Biochimica et Biophysica Acta 1854 (2015) 907–918

Contents lists available at ScienceDirect

Biochimica et Biophysica Acta journal homepage: www.elsevier.com/locate/bbapap

Zinc significantly changes the aggregation pathway and the conformation of aggregates of human prion protein Kai Pan, Chuan-Wei Yi, Jie Chen, Yi Liang ⁎ State Key Laboratory of Virology, College of Life Sciences, Wuhan University, Wuhan 430072, China

a r t i c l e

i n f o

Article history: Received 25 January 2015 Received in revised form 9 April 2015 Accepted 21 April 2015 Available online 25 April 2015 Keywords: Prion protein Prion disease Zinc Protein aggregation Protein conformational change

a b s t r a c t Prion diseases are caused by the conformational change of cellular prion protein PrPC into pathological prion protein PrPSc. Here we study the effect of zinc on the aggregation and conformational change of human prion protein (PrP). As revealed by thioflavin T binding assays, Sarkosyl-soluble SDS-PAGE, and transmission electron microscopy, aggregation of wild-type PrP in the absence of Zn2+ undergoes four steps: amorphous aggregates, profibrils, mature fibrils, and fragmented fibrils. When the molar ratio of Zn2+ to PrP was 9:1, however, aggregation of wild-type PrP undergoes another pathway in which wild-type PrP forms oligomers quickly and then forms short-rod aggregates. Unlike wild-type PrP, the octarepeats deletion mutant PrPΔocta forms typical mature fibrils either with or without zinc. As evidenced by isothermal titration calorimetry, Fourier transform infrared spectroscopy, and proteinase K digestion assays, Zn2+ strongly binds to wild-type PrP monomers with the first binding constant exceeding 107 M−1 under denaturing conditions, and changes the conformation of wild-type PrP aggregates remarkably, but weakly binds to PrPΔocta with binding affinity around 104 M−1 and has no obvious effects on the conformation of PrPΔocta aggregates. Our data demonstrate that zinc significantly changes the aggregation pathway and the conformation of wild-type PrP aggregates mainly via interaction with its octarepeat region. Our findings could explain how zinc modifies pathological PrP conformation associated with prion diseases. © 2015 Elsevier B.V. All rights reserved.

1. Introduction Prion diseases, such as bovine spongiform encephalopathy, scrapie of sheep, and Creutzfeldt–Jakob disease of humans, are caused by the misfolding and conformational change of the normal cellular form of prion protein (PrPC) into an abnormal disease-causing isoform of prion protein (PrPSc) [1–4]. The protein-only hypothesis points out that the infectious agents of prion diseases mainly consist of pathological prion protein PrPSc without nucleic acids [1]. PrPC is a cell surface glycoprotein anchored to cell surface via glycosylphosphatidylinositol (GPI) [4]. There is no difference in sequence between PrP C and PrPSc [1–4]. PrPC and PrPSc differ in their conformations and threedimensional structures, and PrPC is largely α-helical contained but PrPSc is largely β-sheet contained [1]. There are a number of NMR structures of PrP C from different species published so far, but the

Abbreviations: CD, circular dichroism; GPI, glycosylphosphatidylinositol; FTIR, Fourier transform infrared; ITC, isothermal titration calorimetry; PrP, prion protein; PrPΔocta, human prion protein with the deletion of octarepeats (58–89); PrPC, the normal cellular PrP molecule; PrPSc, an abnormal disease-causing isoform of prion protein; IAPP, islet amyloid polypeptide; SDS, sodium dodecyl sulfate; TEM, transmission electron microscopy; ThT, thioflavin T; GdnHCl, guanidine hydrochloride; PK, proteinase K. ⁎ Corresponding author. Tel./fax: +86 27 68754902. E-mail address: [email protected] (Y. Liang).

http://dx.doi.org/10.1016/j.bbapap.2015.04.020 1570-9639/© 2015 Elsevier B.V. All rights reserved.

high-resolution structure of PrPSc is lacking due to its insolubility and heterogeneity [2,5]. The mature human PrPC consists of residues 23–231, with an unstructured N-terminal half 23–120 and a global C-terminal half 121–231. The residues 58–89 are an octapeptide repeats sequence, with GQPHGGGW repeated four times. This sequence can bind to several bivalent metal ions, such as copper, zinc, and manganese [6–9]. Many researches point out that prion protein (PrP) is a copper-binding protein [10,11]. There are several copper binding sites, four of which are in its octarepeat region, two of which just after the octarepeats, and two of which in the helical region [6,7]. However, how copper affects prion diseases is complicated and controversial. Some experiments show that interaction between copper and PrP can accelerate the onset of prion diseases [12–15], but other experiments show that copper can prevent the conversion of PrP into amyloid fibrils and delay the onset of prion diseases [16,17]. The affinity of copper binding to PrP can be above 1010 M−1, much higher than that for zinc binding to PrP (about 104 M−1) [10,11,18,19]. Zinc, as one of the most abundant transition metals in the brain, is present in and essential to all forms of life [20,21]. In the synaptic cleft, where PrP is enriched, the concentration of Zn2 + can be up to 100–300 μM [20,21], remarkably higher than that of copper [6]. Therefore, we want to know whether and how micromolar Zn2+ affects the misfolding and conformational change of PrP, associating with prion

