from the average of three repeated
experiments.
Synthesis of trans-RuCl2(DMSO)4 was carried out referring to earlier
report [37]: Recrystallized cis-RuCl2(DMSO)4 (50 mg) was dissolved in
4 mL of DMSO by gentle heating at 80 °C. The solution was transferred
into a water-cooled photoreactor equipped with a 125-W lamp and
irradiated for 4 h. During the reaction, the solution temperature was
kept close to room temperature. The whole procedure was conducted
under an inert gas atmosphere. The yield was 80%. Similarly, character-
istic IR bands at 1080 cm−1 and 416 cm−1 of the complex trans-
RuCl2(DMSO)4 were also detected.
Synthesis of the complex [(DMSO)2H][trans-Ru(DMSO)2Cl2] [38]:
The amount of 100 mg RuCl3·3H2O was refluxed in 10 mL of ethanol
for 3 h. The deep green solution was then filtered and vacuum-
evaporated to the volume of 1 mL. A 0.1 mL of 37% aqueous HCl and
0.2 mL of DMSO were added, and the solution was heated to 80 °C
under stirring. Within 10 min its color turned to red-orange. 2 mL of
acetone was added to the cooled transparent solution. The formed
red crystals were filtered off, washed with cold acetone and diethyl
ether, and vacuum-dried (yield 70%). The complex was also con-
2.5. Thioflavin T (ThT) assay
After the solution of 0.1 mM PrP106-126 at 10 mM phosphate buffer
at pH 7.2 was aged, equivalent Ru complex was added and the sample
was incubated with 10 μM ThT, and sample fluorescence was monitored
using an LS55 spectrofluorometer (PerkinElmer, USA). The ThT signal
was quantitatively measured by averaging the fluorescence emission at
approximately 500 nm for over 10 s, when excited at 432 nm. The final
spectrum was obtained from the mean of three repeated spectra. In
consideration of possible inner filter effect of Ru-complexes, ThT fluores-
cence spectrum was carried out after addition of aromatic-ring contain-
ing Ru-complexes.
firmed using IR spectrometry. The bands at 345 cm−1, 416 cm−1
,
968 cm−1, 1016 cm−1, 1112 cm−1, 1290 cm−1, 1301 cm−1 and
1398 cm−1 were verified.
Synthesis of NAMI-A [39]: 100 mg [(DMSO)2H][trans-Ru(DMSO)2-
Cl4] was dissolved in 2 mL of acetone and then stirred for 4 h after the
addition of 49 mg imidazole. The deposit turned into brick red from or-
ange. After filtration, the product was washed with acetone and diethyl
ether and vacuum-dried (yield 90%).
For the IC50 determination, the concentrations of Ru complexes were
selected at 0 μM, 20 μM, 40 μM, 60 μM, 100 μM and 200 μM respectively.
Synthesis of Ru(bpy)Cl2 [40]: 86 mg RuCl3·3H2O, 60 mg KCl and
93.6 mg 2,2′-bipyridine were dissolved in 5 mL of 1 M HCl and
stirred at 30 °C for 72 h. The deposit was filtered and washed with
acetone and diethyl ether, and vacuum-dried. The product was
suspended in 5 mL of 0.2 M HCl and bubbled with Cl2 for 15 min,
and then stirred for 1 h. After filtration, the almost black solid was
washed with ice-water and dried in a vacuum (yield 56%).
Synthesis of cis-(bpy)2RuCl2 [41]: 78 mg commercial RuC13·3H2O,
93.6 mg 2,2′-bipyridine and 84 mg LiCl were heated together at reflux
in 5 mL of DMSO for 8 h. After the reaction mixture was cooled to
room temperature, 5 mL of acetone was added and the resultant solu-
tion was cooled at 0 °C overnight. Filtering the yielded red to red–violet
solution, a dark green–black microcrystalline was produced. The solid
was washed three times with water followed by diethyl ether, and
then it was dried by suction (yield 76%).
2.6. Transmission electron microscopy (TEM)
Samples were prepared by mixing equivalent Ru complex with
5 mM peptide solution, and then incubated at 37 °C for 24 h. The final
peptide concentration used in the TEM experiment was 0.1 mM, with
1% DMSO. An aliquot of each sample was spotted onto carbon-coated
600-mesh copper grid and was negatively stained by 2% phosphotungstic
acid. Air-dried specimens were examined and photographed using a
Hitachi H-800 electron microscope (Hitachi, Japan) at 200 kV. The final
photo used was from the mean of three repeated data.
2.7. Atomic force microscopy (AFM)
The aromatic-ring containing ruthenium complexes were confirmed
Samples were prepared by mixing equivalent Ru complex with
5 mM peptide solution, and then incubated at 37 °C for 24 h. The
final peptide concentration used in the AFM experiment was 0.1 mM,
with 1% DMSO. Images were obtained in the tapping mode with a
silicon tip under ambient conditions, a scanning rate of 1 Hz, and a scan-
ning line of 512 using the Veeco D3100 instrument (Veeco Instruments
151 Inc., USA).
by NMR experiment and the 1H NMR spectra were shown in Fig. S1.
