Evaluation of ferritin and transferrin binding to tau protein

    Evaluation of Ferritin and Transferrin Binding to Tau Protein Anna Jahshan, Jose O. Esteves-Villanueva, Sanela Martic-Milne PII: DOI: Reference:

S0162-0134(16)30185-4 doi: 10.1016/j.jinorgbio.2016.06.022 JIB 10024

To appear in:

Journal of Inorganic Biochemistry

Received date: Revised date: Accepted date:

25 January 2016 13 June 2016 16 June 2016

Please cite this article as: Anna Jahshan, Jose O. Esteves-Villanueva, Sanela MarticMilne, Evaluation of Ferritin and Transferrin Binding to Tau Protein, Journal of Inorganic Biochemistry (2016), doi: 10.1016/j.jinorgbio.2016.06.022

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Evaluation of Ferritin and Transferrin Binding to Tau Protein

PT

Anna Jahshan, Jose O. Esteves-Villanueva and Sanela Martic-Milne* Department of Chemistry, Oakland University, 2200 North Squirrel Road, Rochester, MI, 48309,

SC

RI

USA

NU

Corresponding author: Sanela Martic-Milne, e-mail: [email protected], Tel: 248-370-3088, Fax: 248-370-2321

MA

ABSTRACT

D

Tau protein is a neurodegeneration biomarker. Due to the high concentration of metal ions in the

TE

brain, the metallation of tau proteins and their catalytic role in reactive oxygen formation have been identified as a major biochemical pathway of neurodegeneration. High levels of iron ions

AC CE P

have been detected in Alzheimer’s disease brains. One of biological sources of iron ions are ironrich proteins, such as transferrin or ferritin. However, the interactions between tau and these metallo-proteins have not been fully characterized. Here, the interactions between the longest form of full-length tau protein (tau441) with iron-rich proteins were detected using electrochemical impedance spectroscopy. Tau441 was immobilized on Au surface, via Nterminal (N-tau-Au film) or Cys-residues (Cys-tau-Au film), and the charge-transfer resistance, Rct, was monitored prior and post ferritin or transferrin binding. Significant increase in Rct was observed post transferrin binding above 50 µg mL-1, but not ferritin regardless of concentration with N-tau-Au film. Additionally, the electrochemical trend was linear with respect to transferrin concentration. Electrochemical data indicated low binding by ferritin to N-tau-Au or Cys-tau-Au films. The interaction of apotransferrin or apoferritin with tau films was also evaluated. 1

ACCEPTED MANUSCRIPT Electrochemical data may be pointing to the difference in protein binding modes by transferrin compared to ferritin as well as to importance of metal ions in protein-protein interactions.

PT

KEYWORDS

SC

RI

Ferritin, transferrin, tau protein, neurodegeneration, electrochemical impedance spectroscopy

NU

1. INTRODUCTION

MA

Tau protein is naturally occurring neuronal structural protein which maintains the microtubule structure and facilitates nutrient uptake [1]. Upon post-translational modifications tau protein

D

detaches from microtubules, aggregates and leads to neuronal death and cell-to-cell toxicity [2-

TE

4]. Additional mechanism of neurodegeneration is the reactive oxygen species (ROS) formation

AC CE P

and oxidative damage [5-6]. Metal ions may catalyze ROS generation, making the metal homeostasis an important parameter in neurodegeneration. The millimolar concentrations of metal ions, including iron have been detected during post-mortem analysis of Alzheimer’s disease (AD) brains [7-9]. The iron-rich proteins, including transferrin and ferritin, are present in the brain, neurons and their levels are also modulated in AD [10-15]. Iron accumulates in different parts of the brain, including neurons, and has been linked to disease progression [1618]. The decrease in transferrin and increase in ferritin levels in AD brains were reported [10,19]. Free iron is highly reactive and can catalyze formation of ROS giving rise to cell damage [20]. In the brain, extracellular iron binds transferrin and gets intracellular by transferrin receptor-based endocytosis. Transferrin is an iron transporting protein, present in plasma, with high binding affinity to two iron ions. Ferritin protein is an iron storing protein, present within cells, with 24subunits hosting up to 4,500 iron atoms encapsulated in the ferritin cavity as ferrihydrite mineral 2

ACCEPTED MANUSCRIPT (Fe2O3·H2O). In the brain, cytosolic iron is stored in ferritin, but can be transferred to mitochondria among other cellular compartments. Neurons express mostly H-ferritin, which

PT

represents the H-heavy chain of ferritin with ferroxidase center that catalyzes oxidation of Fe(II)

RI

into Fe(III) under stress [17]. The coordination of tau protein to metal ions may contribute to neurodegeneration, however the mechanism of metalation and its impact on ROS formation are

SC

not well understood.

NU

Tau reportedly binds a variety of metal ions: Al(III), Fe(III), Cu(II), Cu(I), Zn(II), Cd(II) and

MA

Hg(II) [21-30]. The coordination of tau peptide fragments to metal ions has been well characterized by biophysical methods, including nuclear magnetic resonance (NMR), mass-

D

spectrometry (MS), and circular dichroism (CD). R1-R4 repeat peptides of tau coordinate Cu(II)

TE

via His residues and backbone amides, while the 287-293 and 310-324 peptide repeats coordinate Cu(II) center via Cys residues [30]. The coordination of the full length tau protein has

AC CE P

also been evaluated with some metal ions [7]. The studies on binding of tau to ferritin or transferrin are scarce. The binding of tau filaments to ferritin has been evaluated by biophysical and biochemical methods, including transmission electron microscopy and Western blotting [31]. The iron ions were found to increase tau filament formation in the presence of ascorbate. The binding of transferrin to tau has been evaluated by biochemical techniques as well [32-34]. Fe(II) and Fe(III) metal ions were found associated with plaques and neurofibrillary tangles, made up of tau, and neurons in AD brains [35]. Monitoring tau protein (nonphosphorylated and phosphorylated) binding to ferritin and transferrin has not been evaluated. However, the evidence of ferritin colocalization with tau filaments in neurodegenerative brains was reported [31], indicating that ferritin interaction with tau protein is relevant especially given the ferritin location intracellular. 3

ACCEPTED MANUSCRIPT Additionally, tau protein has been found to also exist extracellular when it leaks out of the dying neuronal cell [36,37]. In this context, the interactions between the iron-rich proteins

PT

and tau, both intra- and extra-cellular, are of importance. Since tau protein has been also detected

RI

in blood, due to the recent interest in blood biomarkers of neurodegeneration, the interactions between tau protein and transferrin in plasma is also of relevance [38]. In addition, it is unclear

SC

if these protein-protein complexes promote ROS formation.