908

K. Pan et al. / Biochimica et Biophysica Acta 1854 (2015) 907–918

diseases. A valuable evidence of PrPC–zinc interaction is that not only neuronal zinc potently stimulates the endocytosis of PrPC [22,23], but also PrPC enhances neuronal zinc uptake [24]. Clearly, PrP is involved in neuronal zinc homeostasis [21]. In addition, it has been reported that different conformations of PrPSc can lead to different pathological symptoms of prion diseases, and PrPSc conformation is the foundation of prion strain diversity [1,3,25]. The ability of Zn2 + to interact with PrPSc and influence PrPSc conformation directly has widespread implications for understanding the pathogenesis of prion diseases [25]. However, how zinc affects PrP fibrillization and conformational change is still poorly studied. One report shows that the fibrillization of human PrP106-126 is completely inhibited in an environment depleted of transition metals, and Zn2+ can restore PrP106-126 aggregation [26], but another report shows that Zn2+ can inhibit conversion of full-length mouse PrP into amyloid fibrils [16]. Zinc plays an important role in the misfolding and amyloid properties of many different amyloidogenic proteins and peptides like islet amyloid polypeptide (IAPP) [27–32], amyloid β [31–33], and insulin [30,31] in addition to PrP [16,25,26]. Previous studies on IAPP's interactions with zinc have indicated that zinc clearly suppresses the aggregation of IAPP [27–29]. Recent studies have shown that zinc can alter the aggregation pathway of IAPP and amyloid β [32]. Related NMR studies have shown the structural changes accompanied by zinc binding [27–29,32]. In this study, we investigated the effect of zinc on the aggregation and conformational change of wild-type human PrP and its octarepeats deletion mutant PrPΔocta. By using thioflavin T (ThT) binding assays, Sarkosyl-soluble SDS-PAGE, and transmission electron microscopy (TEM), we demonstrated that in the presence of 100 μM Zn2+, wildtype PrP formed oligomers quickly and then formed short-rod aggregates. In the absence of Zn2 +, however, aggregation of wild-type PrP underwent four different stages of amorphous aggregates, profibrils, mature fibrils, and finally fragmented fibrils. By using isothermal titration calorimetry (ITC), circular dichroism (CD), Fourier transform infrared (FTIR) spectroscopy, and proteinase K (PK) digestion assays, we found that Zn2+ strongly bound to the apo form of wild-type PrP under denaturing conditions. We also reported that the binding was accompanied by a remarkable conformational change of wild-type PrP aggregates, whereas such phenomena were not observed for PrPΔocta. Our results indicated that zinc significantly changed the aggregation pathway and the conformation of wild-type PrP aggregates mainly via interaction with its octarepeat region. Our findings could explain how zinc modifies pathological PrP conformation associated with prion diseases and why so many prion strains existed. 2. Materials and methods

on a C4 reversed-phase column (Shimadzu, Kyoto, Japan) as described by Bocharova and co-workers [16,34] with minor changes. Briefly, the inclusion body was dissolved in 8 M urea, purified by Ni-sepharose, refolded in 6 M urea containing 2 mM ethylenediaminetetraacetic acid over night, and further purified by C4 reversed-phase column. After purification, human PrP was dialyzed against 5 mM MES buffer for two times, concentrated and filtered. Purified human PrP was confirmed by SDS-PAGE and mass spectrometry to be single species with an intact disulfide bond. The concentration of human PrP was determined by its absorbance at 280 nm using the molar extinction coefficient value deduced from the composition of the protein online (http://web.expasy. org/protparam/). Metal content was determined by atomic absorption using an AAnalyst-800 atomic absorption spectrometer (PerkinElmer Life Sciences, Shelton, CT) to confirm that human PrP samples were indeed in the apo state, in which the molar ratios of Cu2+, Zn2+, and Ni2+ to human PrP monomer were 0.002:1, 0:1, and 0:1, respectively. 2.3. Fibril formation To form fibrils, freshly prepared (within two weeks) wild-type human PrP and its octarepeats deletion mutant PrPΔocta were diluted to a final concentration of 11 μM in 5 mM MES buffer (pH 7.0) containing 2 M GdnHCl and related concentration of zinc, incubated at 37 °C and shaking at 220 rpm. 2.4. ThT binding assays A 2.5 mM ThT stock solution was freshly prepared in 5 mM MES buffer (pH 7.0) and passed through a 0.22-μm pore size filter before use to remove insoluble particles. A serial of 50 μl samples were taken in chronological order, and mixed with 2.5 ml of 5 mM MES buffer and 12.5 μl of ThT stock solution before measurements. The fluorescence of ThT was excited at 450 nm with a slit-width of 5.0 nm and the emission was measured at 480 nm with a slit-width of 5.0 nm on an LS-55 luminescence spectrometer (PerkinElmer Life Sciences, Shelton, CT). Kinetic parameters were determined by fitting ThT fluorescence intensity versus time to a sigmoidal equation [35–38]: F ¼ F 0 þ ðA þ ct Þ=f1 þ exp½kðt m −t Þg

ð1Þ

where F is the fluorescence intensity, k the rate constant for the growth of fibrils, and tm the time to 50% of maximal fluorescence. F0 describes the initial baseline during the lag time. A + ct describes the final baseline after the growth phase has ended. The lag time is determined to be tm − 2/k.

2.1. Materials 2.5. Sarkosyl-soluble SDS-PAGE ThT was purchased from Sigma-Aldrich (St. Louis, MO). Acetonitrile was obtained from Fisher Scientific (Fairlawn, NJ). Guanidine hydrochloride (GdnHCl) was obtained from Promega (Madison, WI). Proteinase K, Triton X-100 and Sarkosyl were purchased from Amresco (Solon, OH). All other chemicals used were made in China and were of analytical grade.

A serial of 20 μl human PrP samples were taken in chronological order during fibril formation corresponding to ThT binding assays. 20 μl samples were added with 2.5 μl 20% Sarkosyl. The mixture was left at room temperature for 30 min, and then mixed with 5× loading buffer (without β-mercaptoethanol and SDS, and no heating) and separated by 15% SDS-PAGE. Gels were stained by Coomassie Blue R250.

2.2. Plasmid construction and protein purification 2.6. Transmission electron microscopy (TEM) Wild-type human PrP (23–231) and its octarepeats deletion mutant PrPΔocta (23–57, 90–231) were constructed into the pET30a plasmids. PrPΔocta was generated by polymerase chain reaction (PCR) using wild-type PrP plasmid as a template with primers 5′GGTCAAGGAGGT GGCACCCACAGTCAGT/5′CCAGCCACCACCGC CCTGAGGTG. Then ligation reaction was employed to link the two blunt ends of PCR product. The plasmids for wild-type PrP and PrPΔocta were transformed into Escherichia coli. Recombinant wild-type human PrP and its mutant were expressed in E. coli BL21 (DE3) (Novagen), and purified by HPLC

The formation of fibrils by human PrP was confirmed by electron microscopy of negatively stained samples. The preparation for negatively stained samples was described in detail previously [36,38]. Sample aliquots of 10 μl were placed on copper grids and left at room temperature for 1–2 min, rinsed twice with H2O, and then stained with 2% (w/v) uranyl acetate for another 1–2 min. The stained samples were examined using an H-8100 (or an H-7000 FA) transmission electron microscope (Hitachi, Tokyo, Japan) operating at 100 kV.