3.2. ESI-MS study on the binding of Ru complexes to PrP106-126
Duple amounts of Ru complexes were incubated with the peptide
PrP106-126, and then the final solution was analyzed by ESI-MS to
determine whether Ru complexes directly bind to PrP106-126. The

4
X. Wang et al. / Journal of Inorganic Biochemistry 128 (2013) 1–10
resulting ESI-MS deconvoluted spectra are shown in Figs. 1 and 2.
Free PrP106-126 exhibited an intense peak at 1911.99 Da (1+),
corresponding to its expected mass (Fig. 1A). The spectra appeared
different after various Ru complexes were incubated with PrP106-126.
Fig. 1B and C illustrated the mass spectrum of PrP106-126 after incuba-
tion with cis-RuCl2(DMSO)4 and trans-RuCl2(DMSO)4. The peak at
2160.84 Da was referred to the adduct of [PrP106-126 + RuCl2(DMSO)]
(Fig. 1B), whereas the m/z value of 1121.44 Da (2+) was assigned to the
adduct of [PrP106-126 + RuCl2(DMSO)2] (Fig. 1C). Fig. 1D demonstrated
that the treatment of peptide with [(DMSO)2H][trans-Ru(DMSO)2Cl4]
led to the emergence of additional peaks at 2153.77, 2231.78, and
2553.57 Da (1+), which were directly assigned to the metal–peptide
adducts. The increase in mass of 241, 319, and 641 Da matched the
fragments of RuCl−4 , RuCl4(DMSO)−, and 2[RuCl4(DMSO)−], respectively.
This result suggested that the compounds bound to PrP106-126 at 1:1
and 2:1 ratios with one or two displaced DMSO ligands due to hydrolysis
of the compound in aqueous solution.
The incubation of NAMI-A with PrP106-126 (Fig. 2B) produced a
different spectrum with new intensity peaks at 1151.92 and 1347.33
(2+). These results indicated that NAMI-A was bound to the peptide
at 1:1 and 2:1 ratios in the form of [Ru(DMSO)(Im)Cl4], which denoted
that the outside Him+ group in NAMI-A slipped off during the adduct
formation. The new peak at 2317.91 Da (1+) for the treatment
of Ru(bpy)Cl4 with PrP106-126 was referred to the adduct of
[PrP106-126 + Ru(bpy)(DMSO)Cl2] (Fig. 2C). The new adduct may
contribute to the hydrolysis of Ru(bpy)Cl4. The mass spectrum from
the incubation of cis-(bpy)2RuCl2 with PrP106-126 (Fig. 2D) showed
two obvious peaks assigned to the adducts of [PrP106-126 + Ru]
(2009.94 Da) and [PrP106-126 + Ru(bpy)Cl2] (2243.15 Da). The
ESI-MS results revealed that Ru complexes could bind to PrP106-126
in different binding modes. A table was given to show all the detected
m/z values and the estimated data of all adducts (Table S1). The differ-
ence between the detected data and estimated data may be attributed
to the dissociation of the hydrogen.
3.3. The probable binding sites between PrP106-126 and Ru complexes
2D NMR method was used to identify the possible binding site of
Ru Complex with PrP106-126. Here the complexes NAMI-A and
cis-(bpy)2RuCl2 were selected as the model. Fig. 3 compared the
1H–1H TOCSY spectra recorded with or without 2.0 equiv of
NAMI-A. The CδHs, NH of the His111 and CβHs, CγHs, and NH of
Met112 showed chemical shift change and weaker intensity due
to peak relaxation. Other spin system such as Lys110 was also slightly
affected (Fig. S2). Due to the appearance of [Ru(DMSO)(Im)Cl4] in
ESI-MS spectrum, hydrophobic interaction but not direct metal
coordination was assumed for the binding of NAMI-A with
PrP106-126. The possible binding site was from Lys110, His111
and Met112. In addition, the cis-(bpy)2RuCl2 complex did not
change the methionine spin system but the CδHs of His111 as
shown by peak broadening and chemical shift change (Fig. S3).
Moreover, the spin system of Lys110 was not disturbed by the com-
plex cis-(bpy)2RuCl2 (Fig. S4). Since the product of Ru(bpy)Cl2 was
observed in ESI-MS, free metal binding site was remained to the
peptide. And the most possible metal binding site was His111 for
the interaction of cis-(bpy)2RuCl2 and PrP106-126.