NU

Here, we describe the use of electrochemical method, electrochemical impedance spectroscopy

MA

(EIS), for monitoring transferrin and ferritin binding to nonphosphorylated (Tau-Au) tau film on Au surface. The effects of protein orientation and concentration on binding to tau films were

D

investigated and the protein surface characterized by X-ray photoelectron spectroscopy (XPS)

TE

and contact angle. The role of metal ions on protein-protein interactions was evaluated with apoferritin and apotransferrin. In addition, catalytic abilities of the protein films were also

AC CE P

explored with ascorbic acid.

2. RESULTS AND DISCUSSION For measuring ferritin or transferrin binding, the tau protein was immobilized on the Au surface first. The electrochemical signature peak was detected for tau-Au film prior to binding ferritin or transferrin, as well as post. 2.1. Preparation of tau-Au film. Tau-Au film was prepared as depicted in Fig. 1. Briefly, bare Au was incubated in Lipoic acid (a) followed by activation with 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDC)/Nhydroxysuccinimide (NHS) (b). Subsequently, tau441 protein (c) was covalently attached via predominantly NH2 group of N-terminal. The surface immobilization of protein via terminal NH2 4

ACCEPTED MANUSCRIPT group onto Au surface was previously reported [39, 40]. While immobilization may be taking place predominantly via N-terminal of protein, the attachment of tau protein trough Lys residues

PT

may also contribute to surface modification. Next, electrodes were blocked with ethanolamine

AC CE P

TE

D

MA

NU

SC

RI

(d) to remove any –NHS sites and backfilled with hexanethiol (e) to block any free Au surfaces.

Fig. 1. Schematic illustration of tau-Au fabrication: a) Lipoic acid, b) EDC/NHS, c) tau, d) ethanolamine, and e) hexanethiol.

2.2. Electrochemical characterization of tau-Au films. The surface modification steps involved in fabrication of Tau-Au film were monitored using cyclic voltammetry (CV) and EIS. Fig. 2A depicts CVs of stepwise modification of Au surface toward preparation of Tau-Au film. The high current associated with [Fe(CN)6]3-/4- was observed for bare Au (a). Upon immobilizing Lipoic acid (b) on the surface, the current decreased and the separation between the oxidation and reduction peaks increased. This trend was associated with successful surface blocking. The increase in current after tau step may be due to the removal of

5

ACCEPTED MANUSCRIPT nonspecifically adsorbed Lipoic acid layers. Further blocking with ethanolamine (e) and backfilling (f) with hexanethiol gave rise to the final Tau-Au film.

PT

The surface modification was also monitored by EIS. EIS is commonly used technique to

RI

monitor protein adsorption and behavior on metallic surfaces in addition to surface modification

SC

[41]. Nyquist plots of stepwise surface modifications are shown in Fig. 2B. The imaginary impedance, Z’’, was plotted against the real impedance, Z’. The surface blocking was associated

NU

with the increase in the semi-circle portion of the EIS curve. The greatest semi-circle, i.e. charge-

MA

transfer resistance, Rct, was observed after Lipoic acid (b) modification. The dramatic decrease in resistance after EDC/NHS (c) and tau (d) steps may be due to removal of nonspecifically

AC CE P

TE

D

adsorbed Lipoic acid.

Fig. 2. (A) CVs and (B) Nyquist plots of step-wise surface modification steps: a) bare Au, b) lipoic acid, c) EDC/NHS, d) tau, e) ethanolamine, and f) hexanethiol.

2.3. Electrochemical binding studies 6

ACCEPTED MANUSCRIPT Following tau-Au film preparation and characterization, films were exposed to ferritin or transferrin solution in phosphate buffer, pH 7.0. This pH was chosen because it closely mimics

PT

the acidity in the neuronal environment. EIS was measured before and after binding event in the

RI

presence of [Fe(CN)6]3-/4- as a solution redox probe. Nyquist plot of tau-Au (a) is presented in Fig. 3 and it was characterized by a semi-circle in addition to the straight line. All data was fitted

SC

to the equivalent circuit (Fig. 3 inset). The solution resistance, Rs, was associated with the bulk

NU

electrolyte solution, electrode and current leads. The circuit included two time constants in series. The first time constant included the constant phase element, CPE, and the charge-transfer

MA

resistance, R1ct, which was due to resistance of protein film. CPE constant is related to capacitive behavior, and an electrochemical double layer capacitance. The R1ct represents the first semi-

D

circle in the high frequency indicating interfacial resistance of the solution redox probe,

TE

Fe(CN)6]3-/4-, as it moves through the protein film towards the electrode surface. The second time

AC CE P

constant at low frequency produced smaller onset of semi-circle rather than the Warburg constant, at 45˚ angle. Hence, the CPE in parallel with the R2ct may be ascribed to capacitance associated with the diffusive pseudocapacitance while R2ct may be associated with the resistance to mass transfer. For quantitative measurements the Rct value was used to evaluate binding of transferrin or ferritin to tau-Au. Basically, the diameter of the first semi-circle in the Nyquist plot is related to Rct value. After ferritin binding (b) a small increase in semi-circle was observed, i.e. small increase in Rct. However, after transferrin binding (c), the semi-circle portion dramatically increased indicating binding of transferrin to tau-Au film.

7

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

TE

D

([ferritin]=[transferrin]= 100 µg mL-1).