K. Pan et al. / Biochimica et Biophysica Acta 1854 (2015) 907–918

2.7. Isothermal titration calorimetry (ITC)

3. Results

ITC experiments on the interaction between Zn2 + and wild-type human PrP and its octarepeats deletion mutant PrPΔocta were carried out at 25.0 °C using a VP-ITC titration calorimetry (MicroCal, Northampton, MA). Freshly purified wild-type human PrP and PrPΔocta, both without ions, were obtained as described above and dialyzed against 5 mM MES buffer (pH 7.0) in the presence or absence of 2 M GdnHCl for three times at 4 °C. A solution of 20.0 μM human PrP was loaded into the sample cell (1.43 ml), and a solution of 2.0 mM Zn2+ was placed in the injection syringe (300 μ l). The first injection (3 μ l) was followed by 29 injections of 10 μ l. Dilution heats of Zn2+ were measured by injecting Zn2+ solution into buffer alone and were subtracted from the experimental curves prior to data analysis. The stirring rate was 300 rpm. The resulting data were fitted to a three sequential binding sites model using MicroCal ORIGIN software supplied with the instrument, and the standard molar enthalpy change 0 , the binding constant, Kb, and the binding stoichifor the binding, ΔbHm ometry, n, were thus obtained. The standard molar free energy change, 0 0 , and the standard molar entropy change, Δb Sm , for the binding ΔbGm reaction were calculated by the fundamental equations of thermodynamics [37,39]:

3.1. Effects of Zn2+ on the kinetics of fibrillization of human PrP

Δb G0m ¼ −RT ln K b

ð2Þ

  Δb S0m ¼ Δb H 0m −Δb G0m =T:

ð3Þ

2.8. Circular dichroism spectroscopy CD spectra were obtained by using a Jasco J-810 spectropolarimeter (Jasco Corp., Tokyo, Japan) with a thermostated cell holder. Quartz cell with a 1 mm light-path was used for measurements in the far-UV region. Spectra were recorded from 195 to 250 nm for far-UV CD. Human PrP fibril samples were subjected to extensive dialysis against 5 mM MES buffer (pH 7.0) to remove guanidine hydrochloride. Measurements were made at 25 °C. The spectra of all scans were corrected relative to the buffer blank. The mean residue molar ellipticity [θ] (deg·cm2·dmol−1) was calculated using the formula [θ] = (θobs/ 10)(MRW/lc), where θobs is the observed ellipticity in deg, MRW the mean residue molecular weight (109.2 Da for wild-type human PrP and 111.4 Da for PrPΔocta), l the path length in cm, and c the protein concentration in g/ml. 2.9. Fourier transform infrared spectroscopy Attenuated total reflection FTIR spectra were recorded using a Nicolet 5700 FTIR spectrophotometer (Thermo Electron, Madison, WI). Human PrP aggregate samples were prepared in D2O and FTIR spectra were recorded in the range from 400 to 4000 cm−1 at 4 cm−1 resolution. The sample was scanned 128 times in each FTIR measurement, and the spectrum acquired is the average of all these scans. Spectra were corrected for the D2O and water vapors. To analyze the absorbance spectra, OMNIC 8 software was employed. Spectra were processed using 5 point smoothing. Measurements were made at 25 °C. 2.10. PK digestion assays Aggregate samples of wild-type human PrP and its octarepeats deletion mutant PrPΔocta were prepared in 5 mM MES buffer (pH 7.0) containing 2 M GdnHCl and incubated with PK at a PK:PrP molar ratio of 1:100 to 1:50 for 1 h at 37 °C. Digestion was stopped by the addition of 2 mM phenylmethylsulfonyl fluoride (PMSF), and samples were analyzed in 15% SDS-PAGE and detected by silver staining.

909

Because the enhanced fluorescence emission of the dye ThT is a specific marker for the β-sheet conformation of fibril structures [34–38,40], we performed ThT binding assays to monitor the kinetics of amyloid fibril formation of human PrP. In order to semi-quantify the decrease/ increase of human PrP monomers in the presence of Zn2+, we carried out Sarkosyl-soluble SDS-PAGE experiments. For wild-type human PrP and its octarepeats deletion mutant PrPΔocta, time dependent ThT fluorescence and time dependent Sarkosyl-soluble SDS-PAGE experiments are shown in Figs. 1 and 2, respectively. Fig. 1 shows the ThT fluorescence curves based on three independent experiments. To elucidate the detailed effects of Zn2+ on amyloid fibril formation of human PrP, a sigmoidal equation was used to fit the kinetic data in Fig. 1, yielding three kinetic parameters, which are summarized in Table 1. As shown in Fig. 1A–C and Table 1, the presence of 10 μM Zn2+ in 5 mM MES buffer (pH 7.0) containing 2 M GdnHCl inhibited amyloid fibril formation of 11 μM wild-type PrP, accompanied by a moderate decline of the maximum ThT intensity. Furthermore, fibrillization of 11 μM wild-type PrP was remarkably inhibited by 50, 100, or 200 μM Zn2 + on the investigated time scale, accompanied by a remarkable decline of the maximum ThT intensity (Fig. 1A–C). As shown in Fig. 1D–F and Table 1, the presence of 100 μM Zn2 + significantly decelerated the nucleation step of fibril formation of 11 μM PrPΔocta, resulting in a lag time of 23.3 h in the presence of Zn2+, which is two-fold increased compared with that in the absence of Zn2 + (10.4 h), but accompanied by a remarkable enhancement of the maximum ThT intensity. Because GdnHCl is a strong denaturing agent [36,38], we want to know whether the use of GdnHCl in this study disaggregates PrP aggregates. As shown in Fig. S1, a band corresponding to Sarkosyl-soluble wild-type PrP (or PrPΔocta) monomers was not observed when wild-type PrP (or PrPΔocta) aggregates, formed either with or without 100 μM Zn2+, was incubated in the presence of 2 M GdnHCl for 4 days. Clearly, the use of GdnHCl did not disaggregate PrP aggregates under such conditions. To further investigate the kinetics of PrP aggregation, we did the following Sarkosyl-soluble SDS-PAGE experiments (Fig. 2). As shown in Fig. 2A, a clear band corresponding to Sarkosyl-soluble wild-type PrP monomers was still observed when wild-type PrP was incubated in the absence of Zn2+ for 9 h, indicating a lag time over 9 h, which is in agreement with that obtained from ThT binding assays (10.8 h, Table 1). However, the wild-type PrP monomer band was observed when wild-type PrP was incubated with 100 μM Zn2 + for only half hour, indicating a lag time around 1 h (Fig. 2B and C), which is much shorter than that obtained from ThT binding assays (~ 10 h, Table 1). By contrast, a clear band corresponding to Sarkosyl-soluble PrPΔocta monomers was still observed when the octarepeats deletion mutant PrPΔocta was incubated in the absence of Zn2+ for 11.5 h (Fig. 2D) or in the presence of 100 μM Zn2 + for 27.5 h, a much longer time than 11.5 h (Fig. 2E). In agreement with the ThT binding assays, our Sarkosyl-soluble SDS-PAGE experiments indicated that 100 μM Zn2+ strongly inhibited the aggregation of PrPΔocta. Clearly, 100 μM Zn2+ dramatically promoted the formation of Sarkosyl-insoluble aggregates of wild-type PrP but significantly inhibited the formation of Sarkosylinsoluble aggregates of PrPΔocta and the fibrillization of wild-type PrP and its octarepeats deletion mutant. Our data also suggested that wild-type PrP formed amyloid fibrils in the absence of Zn2+, however, in the presence of 100 μM Zn2 +, wild-type PrP formed Sarkosylinsoluble aggregates rather than amyloid fibrils. Does zinc bind to ThT? Some control experiments are needed to clarify this point and accordingly interpret all of the ThT data. As shown in Fig. S2, the data were too small to be fitted, indicating that no binding between zinc and ThT was observed in the conditions used. Clearly, zinc did not bind to ThT during our ThT binding assays.