3.4. PrP106-126 conformational changes induced by Ru complexes
Circular dichroism (CD) spectroscopy was employed to examine
the effect of Ru complexes on the conformation of PrP106-126 at
physiological pH. The peptide in a 5 mM phosphate buffer at pH 7.2
exhibited a predominately negative CD absorbance at 197 nm as the
typical random coil structure. Negative ellipticities were also revealed
in the 210 nm to 230 nm region, suggesting that the peptide was
characterized by the presence of small amounts of secondary struc-
tural elements. The addition of equivalent Ru complexes induced
the disturbance of the negative CD absorbance at 197 nm (Fig. 4).
Cis-RuCl2(DMSO)4 and [(DMSO)2H][trans-Ru(DMSO)2Cl4] displayed
Fig. 1. Deconvoluted ESI-MS spectra of PrP106-126 in the absence (A) and presence of cis-RuCl2(DMSO)4 (B), trans-RuCl2(DMSO)4 (C), and [(DMSO)2H][trans-Ru(DMSO)2Cl4]
(D). The aqueous mixture was prepared by adding double amounts of Ru complexes to PrP106-126. The solution was diluted in water to a final concentration of 50 μM.

X. Wang et al. / Journal of Inorganic Biochemistry 128 (2013) 1–10
5
Fig. 2. Deconvoluted ESI-MS spectra of PrP106-126 in the absence (A) and presence NAMI-A (B), Ru(bpy)Cl4 (C), and cis-(bpy)2RuCl2 (D). The aqueous mixture was prepared by
adding double amounts of Ru complexes to PrP106-126. The solution was diluted in water to a final concentration of 50 μM.
the most effective change in the spectra. The difference is that
cis-RuCl2(DMSO)4 decreased the absorbance at 197 nm, whereas
[(DMSO)2H][trans-Ru(DMSO)2Cl4] increased the absorbance. In addition,
the complex trans-Ru(DMSO)4Cl2 and NAMI-A did not obviously
change the peptide conformation. The change in the CD spectra
of PrP106-126 with Ru(bpy)Cl4/cis-(bpy)2RuC12 was similar to
Fig. 3. Portion of 1H–1H TOCSY spectrum of 1 mM PrP106-126 at pH 5.7 and 25 °C in the absence (light green) and presence (dark blue) of 2.0 equiv of NAMI-A. (For interpretation
of the references to color in this figure legend, the reader is referred to the web version of this article.)

6
X. Wang et al. / Journal of Inorganic Biochemistry 128 (2013) 1–10
Table 1
that of [(DMSO)2H][trans-Ru(DMSO)2Cl4], which indicated the in-
The IC50 value for the inhibition of ruthenium complexes on PrP106-126 aggregation
measured with ThT fluorescence.
fluence of metal complexes on PrP106-126 conformation (Fig. S5).
Ruthenium complex
Inhibition of fibril formation IC50 (μM)
3.5. ThT analysis of PrP106-126 aggregation induced by Ru complexes
Cis-RuCl2(DMSO)4
Trans-RuCl2(DMSO)4
[(DMSO)2H][trans-Ru(DMSO)2Cl4]
NAMI-A
Ru(bpy)Cl4
Cis-(bpy)2RuCl2
81
22
8
10
6
The neuropeptide PrP106-126 is essential to the aggregation of
PrPSc, which is correlated with prion protein toxicity. The aggregation
of PrP106-126 can be monitored by ThT assay that results in a new
excitation at 432 nm and enhanced emission around 500 nm. Fig. 5
shows that the spectrum presented a strong fluorescence signal
when ThT was bound to PrP106-126. However, ThT fluorescence sig-
nificantly decreased after the incubation of PrP106-126 with Ru com-
plexes for 24 h, suggesting that the formation of the fibril structure
was effectively inhibited. Interestingly, cis-(bpy)2RuCl2 and Ru(bpy)
Cl4 complexes exhibited stronger inhibition effects on the aggrega-
tion of PrP106-126 after the incubation of PrP106-126 with Ru
complexes for 10 h, whereas other complexes exhibited generic fluo-
rescence quenching to the peptide (Fig. S6). In consideration of the
inner filter effect, the ThT fluorescence intensity was disturbed a little
bit by the aromatic-ring containing Ru-complexes (Fig. S7), but the
disturbance was negligible compared with strong fluorescence decrease
after the incubation of PrP106-126 with the aromatic-ring containing
Ru-complexes. The concentration-dependent inhibitive effects of differ-
ent ruthenium complexes on PrP106-126 aggregation analyzed by ThT
fluorescence are shown in Fig. S8. Table 1 shows the IC50 value for
different Ru complexes, and Ru(bpy)Cl4 displayed the most effective
inhibition with its IC50 at 6 μM.
7
4. Discussion
PrP106-126 maintains the amyloidogenic and neurotoxic properties
of the entire pathological PrPSc and is usually used as an acceptable
model to study the mechanism of prion diseases [13]. Our previous
study demonstrated that Pt-, Pd- and Au-based complexes altered the
solution properties of PrP106-126 and significantly inhibited prion
fragment aggregation [27,28]. However, the molecular configuration
of selected compounds in the abovementioned types of complexes is
limited. In this study, hexacoordinated Ru complexes were employed
in view of varied ligand configurations to study their inhibition abilities
to PrP106-126.