MA

Fig. 3. Nyquist plots of tau-Au film before (a) and after binding to ferritin (b), or transferrin (c)

AC CE P

The diameter of the semi-circle represents the charge transfer resistance, Rct, value which can be derived by fitting the experimental data to the equivalent circuit (inset Fig 3). The Rct values (representing the first semi-circle in the Nyquist plots) were compared before and after binding. The Rct factor was derived by dividing the Rct-after by Rct-before. Fig. 4 shows Rct factor as a function of ferritin and transferrin concentrations: 0, 0.5, 5.0, 25, 50, 75, 100, 200 µg mL-1 upon binding to Tau-Au film. Notably, Rct factor was unchanged independent of ferritin concentration which may indicate low binding of ferritin to tau-Au film. In contrast, upon increasing transferrin concentration, above 50 µg mL-1 significant increase in Rct factor was observed. After transferrin binding to tau-Au film, the Rct factor increased 11 times, when considering the highest transferrin concentration, indicating a significant binding which can be measured and quantified by EIS.

8

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

TE

D

Fig. 4. Plot of Rct factor of tau-Au as a function of ferritin or transferrin concentrations.

AC CE P

When Tau-Au films were measured by EIS in the absence of [Fe(CN)6]3-/4-, little change in resistance was observed after ferritin or transferrin binding. This indicated that the increase in resistance after transferrin binding may be due to the limited electron-transfer from the solution redox probe to the electrode surface when Tau-Au is being blocked by transferrin. In addition, the two iron centers in transferrin may be buried inside the protein film and inaccessible to communicate with solution redox probe, leading to greater resistance. By contrast when ferritin binds the tau-Au surface, thousands of metal centers may effectively participate in electrontransfer resulting in minimal resistance change, i.e. Rct factor ~1. A direct oxidation/reduction of Fe(III)/Fe(II) redox couple in ferritin adsorbed onto indium tin oxide surface was previously reported [42]. Additionally, the protein-free hydrous ferric oxide was characterized by a similar voltammetric characteristic to that of ferritin on indium tin oxide (ITO) electrodes. The ferritin directly adsorbed onto Au electrode or onto self-assembled monolayers decorated with 9

ACCEPTED MANUSCRIPT carboxylates has shown to undergo electron transfer upon Fe(III) reduction to Fe(II) as well [43]. To clarify the role of electroactivity of iron ions in transferrin or ferritin bound to Tau-Au

PT

modified electrodes, we carried out cyclic voltammetry and square-wave voltammetry. The

RI

cyclic voltammetry was carried in phosphate buffer pH 7.0, in the absence of [Fe(CN)6]3-/4-, using tau-Au films prior and post incubation in ferritin or transferrin. In the -0.5 V to 0.3 V

SC

potential range, no significant current was observed associated with the Fe(II)/Fe(III) redox

NU

couple associated with either ferritin or transferrin. Hence, iron ions of transferrin or ferritin may be assisting in the electron transfer from bulk solution ([Fe(CN)6]3-/4-) to the electrode but are not

MA

the direct source of electrons as evidence by the lack of redox activity of ferritin or transferrin in

D

the present system.

TE

The tau-free Au film was also measured in the presence of ferritin or transferrin and the EIS data collected. The resistance increased in the presence of ferritin or transferrin indicating that both

AC CE P

proteins bind the surface without differentiation. This may be due to non-specific adsorption of ferritin and transferrin onto the surface. To minimize the non-specific adsorption on the surface, the commonly used methods include the hydrophilic diluents such as poly(ethylene)glycolterminated thiol or disulfide, or solution of bovine serum albumin [45,46, 47]. However, in the presence of Tau-Au, the unique increase in resistance, increase in Rct factor, was observed for transferrin over ferritin. This represents a significant differentiation of proteins based on the content of the Tau-Au film. 2.4. X-ray Photoelectron Spectroscpy surface characterization of tau-Au before and after binding

10

ACCEPTED MANUSCRIPT Tau-Au film was characterized by XPS before and after binding to ferritin and transferrin. The tau-Au film was evaluated with respect to sulphur content in order to determine Au-S binding

PT

mode. Fig. 5A shows typical S 2p spectra of tau-Au indicating formation of Au-S. The S 2p1/2

RI

and S 2p3/2 binding energies in the 160-166 eV range were due to the physisorbed thiol and Au-S bond. Additionally, exposure of tau-Au to ferritin (B) or transferrin (C) did not significantly

SC

change the S 2p profile. This is reasonable, since ferritin and transferrin do not bind Au but

AC CE P

TE

D

MA

NU

rather bind to tau-Au film.

Fig. 5. High-resolution XPS spectra of S 2p for (A) tau-Au exposed to (B) ferritin or (C) transferrin.

XPS has been previously used to detect adsorption of a variety of proteins, including ferritin [48]. The iron content was monitored by XPS as well by measuring binding energy associated with Fe 2p. The absorption of ferritin on Au surface was reported previously [49]. In Fig. 6, Fe 2p spectra for tau-Au (a) before and after binding to iron rich proteins are shown. Tau-Au lacks iron peaks as expected. Fig. 6B indicates that ferritin binding to tau-Au is characterized by strong Fe content, unlike transferrin binding (C). This is in contrast to the EIS results which indicated that transferrin binds more than ferritin. However, extremely large iron content in ferritin compared to transferrin may produce iron peaks in XPS despite the lower binding of ferritin. 11

SC

RI

PT

ACCEPTED MANUSCRIPT

NU

Fig. 6. High-resolution XPS spectra of Fe 2p for (A) tau-Au exposed to (B) ferritin or (C)

MA

transferrin. 2.5. Contact angle measurements

D

Typically, a relatively hydrophobic surface would exhibit large contact angle to aqueous solution

TE

but a more hydrophilic surface would decrease the contact angle value [50]. To determine