910

K. Pan et al. / Biochimica et Biophysica Acta 1854 (2015) 907–918

Fig. 1. The presence of Zn2+ altered fibrillization kinetics of human PrP–ThT binding assays. A-C, 11 μM wild-type PrP was incubated with 0–200 μM Zn2+ (open square, 0 μM; solid circle, 10 μM; solid square, 50 μM; solid triangle, 100 μM; and inverted solid triangle, 200 μM). D–F, 11 μM PrPΔocta was incubated with 0–100 μM Zn2+ (open square, 0 μM; and solid triangle, 100 μM). All kinetic experiments were repeated three times.

3.2. Effects of Zn2+ on morphology of human PrP aggregate samples TEM was used to study the morphology of human PrP aggregates formed in the absence and in the presence of 100 μM Zn2 + (Fig. 3). In the absence of Zn2 +, wild-type PrP formed amyloid fibrils with a long and branched structure after incubation for 18 h (Fig. 3A). In the presence of 100 μM Zn2 +, however, wild-type PrP formed short-rod aggregates and ellipsoidal particles rather than typical mature fibrils after incubation for 18 h (Fig. 3C). Unlike wild-type PrP, the octarepeats deletion mutant PrPΔocta formed some long amyloid fibrils and

abundant fibrils with a length of 200–400 nm either with or without 100 μM Zn2+ (Fig. 3B and D). High-resolution TEM was then used to investigate how Zn2+ affects the aggregate morphology and aggregation pathway of human PrP. Figs. 4 and 5 show time-course TEM images of human PrP samples at pH 7.0 after incubation with or without 100 μM Zn2+. After incubation for 4 h, some oligomers were observed for wild-type PrP in the presence of 100 μM Zn2+ (Fig. 4B), but no aggregates were observed for wildtype PrP in the absence of Zn2+ (Fig. 5B). Combining the results from ThT binding assays, Sarkosyl-soluble SDS-PAGE, and TEM, we concluded that aggregation of wild-type PrP in the absence of Zn2 + underwent four steps: amorphous aggregates (Fig. 4C–E), profibrils (Fig. 4F), mature fibrils (Fig. 4G), and at the end, fragmented fibrils (Fig. 4H). When the molar ratio of Zn2+ to PrP was 9:1, however, aggregation of wild-type PrP underwent another pathway in which wild-type PrP formed Sarkosyl-insoluble oligomers quickly (Fig. 5B and C) and then formed short-rod aggregates (Fig. 5E–H). Such short-rod aggregates are 50–100 nm long and pairs of parallel rods (Fig. 5H). Under such conditions, wild-type PrP did not form typical mature fibrils (Fig. 5B–H).

Table 1 Kinetic parameters of human PrP aggregation in the presence of Zn2+ as determined by ThT binding assays at 37 °C. Best-fit values of these kinetic parameters were derived from non-linear least squares modeling of a sigmoidal equation as described in the “Materials and methods” to the data plotted in Fig. 1. The buffer used was 5 mM MES buffer (pH 7.0). Errors shown are standard errors of the mean based on three independent experiments.

Fig. 2. Time-dependent SDS-PAGE analysis of Sarkosyl-soluble human PrP in the presence of Zn2+. A serial of 20 μl samples were taken in chronological order during fibril formation corresponding to ThT binding assays. 20 μl samples were added with 2.5 μl 20% Sarkosyl. The mixture was left at room temperature for 30 min, and then mixed with 5× loading buffer (without β-mercaptoethanol and SDS, and no heating) and separated by 15% SDS-PAGE. Gels were stained by Coomassie Blue R250. Wild-type PrP (A, B, and C) and PrPΔocta (D and E), both at 11 μM, incubated with 0–100 μM Zn2+ (A and D, 0 μM; and B, C, and E, 100 μM) were analyzed.

Human PrP

[Zn2+] μM

A + ct

Lag time h

k h−1

Wild-type PrP

0 10 100 0 100

59.5 ± 3.6 52.8 ± 3.9 ~11 50.5 ± 3.3 154 ± 41

10.8 ± 0.7 12.6 ± 1.4 ~10 10.4 ± 1.5 23.3 ± 1.6

1.39 ± 0.19 1.35 ± 0.05 ND 1.54 ± 0.25 1.30 ± 0.19

PrPΔocta

ND, not determined because the ThT fluorescence data in the present conditions was very small and could not be fitted to such a sigmoidal equation. ~, observed from the ThT fluorescence curves directly.

K. Pan et al. / Biochimica et Biophysica Acta 1854 (2015) 907–918

911

Fig. 3. Transmission electron micrographs of human PrP samples at pH 7.0 after incubation under different conditions. Negative-stain transmission electron micrographs of aggregates produced from 11 μM human PrP (wild-type PrP, A and C; and PrPΔocta, B and D) incubated with 0–100 μM Zn2+ (A and B, 0 μM; and C and D, 100 μM). The incubation time was chosen within a time range of the plateau of each kinetic curve of ThT fluorescence shown in Fig. 1. We used a 2% (w/v) uranyl acetate solution to stain the fibrils negatively. The scale bars represent 200 nm.