4.1. Different binding modes of Ru complexes to PrP106-126
Ru-based complexes can reportedly bind to several peptide frag-
ments or proteins [42,43]. In this study, six hexacoordinated Ru
complexes were used to bind the prion neuropeptide PrP106-126 and
change the physicochemical properties of PrP106-126. In the ESI-MS
experiments, Ru complexes were used to interact with the peptide
after diluting the stock solution. Many other reports proposed the hydro-
lysis property of ruthenium complexes [29,44]. In this study, some
compounds suffered the process, yet they bound to PrP106-126 at
lower concentration (see ESI-MS data for different products), which indi-
cates that the complexes have better binding affinity to PrP106-126. The
ESI-MS spectra demonstrated that Ru complexes can interact with
PrP106-126 in different binding modes as well. The [(DMSO)2H]
[trans-Ru(DMSO)2Cl4] compound bound with PrP106-126 at 1:1 and
1:2 ratios with one or two detached DMSO ligands. The formation of
three adducts is presumably due to the solvolysis of the complex.
NAMI-A has demonstrated its anti-metastatic activity in pre-clinical
studies [35]. It can bind to surface histidyl imidazoles on albumin in the
blood as NAMI [42,43]. In our study, NAMI-A bound to PrP106-126 at
1:1 and 2:1 ratios, with the release of the outer imidazole ring. The MS
3.6. PrP106-126 aggregation images by TEM and AFM
TEM and AFM were performed to ascertain whether Ru complexes
affect peptide aggregation and fibril formation. The aggregates formed
by PrP106-126 displayed a quasi-spherical structure (Fig. 6A), which po-
tentially implies a strong aggregation state, after 24 h of incubation at
37 °C [27,28]. However, the TEM and AFM micrographs of PrP106-126
in the presence of Ru complexes showed that the aggregation was
reversed to different extents. The morphology revealed a distinct shape
after the incubation of Ru complexes with the aggregates for 24 h.
NAMI-A influenced the aggregates of PrP106-126 more effec-
tively than that of [(DMSO)2H][trans-Ru(DMSO)2Cl4] (Fig. 7).
Negatively stained fibrils, in the presence of cis-RuCl2(DMSO)4
and trans-RuCl2(DMSO)4, were branched and weaker than those
of PrP106-126 (Fig. 6). The fibrils for cis-(bpy)2RuCl2 and Ru(bpy)
Cl4 were sparsely scattered, which indicated the more intense
inhibiting ability of aromatic ring-containing complexes (Fig. 8).
Fig. 4. Circular dichroism spectrum of PrP106-126 in the presence and absence (black
line) of 1.0 equiv of ruthenium complexes. The PrP106-126 concentration in solution
was 0.1 mM.
Fig. 5. Evaluation of the ability of the ruthenium complexes to inhibit PrP106-126
aggregation as measured by ThT fluorescence.

X. Wang et al. / Journal of Inorganic Biochemistry 128 (2013) 1–10
7
Fig. 6. TEM and AFM images of PrP106-126 fibrils in the absence (A, D) and presence of cis-RuCl2(DMSO)4 (B, E), and trans-RuCl2(DMSO)4 (C, F). The scale bar is 200 nm for TEM,
and 1.5 μm for AFM.
data illustrated the absence of free metal binding site for NAMI-A binding
to PrP106-126. This interesting phenomenon displayed a distinctive
binding pattern from that of other ruthenium complexes. May non-
bonded interactions such as hydrophobic interaction and electrostatic
interaction contribute to the binding force between NAMI-A and the pep-
tide. After binding with the compound, 2D 1H–1H TOCSY assay showed
the resonance disturbance from His111, Met112, as well as Lys110. More-
over, the imidazole ring of His111 may play a π–π interaction role and the
polar side chain of Lys110 may contribute to the electrostatic interaction.
In addition, MS spectra indicated that cis-(bpy)2RuCl2 and Ru(bpy)Cl4
compounds bound to the peptide in two forms, e.g., only the Ru ion and
relevant fractional group bound to the peptide. A common point of the
two complexes is that the adduct scaffold of the peptide–metal complex
reserved the aromatic ring. For the linear configuration of Ru complexes,
cis-RuCl2(DMSO)4 and trans-RuCl2(DMSO)4 bound to PrP106-126 in
different ways with varied numbers of DMSO ligand that slipped off.
Comparing the relative peak intensities in the MS spectra, [(DMSO)2H]
[trans-Ru(DMSO)2Cl4] showed the strongest binding affinity among the
studied complexes. This strong binding affinity reveals that the confor-
mation of the complexes contributed to the binding affinity with the
peptide.