AC CE P

wettability of the surfaces, contact angle was measured in phosphate buffer, pH 7.0. The bare Au surface had contact angle of 80.5 ± 3.1˚, while Tau-Au was characterized by 62.7 ± 4.7˚, indicating the presence of more hydrophilic and polar groups on the Au surface. Exposure of Tau-Au to ferritin or transferrin solutions produced contact angles of 50.3 ± 1.2˚ and 58.7 ± 7.6˚, respectively. The lower contact angle values may indicate potentially greater polarity of the surface after protein binding. 2.6. Ascorbic acid oxidation To measure the catalytic activity of Tau-Au film, the protein film was immersed in the solution of ascorbic acid and the oxidation current associated with ascorbic acid measured by squarewave voltammetry. Fig. 7A shows current (derived from square-wave voltammograms) for TauAu films which was preincubated in phosphate buffer pH 7.0 (a), transferrin (b) or ferritin (c), 12

ACCEPTED MANUSCRIPT and subsequently measured in ascorbic acid-free solution. As expected no redox activity was observed in the given potential range indicating that ferritin or transferrin, if bound to tau-Au

PT

film were not redox active. In ascorbic acid solution the oxidation of ascorbic acid was expected

RI

at ~0.2 V vs. Ag/AgCl reference electrode [51]. Upon transferrin binding (b), an oxidation peak at 0.23 V was observed which was associated with ascorbic acid oxidation in Fig.7B. Notably,

SC

following ferritin binding to Tau-Au (c), the oxidation peak was dramatically reduced. The ease

NU

of ascorbic acid oxidation after transferrin binding to Tau-Au film may indicate the catalytic properties of iron in the protein film. The iron centers in transferrin may be indirectly aiding in

AC CE P

TE

D

MA

solution electron transfer from ascorbic acid to the electrode.

Fig. 7. Square-wave voltammograms of Tau-Au films after incubation with phosphate buffer (a), transferrin (b) or ferritin (c) in ascorbic acid-free solution (A) and in ascorbic acid solution (B) ([ascorbic acid]=1 mM).

Previously, tau protein in aggregated form isolated from the brains of those affected by neurodegeneration was found to contain ferritin [31]. The ferritin association with tau filaments was evaluated by transmission electron microscopy and immunochemistry by using anti-ferritin 13

ACCEPTED MANUSCRIPT antibodies. In vitro ferritin particles were directly bound to the tau filaments. In solution, the binding mode of nonphosphorylated tau protein to iron ions was evaluated by iron-Sepharose

PT

columns and extensive binding was ascribed to N-terminal (HQDQEGDTD (40-48 amino acid))

RI

and C-terminal of tau rather than the R repeats [31]. Other studies reported that ferritin binding proteins exhibited similar peptide binding sequence HNLGHGHK(H)ERDQGHG [40].

SC

However, the binding constants for these sites have not been reported. Notably, Heme

NU

oxygenase-, enzyme that degrades heme, was found associated with tau filaments in neurodegenerative pathology by using immunoreactivity [52]. These studies point to the

MA

interaction between iron-containing proteins and tau via protein-protein interaction.

D

In this study, the longest tau-441 isoform does not contain either of the above mentioned

TE

sequences. However, since tau protein was immobilized to Au electrode predominantly via Nterminal, the accessibility of protein domains for binding to ferritin or transferrin may be limited.

AC CE P

From the electrochemical data it is unclear which is the binding site. Assuming that the covalent linkage of tau to Au surface was predominantly via N-terminal, its accessibility for binding to ferritin or transferrin has been greatly reduced. By contrast, the R repeat domains or C-terminal of tau may be accessible and may likely be binding sites for the iron-rich proteins to tau in this study. If the ferritin requires N-terminal of tau for binding than the weaker binding by ferritin over transferrin may result in the observed electrochemical data which indicate low ferritin binding under present conditions. To further elucidate the role of tau protein orientation of the surface in binding to transferrin or ferritin we have prepared the Cys-tau-Au film. The Cys-tau-Au film was prepared by using inherent Cys291 and Cys322 residues of tau for direct binding to Au electrode, as previously described [53]. The Cys-tau-Au film may exhibit limited accessibility of R2 and R3 domains of 14

ACCEPTED MANUSCRIPT tau, as these are the repeat domains of tau which contain Cys residues used for binding to Au electrode. Figure 8 depicts the Nyquist plots of Cys-tau-Au before (a) and after binding to ferritin

PT

(b) or transferrin (c). Evidently, the greatest increase in resistance was observed after transferrin

RI

binding indicating that transferrin may interact with tau film even when R2 and R3 repeat domains are inaccessible. By contrast, ferritin binding induced decrease in resistance similar to

SC

the control experiment which contained only phosphate buffer. A decrease in resistance after

NU

incubation may indicate some rearrangement of protein film between incubation, rinsing and

AC CE P

TE

D

MA

measuring steps.

Fig. 8. Nyquist plots of Cys-tau-Au film before (a) and after binding to ferritin (b), or transferrin (c) ([ferritin]=[transferrin]= 100 µg mL-1).

The electrochemical data indicate clear transferrin binding, however, no literature precedence exists about binding sites or binding affinity of transferrin to tau. The transferrin receptor binds two transferrin molecules via Leu122-Asp125 of transferrin [54]. Others have ascribed transferrin receptor binding to the transferrin loops 58-76, 139-145 and 154-167 and Asp356 15

ACCEPTED MANUSCRIPT [55]. Our electrochemical data show that tau is accessible for binding to transferrin even when the tau protein was attached to Au surface via N-terminal predominantly or Cys residues of R

PT

repeat domains.

RI

In order to evaluate importance of metal ion in ferritin or transferrin for binding to tau-Au films,

SC

the binding studies were carried out with apoferritin and apotransferrin. Both surfaces, N-tau-Au and Cys-tau-Au were used to evaluate binding of metal ion-free proteins. Fig. 9A shows

NU

significant increase in resistance after apotransferrin (c) binding to N-tau-Au, which was similar

MA

to transferrin binding. Notably, apoferritin (b) resulted in a significant increase in resistance as well, in contrast to ferritin with the same protein surface. When Cys-tau-Au was used for binding

D

to metal ion-free proteins, Fig. 9B, significant increase in resistance was observed after

TE

apoferritin, while apotransferrin resulted in a much smaller change in resistance. The electrochemical data for Cys-tau-Au indicate that presence of metal ions may have significantly

AC CE P

affected binding affinity.