Fig. 4. Time-course transmission electron micrographs of human PrP samples at pH 7.0 after incubation without zinc. Negative-stain transmission electron micrographs of aggregates produced from 11 μM wild-type PrP incubated in the absence of Zn2+ for 0 (A), 4 (B), 8 (C), 9 h (D), 10.5 (E), 11 (F), 14 (G), and 20 h (H). We used a 2% (w/v) uranyl acetate solution to stain the fibrils negatively. The scale bars represent 200 nm.

912

K. Pan et al. / Biochimica et Biophysica Acta 1854 (2015) 907–918

Fig. 5. Time-course transmission electron micrographs of human PrP samples at pH 7.0 after incubation with zinc. Negative-stain transmission electron micrographs of aggregates produced from 11 μM wild-type PrP incubated in the presence of 100 μM Zn2+ for 0 (A), 4 (B), 6 (C), 8 (D), 12 (E), 17.5 (F), and 20 h (G and H). The black scale bars represent 200 nm, and the white scale bar represents 50 nm.

Different from wild-type PrP, the octarepeats deletion mutant PrPΔocta did form typical mature fibrils either with or without zinc (Fig. 3B and D). Our data demonstrated that zinc significantly changed the aggregation pathway of wild-type PrP but had no obvious effects on the aggregation pathway of PrPΔocta.

3.3. Thermodynamics of the binding of Zn2+ to human PrP ITC provides a direct route to the complete thermodynamic characterization of non-covalent, equilibrium interactions [37,41]. ITC profiles for the binding of Zn2 + to wild-type PrP and its octarepeats deletion

Fig. 6. ITC profiles for the binding of Zn2+ to wild-type PrP and its octarepeats deletion mutant under denaturing conditions and at 25.0 °C. The top panels (A and B) represent the raw data for sequential 10-μl injections of 2.0 mM Zn2+ into 20.0 μM wild-type PrP (A) and 20.0 μM PrPΔocta (B) in 5 mM MES buffer (pH 7.0) containing 2 M GdnHCl, respectively. The bottom panels (C and D) show the plots of the heat evolved (kilocalories) per mole of Zn2+ added, corrected for the heat of Zn2+ dilution, against the molar ratio of Zn2+ to human PrP. The data (solid squares) of panels C and D were best fitted to a three sequential binding sites model, and the solid lines represent the best fit. E represents the native gel electrophoresis of wild-type PrP. Lane 1, wild-type PrP in the absence of Zn2+. Lane 2, wild-type PrP in the presence of Zn2+ (the sample was taken from the calorimeter cell after the ITC experiments). Only one population of wild-type PrP, the monomeric wild-type PrP (M), was visible in lanes 1 and 2. The samples were mixed with 2× loading buffer and separated by 7% native PAGE. Gel was stained by Coomassie Blue R250.

K. Pan et al. / Biochimica et Biophysica Acta 1854 (2015) 907–918

913

Fig. 7. ITC profiles for the binding of Zn2+ to wild-type PrP and its octarepeats deletion mutant under non-denaturing conditions and at 25.0 °C. The top panels represent the raw data for sequential 10-μl injections of 2.0 mM Zn2+ into 25.8 μM wild-type PrP (A) and 26.0 μM PrPΔocta (B) in 5 mM MES buffer (pH 7.0), respectively. The bottom panels (C and D) show the plots of the heat evolved (kilocalories) per mole of Zn2+ added, corrected for the heat of Zn2+ dilution, against the molar ratio of Zn2+ to human PrP. The data (solid squares) of panel C were best fitted to a three sequential binding sites model, and the solid lines represent the best fit. However, the data (solid square) were too small to be fitted, indicating that no binding was observed in the conditions used (D).

mutant PrPΔocta at 25.0 °C are shown in Figs. 6 and 7. The top panels in Fig. 6 show representatively raw ITC curves resulting from the injections of Zn2+ into a solution of wild-type PrP (Fig. 6A) and PrPΔocta (Fig. 6B) in 5 mM MES buffer (pH 7.0) containing 2 M GdnHCl, and the top panels in Fig. 7 show representatively raw ITC curves resulting from the injections of Zn2+ into a solution of wild-type PrP (Fig. 7A) and PrPΔocta (Fig. 7B) in 5 mM MES buffer (pH 7.0) in the absence of GdnHCl. The titration curves show that Zn2+ binding to wild-type PrP and PrPΔocta was exothermic under denaturing conditions, resulting in negative peaks in the plots of power versus time. The bottom panels in Fig. 6 show the plots of the heat evolved per mole of Zn2+ added, corrected for the heat of Zn2+ dilution, against the molar ratio of Zn2+ to wildtype PrP (Fig. 6C) and PrPΔocta (Fig. 6D) in 5 mM MES buffer (pH 7.0) containing 2 M GdnHCl, and the bottom panels in Fig. 7 show the

plots of the heat evolved per mole of Zn2 + added, corrected for the heat of Zn2+ dilution, against the molar ratio of Zn2+ to wild-type PrP (Fig. 7C) and PrPΔocta (Fig. 7D) in 5 mM MES buffer (pH 7.0) in the absence of GdnHCl. The calorimetric data of wild-type PrP in the presence or absence of 2 M GdnHCl (Figs. 6C and 7C) and PrPΔocta in the presence of 2 M GdnHCl (Fig. 6D) were best fitted to a three sequential binding sites model. However, the data of PrPΔocta in the absence of GdnHCl (Fig. 7D) were too small to be fitted, indicating that no binding was observed under such non-denaturing conditions. The thermodynamic parameters for the binding of Zn2 + to the apo form of human PrP are summarized in Table 2. To our surprise, Zn2 + strongly bound to the apo form of wild-type human PrP with the first binding constant exceeding 107 M− 1 under denaturing conditions (Table 2), which is three orders of magnitude higher than that for the binding reaction of

Table 2 0 and n, were determined using a Thermodynamic parameters for the binding of Zn2+ to the apo form of human PrP as determined by ITC at 25.0 °C. Thermodynamic parameters, Kb, ΔbHm 0 0 ) and the standard molar binding entropy (Δb Sm ) for the binding reaction were calculated using three sequential binding sites model. The standard molar binding free energy (ΔbGm Eqs. (2) and (3), respectively. The buffer used was 5 mM MES buffer (pH 7.0) containing 2 M GdnHCl. Errors shown are standard errors of the mean. Human PrP