The 1H–1H TOCSY spectrum demonstrated a probable binding mode
between Ru complex and PrP106-126. NAMI-A binds to the peptide
through non-bonded interaction, because the product [Ru(DMSO)(Im)
Cl4] from ESI-MS shows no direct metal coordination to the peptide.
The 2D NMR results displayed that the spin system of Lys110, His111,
Fig. 7. TEM and AFM images of PrP106-126 fibrils in the absence (A, D) and presence of [(DMSO)2H][trans-Ru(DMSO)2Cl4] (B, E), and NAMI-A (C, F). The scale bar is 200 nm for
TEM, and 1.5 μm for AFM.

8
X. Wang et al. / Journal of Inorganic Biochemistry 128 (2013) 1–10
Fig. 8. TEM and AFM images of PrP106-126 fibrils in the absence (A, D) and presence of cis-(bpy)2RuCl2 (B, E), and Ru(bpy)Cl4 (C, F). The scale bar is 200 nm for TEM, and 1.5 μm for AFM.
and Met112 was disturbed either in the peak shape or in the chemical
shift. A presumption was made that the imidazole ring of complex
NAMI-A interacts with the peptide at the N-terminal region, causing the
change of the peptide biophysical properties. When it comes to the
complex cis-(bpy)2RuCl2, only the chemical shift change of His111 CδHs
was observed, indicating the probable metal binding site to the peptide
according to the ESI-MS result.
Different binding modes were also reflected by the CD spectra. The
coordination of Ru complexes to PrP106-126 changed the secondary
structure of the peptide at different degrees. Our conformational
study revealed that cis-RuCl2(DMSO)4 induced an opposite effect on
the conformation of PrP106-126, unlike other complexes. This opposite
effect may be a result of the different binding modes of this complex
with PrP106-126. Furthermore, the influence of trans-RuCl2(DMSO)4
and NAMI-A on the conformation of PrP106-126 was not obvious.
However, the binding of the two compounds with PrP106-126 is solid
from ESI-MS. Even though notable conformational change is not
observed, the inhibitory effects of the two compounds on peptide
aggregation are evident. This indicates no direct relationship between
the binding mode and the inhibition of peptide aggregation for the
two compounds. Maybe they affect some factors relative to fibril
formation but not visible secondary structure change, such as spatial
obstacle and β-sheet orientation.
with metal inhibitors of Aβ protein, the mentioned compounds have a
better IC50 value under 10 μM [47]. All the results corresponded to
the observed AFM and TEM images.
The peptide aggregation resulted from Ru complexes were also
confirmed by TEM and AFM images, in which method no ThT binding
competition was worried. The morphology of the aggregates formed
by the prion peptide after incubation for 24 h is large in scale and
volume. However, after treating with the Ru complexes, the aggregated
neuropeptide changed into scattered and filamentous aggregates. All Ru
complexes reversed the aggregation behavior of PrP106-126, despite
varied inhibitory effects for different complexes. NAMI-A displayed a
better inhibitory ability than that of [(DMSO)2H][trans-Ru(DMSO)2Cl4].
The only difference between the two complexes was their ligands.
NAMI-A bound to PrP106-126 in the form of [Ru(DMSO)(Im)Cl]. The
adduct formed by PrP106-126 and NAMI-A reserved the aromatic ring
imidazole, which may enlighten us with the proposal that the aromatic
ring interacts with the peptide side chain such as imidazole group of
His111. Ultimately, the conformation of aggregated peptide was affect-
ed and simultaneously the change of morphology of aggregates.
Former research proposed that the protofibril was an inevitable
step during the process of aging fibrous peptides such as Aβ and prion
peptides [49,50]. The observed protofibril after treating aggregated
prion peptides with NAMI-A indicated that the complex reversed the
fibrous state of PrP106-126. It has also been reported that oligomers
formed by prion peptides or Aβ peptides maintained the structure of
spherical like aggregates that can be detected under AFM or TEM instru-
ments [51]. Although the morphology of aggregates was changed after
treating cis-(bpy)2RuCl2 and Ru(bpy)Cl4 with the aggregated fibril,
clearance of the oligomers still needs further efforts.
4.2. Effective inhibition of Ru complexes on PrP106-126 aggregation
The binding interactions of Ru complexes with PrP106-126 were
well identified by ESI-MS spectra. In addition, Ru complexes significant-
ly affected the formation of the amyloid structures of PrP106-126
through ThT assay. Although minor inner filter effect did exist in the
ThT assay, especially to those aromatic compounds, the main results
were not impacted. Further, several earlier studies have employed the
ThT assay to assess the effect of inhibitors on the fibril formation of
Aβ protein [45–48]. And the IC50 values were also determined to com-
pare the inhibition of different ruthenium complexes on preformed
PrP106-126 fibrils. The IC50 results revealed that aromatic-ring con-
taining complexes such as NAMI-A, Ru(bpy)Cl4, and cis-(bpy)2RuCl2
exerted a stronger inhibitive effect on the fibrils formation. Compared
However, cis-(bpy)2RuCl2 and Ru(bpy)Cl4 demonstrated better
inhibitory effects than that of cis- and trans-RuCl2(DMSO)4, which
may be consequent to their aromatic pyridine ligand. Hence, potential
π–π interaction between the aromatic compounds and PrP106-126 is
supposed, aside from metal coordination and hydrophobic interac-
tion. The aggregates may be composed of monomers, oligomers and
fibrils. Particularly, the oligomers are cytotoxic. Therefore, more Ru
compounds need to be studied for better deconstruction of oligomers
and elimination of fibrils.