Fig. 9. Nyquist plots of (A) N-tau-Au and (B) Cys-tau-Au film before (a) and after binding to apoferritin (b), or apotransferrin (c) ([apoferritin]=[apotransferrin]= 100 µg mL-1).

16

ACCEPTED MANUSCRIPT

Since phosphorylation and aggregation of tau protein may be synergistic events

PT

understanding of how they modulate protein-protein interactions is of importance. The impact of

RI

tau phosphorylation on binding to iron-rich proteins in not currently understood. There is

SC

evidence of the phosphotau protein strongly binding to Fe(III) ions [56]. However, how introduction of doubly negative charge, due to phosphorylation, affects binding of tau to ferritin

NU

or transferrin remains unknown. Importantly, tau protein is hyperphosphorylated, with multiple

MA

phospho groups introduced in tau protein sequence, which may additionally modulate proteinprotein interactions between tau and a variety of proteins, not limited to transferrin or ferritin,

D

such as protein chaperones. Future investigation of hyperphosphorylated tau with metal ions or

AC CE P

3. CONCLUSIONS

TE

metal-containing proteins is needed for complete understanding of tau biochemistry.

The electrochemical impedance spectroscopy was used to monitor binding of iron-rich proteins transferrin and ferritin to non-phosphorylated tau441 on Au surface. The results indicate weak binding by ferritin, but extremely good binding by transferrin for tau protein films when immobilized via N-terminal or Cys-residues. The transferrin binding was also concentration dependent. The iron-rich film was catalytic with biological reductant such as ascorbic acid. The interaction between the apoferritin and apotransferrin with tau protein films on Au surfaces indicated that some protein-protein interactions may be metal ion dependent. The current assay may be used for detection of other protein-protein interactions related to degeneration or metalloproteins systems. The synergistic effects of post-translational modifications, such as phosphorylation, and aggregation of tau on binding to other proteins also needs to be considered.

17

ACCEPTED MANUSCRIPT 4. EXPERIMENTAL SECTION

PT

4.1. Materials and Methods All proteins were used as received. Tau441 protein was purchased from rPeptides (GA, USA).

RI

Transferrin (human), apotransferrin (human) and apoferritin (equine spleen) were purchased

SC

from Sigma (MO, USA) and ferritin (human liver) was purchased from Calbiochem (CA, USA).

NU

A stock solution of phosphate buffer, pH 7.0, was prepared using 0.06 M sodium phosphate dibasic, anhydrous, obtained from J.T. Baker (NJ, USA) and 0.06 M sodium phosphate

MA

monobasic, anhydrous, obtained from Fisher Scientific (NJ, USA). The pH was adjusted with sodium hydroxide obtained from Fisher Scientific (NJ, USA) and phosphoric acid obtained from

D

Fisher Scientific (NY, USA). 1-Ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDC) and N-

TE

Hydroxysuccinimide (NHS) were purchased from Thermo Scientific (IL, USA). Ethanolamine,

AC CE P

hexanethiol, potassium ferricyanide (III) and potassium hexacyanoferrate (II) trihydrate were obtained from Sigma-Aldrich (MO, USA). 4.2. Preparation of the N-Tau-Au Surface The Au electrodes were cleaned as follows: (1) hand polishing in alumina (1 mm, 0.3 mm and 0.05 mm) for 1 min each, and (3) sonicating in deionized water (DI) for 5 minutes, while every 2.5 minutes disposing the water and adding new MilliQ H2O. Electrochemical cleaning was performed by CV in 0.5 M KOH solution at 0.5 V s-1 in the -2–0 V potential range, followed by 2 cyclic voltammetry scans in 0.5 M H2SO4 at 0.5 V s-1 in the 0–1.5 V potential range. The clean Au electrodes were rinsed with DI water, dried under a N2 flow, and then incubated in a 2 mM solution of lipoic acid in ethanol for 24 h at 5°C. Then, the Lip–Au electrodes were rinsed with ethanol. For tau protein immobilization, Lip-Au electrodes were incubated 20 mM EDC and 18

ACCEPTED MANUSCRIPT 50 mM NHS at room temperature for 1 h. These electrodes were then rinsed with 10 mM phosphate buffer, pH 7.0. For tau protein immobilization, these electrodes were incubated with

PT

10 μg mL-1 tau (50 mM 2-(N-morpholino)ethanesulfonic acid (MES), pH 6.8, 100 mM NaCl,

RI

and 0,5 mM EDTA) overnight at 5 °C. The tau-Au electrodes were rinsed 10 mM phosphate buffer, pH 7.0, and immersed in 100 mM ethanolamine solution for 1 h at room temperature.

SC

After rinsing, the electrodes were next, incubated in 10 mM hexanethiol solution for 20 min at

NU

room temperature. Subsequently, the electrodes were rinsed with 10 mM phosphate buffer

MA

solution, pH 7.0, prior to electrochemical measurements. 4.2. Preparation of the Cys-tau-Au Surface

D

The Cys-tau-Au film was prepared by immersing clean Au electrode in 10 µg mL-1 solution of

TE

tau protein (50 mM MES, pH 6.8, 100 mM NaCl, and 0.5 mM EDTA) for 24 h at 5˚ C. The tau-

AC CE P

Au electrodes were rinsed with 10 mM phosphate buffer, pH 6.8, and backfilled with 10 mM hexanethiol solution for 20 min at 25˚ C. Next, the tau-Au electrodes were rinsed with phosphate buffer, pH 6.8 prior to electrochemical measurements. 4.3. Electrochemical Measurements Electrochemical experiments were carried out using a CHI660D potentiostat from CH Instruments Inc. (TX, USA). The gold disk electrodes (0.0314 cm2 surface area) were purchased from CH Instruments Inc. (TX, USA). A conventional three-electrode system consisting of a gold electrode as the working electrode, platinum wire as the auxiliary electrode and Ag/AgCl/1.0 M KCl as the reference electrode was used for all experiments. All electrochemical measurements were performed in 10 mM K3[Fe(CN)6], 10 mM K4[Fe(CN)6], and 0.06 M phosphate buffer, pH 7.0, unless otherwise specified. EIS was carried out starting at an open 19

ACCEPTED MANUSCRIPT circuit potential (OCP) of 0.2 V, a frequency range between 0.01 Hz and 100000 Hz, and an applied amplitude of 5 mV. Experimental EIS data were plotted to an equivalent circuit using

PT

ZSimp Win 3.22 (Princeton Applied Research). Fitted and experimental data were presented in

RI

the form of Nyquist plots. The charge transfer resistance, Rct, was determined by plotting the

SC

impedance data to the appropriate equivalent circuit and was expressed in Ω. All experiments were performed in triplicate.