Wild-type PrP

PrPΔocta

Wild-type PrPa

PrPΔoctaa

n

1st 2nd 3rd 1st 2nd 3rd 1st 2nd 3rd –

Kb × 10−4

0 ΔbHm

0 ΔbGm

0 ΔbSm

M−1

kcal mol−1

kcal mol−1

cal mol−1 K−1

3720 ± 3200 1.15 ± 0.09 0.094 ± 0.011 5.7 ± 5.5 1.1 ± 1.5 0.095 ± 0.068 3.30 ± 0.84 0.274 ± 0.058 0.57 ± 0.12 NB

−2.00 ± 0.02 −8.82 ± 0.51 −24.8 ± 1.7 −0.67 ± 0.40 −1.1 ± 1.7 −10.4 ± 5.5 −4.65 ± 0.65 −4.5 ± 3.1 −7.9 ± 2.8 –

−10.32 ± 0.51 −5.54 ± 0.05 −4.05 ± 0.07 −6.48 ± 0.58 −5.53 ± 0.78 −4.06 ± 0.42 −6.16 ± 0.15 −4.69 ± 0.13 −5.12 ± 0.12 –

27.9 ± 1.8 −11.0 ± 1.9 −69.7 ± 6.1 19.5 ± 3.3 14.8 ± 8.2 −21 ± 20 −5.1 ± 2.7 0.79 ± 11 −9.4 ± 9.7 –

NB, no binding observed under such non-denaturing conditions. a The buffer used was 5 mM MES buffer (pH 7.0) in the absence of GdnHCl.

914

K. Pan et al. / Biochimica et Biophysica Acta 1854 (2015) 907–918

zinc with wild-type human PrP (Table 2) or wild-type mouse PrP under non-denaturing conditions [18,19]. Furthermore, zinc did weakly bind to human PrPΔocta with binding affinity around 104 M− 1 under denaturing conditions but did not bind to human PrPΔocta under non-denaturing conditions (Table 2). Our ITC data clearly indicated that at pH 7.0, zinc bound to the apo form of wild-type PrP mainly via interaction with its octarepeat region under denaturing conditions, thereby significantly changing its aggregation pathway. Interestingly, we found that the second and third binding constants for Zn2+-wildtype–PrP interaction (about 104 and 103 M−1, respectively) were similar to those for Zn2+–PrPΔocta interaction under denaturing conditions (Table 2). Therefore we conclude that under denaturing conditions wild-type PrP has at least three classes of zinc-binding sites, one major class of which in its octarepeat region and two minor classes of which outside the octarepeats. Our ITC data (Table 2) also indicated that at pH 7.0, zinc weakly bound to the apo form of PrPΔocta under denaturing conditions, thereby greatly extending PrPΔocta aggregation lag time and increasing plateau ThT fluorescence but not significantly changing its aggregation pathway. Because any aggregation of PrP could greatly affect interpretation of our ITC data, we need to demonstrate that the protein remains monomeric during ITC experiments that require rapid stirring. We then turned to native gel electrophoresis. As shown in Fig. 6E, we observed only one population of wild-type PrP, the monomeric wild-type PrP, but neither dimers nor oligomers, in the presence of Zn2+. Lane 1 serves as a standard where wild-type PrP was in the absence of Zn2+, resulting in a purely monomeric population (Fig. 6E). Clearly, the protein remained monomeric during our ITC experiments under denaturing conditions. We will use NMR spectroscopy to confirm such a conclusion in the future studies.

Fig. 8. Secondary structure changes of the apo form of human PrP incubated with Zn2+ monitored by far-UV CD at 25 °C. Far-UV CD spectra of 20 μM wild-type PrP (A) and PrPΔocta (B) with 100 μM Zn2+ (b) or without Zn2+ (a).

3.4. Effects of Zn2+ on the secondary structure of human PrP CD spectroscopy was used to detect secondary structure changes of the apo form of human PrP incubated with Zn2+. Fig. 8 shows far-UV CD spectra of wild-type PrP and PrPΔocta either with or without Zn2+. As shown in Fig. 8A and B, in the absence and presence of 100 μM Zn2+, the CD spectra of either wild-type PrP or PrPΔocta had double minima at 208 and 222 nm, indicative of predominant α-helical structure. Our CD data demonstrated that Zn2+ had no obvious effects on the secondary structure of wild-type PrP and its octarepeats deletion mutant. 3.5. Effects of Zn2+ on the secondary structure of human PrP aggregates CD spectroscopy was then used to detect secondary structure changes of human PrP aggregates incubated with Zn2+. Fig. 9 shows far-UV CD spectra of aggregates formed by either wild-type PrP or PrPΔocta with or without Zn2+. As shown in Fig. 9A, in the absence and presence of 100 μM Zn2+, the CD spectra of aggregates formed by wild-type PrP had a single minimum at 218 nm, indicative of predominant β-sheet structure [38]. In the absence of Zn2+, the CD spectrum of aggregates formed by wild-type PrP had a maximum at 204 nm (Fig. 9A). In the presence of 100 μM Zn2+, however, the CD spectrum of aggregates formed by wild-type PrP did not have such a maximum (Fig. 9A). Unlike wild-type PrP, the CD spectra of aggregates formed by the octarepeats deletion mutant PrPΔocta had a single minimum at 220 nm and a maximum at 204 nm either with or without 100 μM Zn2+ (Fig. 9B). Our CD data demonstrated that Zn2+ had obvious effects on the secondary structure of aggregates formed by wild-type PrP, but had no obvious effects on that of aggregates formed by PrPΔocta.

Fig. 9. Secondary structure changes of human PrP aggregates incubated with Zn2+ monitored by far-UV CD at 25 °C. Far-UV CD spectra of human PrP aggregates formed by 11 μM wild-type PrP (A) or PrPΔocta (B) with 100 μM Zn2+ (b) or without Zn2+ (a). The incubation time was chosen within a time range of the plateau of each kinetic curve of ThT fluorescence shown in Fig. 1.