X. Wang et al. / Journal of Inorganic Biochemistry 128 (2013) 1–10
9
4.3. Comparison of Ru complexes binding affinity with inhibitory effect
on PrP106-126
Chemical Resource Engineering of Beijing University of Chemical Tech-
nology (CRF-2012-C-102).
The MS spectra commonly display several adducts after the incu-
bation of the complexes with PrP106-126. Interestingly, the Ru
complexes, except for NAMI-A, bound to PrP106-126 in the metal
coordination mode because Ru ion-contained fragment was always
observed in the MS spectra of the peptide-complex mixing solution.
The MS data suggested that the binding affinity of Ru complexes is
not better than that of Pd complexes [28], whereas the inhibitory
effect of ad hoc complexes is greater in NAMI-A, Ru(bpy)Cl4, and
cis-(bpy)2RuCl2. The MS data collected after the incubation of the
three complexes with PrP106-126 reflected relatively weak peaks,
indicating that their binding affinity was not stronger compared
with [(DMSO)2H][trans-Ru(DMSO)2Cl4]. Interestingly, the three
complexes displayed intensive inhibitory effect on the aggregation
property of the peptide in the TEM and AFM assays. Hence, no direct
relationship exists between binding affinity and inhibitory ability.
The oligomers formed by PrP106-126 are stacked by the interface
of hydrophobic C-terminal residues 113–126, which may increase
the rate of fibril growth [9,19]. The difference between binding affin-
ity and inhibitory effect may also arise from non-bonded interactions
including hydrophobic, electrostatic, and π–π interaction, aside from
the molecular configuration contribution [52]. Using NAMI-A as an
example, the MS data indicated that the complex bound to the pep-
tide in the form of [Ru(DMSO)(Im)Cl4], which denoted no remaining
free metal binding site for the complex to bind with PrP106-126,
which was indicated by NMR as well. The hydrophobic interaction
between NAMI-A and PrP106-126 interfered with the interface be-
tween PrP106-126 molecules, inhibiting the aggregation of the peptide.
Therefore, the disaggregation of PrP106-126 does not necessarily rely
on metal coordination, specifically, if intense non-bonded interaction
can be built between the neuropeptide and the metal complexes.
In summary, the study on the interaction of Ru complexes with
PrP106-126 indicated that Ru complexes can bind to PrP106-126 in
different binding modes, either through metal coordination or through
non-bonded interactions. The complexes with different molecular
configurations inhibited the aggregation of PrP106-126 at different de-
grees. Moreover, aromatic ring-containing Ru complexes showed better
inhibitory effects. In addition, the binding affinity does not determine
the inhibitory ability of Ru complexes. Ru complexes are compounds
that have relatively low cellular toxicity compared with platinum,
palladium, and gold complexes. Thus, our study paves the way for effec-
tive pharmaceutical products against prion diseases, and further inves-
tigation should be carried out to discover potential metallodrugs based
on ruthenium complexes.
Appendix A. Supplementary data
Supplementary data to this article can be found online at http://
dx.doi.org/10.1016/j.jinorgbio.2013.07.009.
References
[1] S.B. Prusiner, Science 278 (1997) 245–251.
[2] S.B. Prusiner, Proc. Natl. Acad. Sci. U. S. A. 95 (1998) 13363–13383.
[3] C. Korth, B. Stierli, P. Streit, M. Moser, O. Schaller, R. Fischer, W. Schulz-Schaeffer, H.
Kretzschmar, A. Raeber, U. Braun, F. Ehrensperger, S. Hornemann, R. Glockshuber, R.
Riek, M. Billeter, K. Wuthrich, B. Oesch, Nature 390 (1997) 74–77.
[4] J.F. Diedrich, P.E. Bendheim, Y.S. Kim, R.I. Carp, A.T. Haase, Proc. Natl. Acad. Sci. U.
S. A. 88 (1991) 375–379.
[5] M.F. Jobling, X. Huang, L.R. Stewart, K.J. Barnham, C. Curtain, I. Volitakis, M.