NU

4.4. Electrochemical Measurements

MA

The binding studies with ferritin and transferrin were carried out by incubating Tau-Au films into 100 µg mL-1 (unless otherwise specified) ferritin or transferrin solution in phosphate buffer, pH

D

7.0 for 2 h at 37 ˚C. After binding, electrodes were rinsed with phosphate buffer, pH 7.0 and

TE

measured.

AC CE P

4.5. Ascorbic acid measurements

The prepared films of Tau-Au incubated with ferritin or transferrin were immersed in the solution of 1 mM ascorbic acid (phosphate buffer, pH 7.0) and square-wave voltammograms were acquired. The Tau-Au incubated in phosphate buffer pH 7.0, instead of ferritin or transferrin, was used as a control. 4.6. Contact Angle The samples for surface characterization were prepared using Au sputtered silicon wafers (Nanofabrication Facility, Western University, Canada). The Au wafers were cleaned by etching with piranha solution for 5 min and rinsing with copious amount of MilliQ water. Next, the substrates were rinsed with ethanol and dried with N2. For tau film preparation the stepwise

20

ACCEPTED MANUSCRIPT modification steps were identical to those described previously. Once tau films were prepared, the samples were incubated in 100 μg mL-1 ferritin or transferrin for 2 h at 37 °C. Finally,

PT

substrates were rinsed and measured. Contact angle (θ) was measured using the method based on

RI

the diameter of a 10 mL sessile drop of 10 mM phosphate buffer at a pH of 7.0 to distinguish surface wettability of tau–Au before and after binding ferritin and transferrin. Each sample was

NU

SC

tested in duplicate measurements.

MA

4.7. X-ray photoelectron spectroscopy

The samples for surface characterization were prepared using Au sputtered silicon wafers

D

(Nanofabrication Facility, Western University, Canada). The Au wafers were cleaned by etching

TE

with piranha solution for 5 min and rinsing with copious amount of MilliQ water. Next, the substrates were rinsed with ethanol and dried with N2. For tau film preparation the stepwise

AC CE P

modification steps were identical to those described previously. Once tau films were prepared, the samples were incubated in 100 μg mL-1 ferritin or transferrin for 2 h at 37 °C. Finally, substrates were rinsed and measured. The samples were analyzed by X-ray photoelectron spectroscopy (XPS) using a Kratos Axis Ultra X-ray photoelectron spectrometer. XPS can detect all elements except hydrogen and helium, probes the surface of the sample to a depth of 710 nanometers, and has detection limits ranging from 0.1 to 0.5 atomic percent depending on the element. Survey scan analyses were carried out with an analysis area of 300 x 700 microns and a pass energy of 160 eV. High resolution analyses were carried out with an analysis area of 300 x 700 microns and a pass energy of 20 eV. High resolution spectra were charge corrected to Au 4f7/2 set to 83.95 eV. ABBREVIATIONS 21

ACCEPTED MANUSCRIPT Electrochemical impedance spectroscopy

CV

Cyclic voltammetry

Rct

Charge-transfer resistance

NHS

N-hydroxysuccinimide

XPS

X-ray photoelectron spectroscopy

NU

SC

EDC 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide

tau immobilized via N-terminal

Cys-tau-Au film

tau immobilized via Cys-residues

AD

Alzheimer’s Disease

ITO

Indium tin oxide

Z’’

imaginary impedance

Z’

real impedance

CPE

constant phase element

AC CE P

TE

reactive oxygen species

D

MA

N-tau-Au film

ROS

NMR nuclear magnetic resonance MS

mass spectrometry

CD

circular dichroism

RI

PT

EIS

22

PT

ACCEPTED MANUSCRIPT

RI

ACKNOWLEDGEMENTS

SC

Authors would like to thank Department of Chemistry and Oakland University for support. A. J.

APPENDIX A. SUPPLEMENTARY DATA

NU

thanks Honors College for support.

AC CE P

REFERENCES

TE

D

MA

Supplementary data to this article can be found online at .

[1] M. D. Wingarten, A. H. Lockwood, S. Y. Hwo, M. W. Kirschner, Proc. Natl. Acad. Sci. U. S. A. 72 (1975) 1858-1862.

[2] A. D. Alonso, J. Di Clerico, B. Li, C. P. Corbo, M. E. Alaniz, I. Grundke-Iqbal, K. Iqbal, J, Biol. Chem. 285 (2010) 30851-30860. [3] A. De Calignon, M. Polydoro, M. Suarez-Calvet, C. William, D. H. Adamowicz, K. J. Kopeikina, R. Pitstick, N. Sahara, K.H. Ashe, G. A. Carlson, T. L. Spires-Jones, B.T. Hyman, Neuron. 73 (2012) 685-697. [4] N. Kfoury, B.B. Holmes, H. Jiang, D.M. Holtzman, M. I. Diamond, J. Biol. Chem. 287 (2012) 19440-1945. 23

ACCEPTED MANUSCRIPT [5] E. R. Stadtman, Science. 257 (1992) 1220-1224. [6] M. A. Smith, G. Perry, P. L. Richey, L. M. Sayre, V. E. Anderson, M. F. Beal, N. Kowall,

PT

Nature. 382 (1996) 120-121.

RI

[7] M. A. Lovell, J. D. Robertson, W. J. Teesdale, J. L. Campbell, W. R. Markesbery, J.

SC

Neurological Sci. 158 (1998) 47-52.