K. Pan et al. / Biochimica et Biophysica Acta 1854 (2015) 907–918

As CD spectra of highly aggregated PrP could be distorted by light scattering effects, FTIR was then used to detect more precisely secondary structure changes of human PrP aggregates incubated with Zn2+. Fig. 10 shows FTIR spectra of aggregates formed by either wild-type PrP or PrPΔocta with or without Zn2 +. As shown in Fig. 10A, in the absence of Zn2+, the FTIR spectrum of aggregates formed by wild-type PrP had an amide I′ peak in the range of 1620–1630 cm−1, which is a characteristic for cross-β-sheet structure in amyloid fibrils [38,42]. In the presence of 100 μM Zn2+, however, the FTIR spectrum of aggregates formed by wild-type PrP did not have such an amide I′ peak but had another amide I′ peak in the range of 1640–1650 cm−1 (Fig. 10A), supporting the conclusion reached by TEM that wild-type PrP did not form typical mature fibrils in the presence of 100 μM Zn2 + (Fig. 3C). Unlike wild-type PrP, the FTIR spectra of aggregates formed by the octarepeats deletion mutant PrPΔocta did have an amide I′ peak in the range of 1620–1630 cm− 1 either with or without 100 μM Zn2 + (Fig. 10B), supporting the conclusion reached by TEM that PrPΔocta did form typical mature fibrils either with or without 100 μM Zn2 + (Fig. 3B and D). Our FTIR data once again demonstrated that Zn2+ had obvious effects on the secondary structure of aggregates formed by wild-type PrP, but had no obvious effects on that of aggregates formed by PrPΔocta. 3.6. Effects of Zn2+ on the PK-resistant features and conformation of human PrP aggregates PK digestion assays have been used to detect multiple prion strains associated with PrP conformations [25]. We thus used PK digestion assays to detect the significant differences in PK resistance activities

915

and conformation between wild-type PrP aggregates formed in the presence of zinc and those formed in the absence of zinc (Fig. 11). As shown in Fig. 11A, wild-type PrP aggregates formed in the absence of zinc did generate PK-resistant core fragment of 15–16-kDa and two short PK-resistant fragments (12- and 10-kDa bands) after PK digestion for 1 h, which are similar to those of human PrP reported previously [36, 38], but wild-type PrP aggregates formed in the presence of 100 μM Zn2 + generated PK-resistant core fragment of 15–16-kDa and only one short PK-resistant fragment (13–14-kDa band). Furthermore, the presence of 100 μM Zn2 + caused a remarkable decrease in optical density of the 15–16-kDa band, compared with that in the absence of zinc (Fig. 11A). Unlike wild-type PrP aggregates, PrPΔocta aggregates incubated with or without Zn2+ produced the same PK-resistant fragments, the 15–16-kDa, 13–14-kDa, and 12 kDa bands (Fig. 11B). Clearly, wild-type PrP aggregates had PK-resistant features and conformation in the presence of zinc different from those in the absence of zinc, but PrPΔocta aggregates had the same PK-resistant features and conformation in the presence of zinc as those in the absence of zinc. Taken together, our ITC, CD, FTIR, and PK digestion data demonstrated that Zn2+ strongly bound to the apo form of wild-type PrP with the first binding constant exceeding 107 M−1 under denaturing conditions and thus changed the structure and conformation of wild-type PrP aggregates remarkably, but weakly bound to PrPΔocta with binding affinity around 104 M−1 under denaturing conditions and had no obvious effects on the structure and conformation of PrPΔocta aggregates. We will use solidstate NMR experiments to obtain more insights into the zinc-bound aggregates of PrP in the future studies. 4. Discussion Prion hypothesis, also known as the protein-only hypothesis, has been proved by many landmark discoveries since its first proposal more than 40 years ago [1,43–47]. Although scientists can generate PrPSc in vitro de novo from noninfectious PrPC [46–52], and have also

Fig. 10. Secondary structure changes of human PrP aggregates incubated with Zn2+ monitored by FTIR at 25 °C. FTIR spectra of human PrP aggregates formed by 11 μM wild-type PrP (A) or PrPΔocta (B) with 100 μM Zn2+ (b) or without Zn2+ (a). The incubation time was chosen within a time range of the plateau of each kinetic curve of ThT fluorescence shown in Fig. 1.

Fig. 11. Concentration-dependent proteinase K-digestion assays of wild-type PrP (A) and its octarepeats deletion mutant (B) aggregates with zinc. Samples were treated with PK for 1 h at 37 °C at PK:PrP molar ratios as follows: 1:100 (lanes 3 and 7) and 1:50 (lanes 4 and 8). PK concentration: 0.4 μg/ml (lanes 3 and 7) and 0.8 μg/ml (lanes 4 and 8). The controls with zero protease were loaded in lanes 1 and 5. Protein molecular weight markers were loaded on lanes 2 and 6: restriction endonuclease Bsp98 I (25.0 kDa), β-lactoglobulin (18.4 kDa), and lysozyme (14.4 kDa). Aggregates were produced from wild-type PrP and its octarepeats deletion mutant in the presence of 100 μM Zn2+ (lanes 5–8), compared with in the absence of Zn2+ (lanes 1–4). The incubation time was chosen within a time range of the plateau of each kinetic curve of ThT fluorescence shown in Fig. 1. Protein fragments were separated by SDS-PAGE and detected by silver staining.

916

K. Pan et al. / Biochimica et Biophysica Acta 1854 (2015) 907–918

demonstrated that PrPSc is the main constituent of the pathogen of prion diseases [1,43–47], co-factors, which are not limited to proteins and include lipids, polyanions, and metal ions, can contribute to the concept of prion strains and prion generation pathway [1,3,25,45,47,53]. In this paper, we investigated how one of co-factors, Zn2+, changed the aggregation pathway and the structure and conformation of aggregates of wild-type human PrP and its octarepeats deletion mutant PrPΔocta. A variety of biophysical measurements – including ThT binding assays, Sarkosyl-soluble SDS-PAGE, TEM, ITC, CD, FTIR spectroscopy, and PK digestion assays – were utilized in the investigation of zinc–PrP binding properties. We found that aggregation of wild-type PrP underwent typical fibril formation pathway in the absence of Zn2+, forming amorphous aggregates, profibrils, mature fibrils, and fragmented fibrils step by step. In the presence of 100 μM Zn2 +, however, aggregation of wild-type PrP underwent a novel pathway, forming oligomers quickly and then short-rod aggregates. Different from wild-type PrP, the octarepeats deletion mutant PrPΔocta formed typical mature fibrils either with or without zinc. It has been reported that Zn2 + binds to the histidines within the octarepeat domain and drives an N-terminal to C-terminal tertiary interaction in wild-type mouse PrP [54]. Our ITC data indicate that under denaturing conditions Zn2+ did strongly bind to wild-type human PrP monomers with the first binding constant exceeding 107 M−1, but weakly bound to PrPΔocta with binding affinity around 104 M− 1. Therefore, Zn2 + bound to wild-type human PrP mainly via interaction with its octarepeat region. Our CD, FTIR, and PK digestion data demonstrated that the binding of Zn2+ to wild-type PrP monomers did change the structure and conformation of wild-type PrP aggregates remarkably, but zinc had no obvious effects on the structure and conformation of PrPΔocta aggregates. These data could explain why only wild-type PrP could form short-rod aggregates with unique structure and conformation in the presence of Zn2 +, but not its octarepeats deletion mutant. For the study of interaction between Zn2 + and PrP, some of the researches have been focused on prion peptide [55,56], but other researchers have noticed the interaction between zinc and wild-type mouse PrP [18,19]. They have found that Zn2 + binds to wild-type mouse PrP with binding constants around 104 M−1 [18,19]. By using ITC and other biophysical methods, we found that under denaturing conditions, Zn2 + strongly bound to wild-type human PrP with the first binding constant exceeding 107 M−1 mainly via interaction with