Perugini, A.R. White, R.A. Cherny, C.L. Masters, C.J. Barrow, S.J. Collins, A.I. Bush,
R. Cappai, Biochemistry 40 (2001) 8073–8084.
[6] E. Gaggelli, F. Bernardi, E. Molteni, R. Pogni, D. Valensin, G. Valensin, M. Remelli,
M. Luczkowski, H. Kozlowski, J. Am. Chem. Soc. 127 (2004) 996–1006.
[7] H.Z. Soprana, L.C. Souza, V. Debbas, F.R.M. Laurindo, Exp. Toxicol. Pathol. 63 (2011)
229–236.
[8] P. Stańczak, H. Kozłowski, Biochem. Biophys. Res. Commun. 352 (2007) 198–202.
[9] K.M. Pan, M. Baldwin, J. Nguyen, M. Gasset, A. Serban, D. Groth, I. Mehlhorn, Z.
Huang, R.J. Fletterick, F.E. Cohen, Proc. Natl. Acad. Sci. U. S. A. 90 (1993)
10962–10966.
[10] M.F. Jobling, L.R. Stewart, A.R. White, C. McLean, A. Friedhuber, F. Maher, K. Beyreuther,
C.L. Masters, C.J. Barrow, S.J. Collins, R. Cappai, J. Neurochem. 73 (1999) 1557–1565.
[11] F. Tagliavini, F. Prelli, L. Verga, G. Giaccone, R. Sarma, P. Gorevic, B. Ghetti, F. Passerini,
E. Ghibaudi, G. Forloni, Proc. Natl. Acad. Sci. U. S. A. 90 (1993) 9678–9682.
[12] D.R. Brown, B. Schmidt, H.A. Kretzschmar, Nature 380 (1996) 345–347.
[13] J. Chabry, C. Ratsimanohatra, I. Sponne, P.P. Elena, J.P. Vincent, T. Pillot, J. Neurosci.
23 (2003) 462–469.
[14] D.R. Brown, Biochem. J. 352 (2000) 511–518.
[15] A. Corsaro, S. Thellung, V. Villa, M. Nizzari, T. Florio, Int. J. Mol. Sci. 13 (2012) 8648–8669.
[16] D.L. Rymer, T.A. Good, J. Neurochem. 75 (2000) 2536–2545.
[17] C. Selvaggini, L. Degioia, L. Cantu, E. Ghibaudi, L. Diomede, F. Passerini, G. Forloni, O.
Bugiani, F. Tagliavini, M. Salmona, Biochem. Biophys. Res. Commun. 194 (1993)
1380–1386.
[18] F. Chiti, C.M. Dobson, Annu. Rev. Biochem. 75 (2006) 333–366.
[19] M. Grabenauer, C. Wu, P. Soto, J.E. Shea, M.T. Bowers, J. Am. Chem. Soc. 132 (2009)
532–539.
[20] P. Walsh, P. Neudecker, S. Sharpe, J. Am. Chem. Soc. 132 (2010) 7684–7695.
[21] R. Liu, R. Su, M. Liang, R. Huang, M. Wang, W. Qi, Z. He, Curr. Med. Chem. 19
(2012) 4157–4174.
[22] Y. Liu, Y. Han, J. Song, Y. Wang, Y. Jing, Q. Shi, C. Tian, Z. Wang, C. Li, J. Han, X. Dong,
Int. J. Biochem. Cell Biol. 43 (2011) 768–774.
[23] I. Turi, C. Kállay, D. Szikszai, G. Pappalardo, G. Di Natale, P. De Bona, E. Rizzarelli, I.
Sóvágó, J. Inorg. Biochem. 104 (2010) 885–891.
[24] K.J. Barnham, V.B. Kenche, G.D. Ciccotosto, D.P. Smith, D.J. Tew, X. Liu, K. Perez,
G.A. Cranston, T.J. Johanssen, I. Volitakis, A.I. Bush, C.L. Masters, A.R. White, J.P.
Smith, R.A. Cherny, R. Cappai, Proc. Natl. Acad. Sci. U. S. A. 105 (2008) 6813–6818.
[25] M.P. Cuajungco, C.J. Frederickson, A.I. Bush, Subcell. Biochem. 38 (2005) 235–254.
[26] T. Storr, M. Merkel, G.X. Song-Zhao, L.E. Scott, D.E. Green, M.L. Bowen, K.H. Thompson,
B.O. Patrick, H.J. Schugar, C. Orvig, J. Am. Chem. Soc. 129 (2007) 7453–7463.
[27] Y. Wang, J. Xu, L. Wang, B. Zhang, W. Du, Chem. Eur. J. 16 (2010) 13339–13342.
[28] Y. Wang, L. Feng, B. Zhang, X. Wang, C. Huang, Y. Li, W. Du, Inorg. Chem. 50 (2011)
4340–4348.
Abbreviations
[29] M.J. Clarke, Coord. Chem. Rev. 236 (2003) 209–233.
[30] C.G. Hartinger, M.A. Jakupec, S. Zorbas-Seifried, M. Groessl, A. Egger, W. Berger, H.