MA

Frederickson, Brain Res. 852 (2000) 274-278.

NU

[8] S. W. Suh, K. B. Jensen, M. S. Jensen, D. S. Silva, P. J. Kesslak, G. Danscher, C. J.

[9] P. Faller, C. Hureau, O. Berthoumieu, Inorg. Chem. 2013, 52, 12193-12206.

D

[10] P. Fischer, M.E. Giitz, W. Danielczyk, W. Gsell, P. Riederelf, Life Sci. 60 (1997) 2273-

TE

2278.

AC CE P

[11] P. De Sole, C. Rossi, M. Chiarpotto, G. Ciasca, B. Bocca, A. Alimonti, A. Bizzarro, C. Rossi, C. Masullo, Clinical Biochem. 46 (2013) 89-93. [12] H. Yang, M. Yang, H. Guan, Z. Liu, S. Zhao, S. Takeuchi, D. Yanagisawa, I. Tooyama, Neurosci. Res. 77 (2913) 1-7.

[13] M. A. Smith, G. Perry, J. Neurol. Sci. 134 (1995) 92-94. [14] M. A. Smith, P.L.R. Harris, L. M. Sayre, G. Perry, Proc. Natl. Acad. Sci. USA, 94 (1997) 9866-9868. [15] E. Maerelli, P. Sola, P. Faglioni, V. Pavarotti, Riv. Patol. Nerv. Ment. 103 (1982) 253-261. [16] C. E. Arber, L. Abi, H. Houlden, S. Wray, Neuropathol. Appl. Neurobiol. 47 (2015) 1-22.

24

ACCEPTED MANUSCRIPT [17] L. Zecca, M. B. H. Youdim, P. Riederer, J. R. Connor, R. R. Crichton, Nat. Rev. Neurosci. 5 (2004) 863-873.

PT

[18] K. Jellinger, W. Paulus, I. Grundke-Iqbal, P. Riederer, M. B. Youdim, Disease. J. Neural.

RI

Transm. Park. Dis. Dement. Sect. 2 (1990) 327-340.

SC

[19] I. Grundke-Iqbal, J. Fleming, Y. C. Tung, H. Lassmann, K. Iqbal, J. G. Joshi, Acta

NU

Neuropathol. 81 (1990) 105-110.

MA

[20] R. R. Crichton, S. Wilmet, R. Legssyer, R. J. Ward, J. Inorg. Biochem. 91 (2002) 9-18. [21] A. Soragni, B. Zambelli, M. D. Mukrasch, J. Biernat, S. Jeganathan, C. Griesinger, S. Ciurli,

D

E. Mandelkow, M. Zweckstetter, Biochemistry. 47 (2008) 10841-110851.

AC CE P

(2006) 841-849.

TE

[22] Q. Ma, Y. Li, J. Du, H. Liu, K. Kanazawa, T. Nemoto, H. Nakanishi, Y. Zhao, Peptides. 27

[23] D. J. Yang, S. Shi, T. M. Yao, L. N. Ji, Biometals. 25 (2011) 361-372. [24] Q. F. Ma, Y. M. Li, J. T. Du, K. Kanazawa, T. Nemoto, H. Nakanishi, Y. F. Zhao, Biopolymers. 79 (2005) 74-85.

[25] B. Bader, G. Nubling, A. Mehle, S. Nobile, H. Kretzschmar, A. Giese, Biochem. Biophys. Res. Commun. 411 (2011) 190-196. [26] A. Yamamoto, R. W. Shin, K. Hasegawa, H. Naiki, H. Sato, F. Yoshimasu, T. Kitamoto, J. Neurochem. 82 (2002) 1137-1147. [27] M. G. Savelieff, S. Lee, Y. Liu, M. H. Lim, ACS Chem. Biol. 8 (2013) 856-865.

25

ACCEPTED MANUSCRIPT [28] B. I. Giasson, D. M. Sampathu, C. A. Wilson, V. Vogelsber-Ragaglia, W. E. Mushynski, V. M. Y. Lee, Biochemistry. 41 (2002) 15376-15387.

PT

[29] X. Du, Y. Zheng, Z. Wang, Y. Chen, R. Zhou, G. Song, J. Ni, Q. Liu, Inorg. Chem. 53

RI

(2014) 11221-11230.

SC

[30] B-k. Shin, S. Saxena, J. Phys. Chem. B. 115 (2011) 15067-15078.

NU

[31] M. Perez, J. M. Valpuesta, E. Montejo de Garcini, C. Quintana, M. Arrasate, J. L. Lopez-

MA

Carrascosa, A. Rabano, J. Garcia de Yebenes, J. Avila, Am. J. Pathol. 152 (1998) 1531-1539. [32] B. Gallo, F. Bracco, S. Morara, L. Battistin, B. Tavolato, J. Neurol. Sci. 70 (1985) 81-92.

D

[33] J. R. Connor, S. L. Menzies, S. M. St. Martin, E. J. Mufson, J. Neurosci. Res. 31 (1992) 75-

TE

83.

AC CE P

[34] J. R. Connor, S. L. Menzies, J. Neurol. Sco. 134 (1995) 33-44. [35] M. A. Smith, P. L. R. Harris, L. M. Sayre, G. Perry, Proc. Natl. Acad. Sci. USA, 94 (1997) 9866-9868.

[36] F. Clavaguera, T. Bolmont, R. A. Crowther, D. Abramowski, S. Frank, A. Probst, G. Fraser, A. K. Stalder, M. Beiber, M., Staufenbiel, M. Jucker, M. Goedert, M. Tolnay, Nat. Cell. Biol. 2009, 11, 909-913. [37] D. W. Sanders, S. K. Kaufman, S. L. DeVos, A. M. Sharma, H. Mirbaha, A. Li, S. J. Barker, A. C. Foley, J. R. Thorpe, L. C. Serpell, T. M. Miller, L. T. Grinber, W. W. Seeley, M. I. Diamond, Neuron. 2014, 82, 1271-1288.