Zorbas, P.J. Dyson, B.K. Keppler, Chem. Biodivers. 5 (2008) 2140–2155.
[31] M.J. Clarke, F. Zhu, D.R. Frasca, Chem. Rev. 99 (1999) 2511–2533.
[32] H. Chen, J.A. Parkinson, S. Parsons, R.A. Coxall, R.O. Gould, P.J. Sadler, J. Am. Chem.
Soc. 124 (2002) 3064–3082.
[33] D. Valensin, P. Anzini, E. Gaggelli, N. Gaggelli, G. Tamasi, R. Cini, C. Gabbiani, E.
Michelucci, L. Messori, H. Kozlowski, G. Valensin, Inorg. Chem. 49 (2010) 4720–4722.
[34] N.P. Cook, V. Torres, D. Jain, A.A. Marti, J. Am. Chem. Soc. 133 (2011) 11121–11123.
[35] I.P. Evans, A. Spencer, G. Wilkinson, J. Chem. Soc. Dalton Trans. (1973) 204–209.
[36] B.R. James, E. Ochiai, G.L. Rempel, Inorg. Nucl. Chem. Lett. 7 (1971) 781–784.
[37] E. Alessio, G. Mestroni, G. Nardin, W.M. Attia, M. Calligaris, G. Sava, S. Zorzet,
Inorg. Chem. 27 (1988) 4099–4106.
PrPC
Cellular isoform of prion protein
Abnormal disease-causing isoform of prion protein
PrPSc
PrP106-126 Prion peptide 106-KTNMKHMAGAAAAGAVVGGLG-126
Aβ
TOCSY
NAMI-A [Trans-RuCl4(DMSO)(Im)][ImH]
ThT Thioflavin-T
Amyloid beta protein
Total correlation spectroscopy
[38] E. Alessio, G. Balducci, M. Calligaris, G. Costa, W.M. Attia, G. Mestroni, Inorg. Chem.
30 (1991) 609–618.
[39] G. Mestroni, E. Alessio, G. Sava, WO Patent, WO/1998/000, 4311998.
[40] F.P. Dwyer, H.A. Goodwin, E.C. Gyarfas, Aust. J. Chem. 16 (1962) 42–50.
[41] B.P. Sullivan, D.J. Salmon, T.J. Meyer, Inorg. Chem. 17 (1978) 3334–3341.
[42] M.I. Webb, R.A. Chard, Y.M. Al-Jobory, M.R. Jones, E.W.Y. Wong, C.J. Walsby, Inorg.
Chem. 51 (2011) 954–966.
Acknowledgments
This work was supported by the National Natural Science Founda-
tion of China (No. 21271185), the National Basic Research Program
(No. 2011CB808503), and the Open Fund of State Key Laboratory of
[43] M.I. Webb, C.J. Walsby, Dalton Trans. 40 (2011) 1322–1331.
[44] I. Kostova, Curr. Med. Chem. 13 (2006) 1085–1107.

10
X. Wang et al. / Journal of Inorganic Biochemistry 128 (2013) 1–10
[45] S.R. Byeon, J.H. Lee, J.H. Sohn, D.C. Kim, K.J. Shin, K.H. Yoo, I.M. Jung, W.K. Lee, D.J.
Kim, Bioorg. Med. Chem. Lett. 17 (2007) 1466–1470.
[46] J.H. Byun, H.Y. Kima, Y.S. Kima, I.M. Jung, D.J. Kim, W.K. Lee, K.H. Yoo, Bioorg. Med.
Chem. Lett. 18 (2008) 5591–5593.
[47] V.P.S. Chauhan, I. Ray, A. Chauhan, J. Wegiel, H.M. Wisniewski, Neurol. Res. 7
(1997) 805–809.
[48] B.J. Blanchard, A. Chen, L.M. Rozeboom, K.A. Stafford, P. Weigele, V.M. Ingram,
Proc. Natl. Acad. Sci. U. S. A. 101 (2004) 14326–14332.
[49] B.A. Chromy, R.J. Nowak, M.P. Lambert, K.L. Viola, L. Chang, P.T. Velasco, B.W.
Jones, S.J. Fernandez, P.N. Lacor, P. Horowitz, C.E. Finch, G.A. Krafft, W.L. Klein,
Biochemistry 42 (2003) 12749–12760.
[50] M. Arimon, I. Díez-Pérez, M.J. Kogan, N. Durany, E. Giralt, F. Sanz, X.
Fernàndez-Busquets, FASEB J. 19 (2005) 1344–1346.
[51] J. Laurén, D.A. Gimbel, H.B. Nygaard, J.W. Gilbert, S.M. Strittmatter, Nature 457 (2009)
1128–1132.
[52] G. Ma, F. Huang, X. Pu, L. Jia, T. Jiang, L. Li, Y. Liu, Chem. Eur. J. 17 (2011) 11657–11666.