26

ACCEPTED MANUSCRIPT [38] E. Kovesdi, J. Luckl, P. Bukovics, O. Farkas, J. Pal, E. Czeiter, D. Szellar, T. Doczi, S. Komoly, A. Buki, Acta Neurochir 2010, 152, 1-17.

PT

[39] S. Balasubramanian, A. Revzin, A. Simonian, Electroanalysis, 18 (2006) 1885-1892.

RI

[40] M. Tolba, M. U. Amhed, C. Tlili, F. Eichenseher, M. J. Loessner, M. Zourob, Analyst, 137

SC

(2012) 5749-5756.

NU

[41]M. A. MacDonald, H. A. Andreas, Electrochim. Acta 129 (2014) 290-299.

MA

[42] F. Marken, D. Patel, C. E. Madden, R. C. Millward, S. Fletcher, New. J. Chem. 26 (2002) 259-263.

D

[43] T. D. Martin, S. A. Monheit, R. J. Niichel, S. C. Peterson, C. H. Campbell, D. C. Zapien, J.

TE

Electroanal. Chem. 420 (1997) 279-290.

AC CE P

[44] D. C. Zapien, M. A. Johnson, J. Electoanal. Chem. 494 (2000) 114-120. [45] W. Senaratne, K. Takada, R. Das, J. Cohen, B. Baird, H. D. Abruna, C. K. Ober, Biosens. Bioelectron. 22 (2006), 63-70.

[46] M. Wasowics, S. Viswanathan, A. Dvornyk, K. Grzelak, B., Kludkiewics, H. Radecka, Biosens. Bioelectron. 24 (2008) 284-289. [47] K. L. Prime, G. M. Whitesides, J. Am. Chem. Soc. 115 (1993) 10714-10721. [48] K. C. Martin, S. M., Villano, P. R. McCurdy, D. C. Zapien, Langmuir. 19 (2003) 58085810. [49] F. Caruso, D. N. Furlong, P. Kingshott, J. Colloid Interface Sci. 186 (1997) 129-140.

27

ACCEPTED MANUSCRIPT [50] T. Onda, S. Shinuichi, N. Satoh, K. Tsujii, Langmuir. 12 (1996) 2125-2127.

PT

[51] S. Martic, M. K. Rains, H. B. Kraatz, Analytical Biochemistry. 442 (2013) 130-137. [52] M. A. Smith, R. K. Kutty, P. L. Richey, S.D. Yan, D. Stem, G. J. Chader, B. Wiggert, P. B.

RI

Pertsen, G. Perry. Am. J. Pathol. 145 (1994), 42-47.

SC

[53] H. Trzeciakiewicz, J. O. Esteves-Villanueva, N. Carlin, S. Martic, Electrochim. Acta, 162,

NU

2015, 24-30.

[54] P. M. Kelly, C. Aberg, E. Polo, A. O’Connell, J. Cookman, J. Fallon, Z., Krpetic, K. A.

MA

Dawson, Nat. Nanotech. 10 (2015) 472-479.

[55] B. E. Eckenroth, A. N. Steere, N. D. Chasteen, S. J. Everse, A. B. Mason, Proc. Natl. Acad.

TE

D

Sci. U.S.A. 108 (2011) 12089-12094.

AC CE P

[56] A. Yamamoto, R. W. Shin, K. Hasegowa, H. Naiki, H. Sato, F. Yoshimasu, T. Kitamoto, J. Neurochem. 82 (2002) 1137-1147.

Figure Captions

Fig. 1. Schematic illustration of tau-Au fabrication: a) Lipoic acid, b) EDC/NHS, c) tau, d) ethanolamine, and e) hexanethiol. Fig. 2. (A) CVs and (B) Nyquist plots of step-wise surface modification steps: a) bare Au, b) lipoic acid, c) EDC/NHS, d) tau, e) ethanolamine, and f) hexanethiol. Fig. 3. Nyquist plots of tau-Au film before (a) and after binding to ferritin (b), or transferrin (c) ([ferritin]=[transferrin]= 100 µg mL-1). Fig. 4. Plot of Rct factor of tau-Au as a function of ferritin or transferrin concentrations. 28

ACCEPTED MANUSCRIPT Fig. 5. High-resolution XPS spectra of S 2p for (A) tau-Au exposed to (B) ferritin or (C) transferrin.

PT

Fig. 6. High-resolution XPS spectra of Fe 2p for (A) tau-Au exposed to (B) ferritin or (C)

RI

transferrin.

SC

Fig. 7. Square-wave voltammograms of Tau-Au films after incubation with phosphate buffer (a),

NU

transferrin (b) or ferritin (c) in ascorbic acid-free solution (A) and in ascorbic acid solution (B) ([ascorbic acid]=1 mM).

MA

Fig. 8. Nyquist plots of Cys-tau-Au film before (a) and after binding to ferritin (b), or transferrin

D

(c) ([ferritin]=[transferrin]= 100 µg mL-1).

TE

Fig. 9. Nyquist plots of (A) N-tau-Au and (B) Cys-tau-Au film before (a) and after binding to

AC CE P

apoferritin (b), or apotransferrin (c) ([apoferritin]=[apotransferrin]= 100 µg mL-1).

29

ACCEPTED MANUSCRIPT

Synopsis of Graphical Abstract

AC CE P

TE

D

MA

NU

SC

RI

PT

Transferrin binding to immobilized tau protein on Au electrode was detected by electrochemical impedance spectroscopy. The increase in the charge-transfer resistance was observed for transferrin over ferritin binding. The protein orientation on the surface and the iron ions affected protein-protein interactions.

30

ACCEPTED MANUSCRIPT

AC CE P

TE

D

MA

NU

SC

RI

PT

Graphical abstract

31

ACCEPTED MANUSCRIPT Highlights

TE

D

MA

NU

SC

RI

PT

Tau was immobilized via N-terminal (N-tau-Au film) or Cys-residues (Cys-tau-Au film) Transferrin binding to N-tau or Cys-tau-Au films increased charge transfer resistance Ferritin binding to N-tau-Au and Cys-tau-Au was not detected Metal ions affected protein binding to tau films

AC CE P

   

32