its octarepeat region, thereby significantly changing its aggregation pathway. We also found that zinc did weakly bind to PrPΔocta under denaturing conditions, thereby delaying the lag time of the fibrillization of PrPΔocta but not significantly changing its aggregation pathway. The role of zinc in prion-like amyloid β aggregation has been previously investigated [31–33,57–59]. This paper reports a timely biophysical study on the effect of zinc binding to PrP and the data presented suggests that zinc changes the aggregation pathway and conformation of wild-type PrP aggregates by interacting with the octarepeat region, providing important clues on how this process is regulated by chemical cues in the cellular environment. The major finding of our study is that Zn2 + did induce wild-type PrP to form Sarkosyl-insoluble oligomers quickly, which could be converted into structured short-rod aggregates. This finding indicated that zinc mediated wild-type PrP aggregation in an unusual way, which not only dramatically promoted the formation of Sarkosyl-insoluble oligomers, but also changed the aggregation pathway and the conformation of aggregates of human PrP. We also demonstrated that zinc affected aggregation of wild-type PrP and its octarepeats deletion mutant in a different way via strongly interacting with wild-type PrP but weakly interacting with PrPΔocta under denaturing conditions. Based on our own data, we propose a hypothetical model to show how zinc influences human PrP aggregation (Fig. 12). Aggregation of wild-type PrP in the absence of Zn2+ undergoes four steps: amorphous aggregates, profibrils, mature fibrils, and at the end, fragmented fibrils. In the presence of Zn2 +, however, aggregation of wild-type PrP undergoes another pathway in which wild-type PrP forms oligomers quickly and then forms short-rod aggregates. Unlike wild-type PrP, the octarepeats deletion mutant PrPΔocta forms typical mature fibrils either with or without Zn2 +. Such differences could be caused by the major interaction between Zn2 + and the octarepeat region of human PrP, and such an interaction could facilitate the metal-driven interdomain interaction between the C-terminal globular domain and the N-terminal flexible disordered domain of the protein. Our data demonstrate that zinc significantly changed the aggregation pathway and the structure and conformation of wild-type PrP aggregates mainly via interaction with its octarepeat region. Our findings could explain how zinc modifies pathological PrP conformation associated with prion diseases and link abnormal aggregation of prion protein modified by zinc to pathogenesis of prion diseases and prion strain diversity. It is

Fig. 12. A hypothetical model to show how zinc influences human PrP aggregation. Aggregation of wild-type PrP in the absence of Zn2+ undergoes four steps: amorphous aggregates, profibrils, mature fibrils, and at the end, fragmented fibrils. In the presence of Zn2+, however, aggregation of wild-type PrP undergoes another pathway in which wild-type PrP forms oligomers quickly and then forms short-rod aggregates. Different from wild-type PrP, the octarepeats deletion mutant PrPΔocta forms typical mature fibrils either with or without Zn2+. Such differences could be caused by the interaction between Zn2+ and the octarepeat region of human PrP, and such an interaction could facilitate the metal-driven interdomain interaction between the C-terminal globular domain and the N-terminal flexible disordered domain.

K. Pan et al. / Biochimica et Biophysica Acta 1854 (2015) 907–918

worth mentioning that zinc has also been found to modify the conformation and aggregation pathways of other amyloid-forming proteins/ peptides, including amyloid β [58,59], Tau protein [37,59], and superoxide dismutase 1 [59,60]. Transparency document The Transparency document associated with this article can be found in the online version. Author contributions Planned the experiments: YL. Performed the experiments: KP C-WY. Analyzed the data: KP YL. Contributed reagents/materials: JC. Wrote the paper: KP YL. Acknowledgements We sincerely thank Prof. Geng-Fu Xiao (Wuhan Institute of Virology, Chinese Academy of Sciences) for his kind gift of the human PrPC plasmid. We thank Dr. Li Li in this college and Dr. Zhi-Ping Zhang (Wuhan Institute of Virology, Chinese Academy of Sciences) for their technical assistances on TEM, and Dr. Fei Liu (Analytical and Testing Center, Wuhan University) for her technical assistance on FTIR. This study was supported by the National Key Basic Research Foundation of China Grants 2013CB910702 and 2012CB911003, the National Natural Science Foundation of China Grants 31370774 and 31170744, and the Fundamental Research Funds for the Central Universities of China (410500058 and 410500064). Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.bbapap.2015.04.020. References [1] S.B. Prusiner, Prions, Proc. Natl. Acad. Sci. U. S. A. 95 (1998) 13363–13383. [2] B. Caughey, G.S. Baron, B. Chesebro, M. Jeffrey, Getting a grip on prions: oligomers, amyloids, and pathological membrane interactions, Annu. Rev. Biochem. 78 (2009) 177–204. [3] A. Aguzzi, C. Sigurdson, M. Heikenwaelder, Molecular mechanisms of prion pathogenesis, Annu. Rev. Pathol. 3 (2008) 11–40. [4] A. Aguzzi, F. Baumann, J. Bremer, The prion's elusive reason for being, Annu. Rev. Neurosci. 31 (2008) 439–477. [5] R. Diaz-Espinoza, C. Soto, High-resolution structure of infectious prion protein: the final frontier, Nat. Struct. Mol. Biol. 19 (2012) 370–377. [6] A. Rana, D. Gnaneswari, S. Bansal, B. Kundu, Prion metal interaction: is prion pathogenesis a cause or a consequence of metal imbalance? Chem. Biol. Interact. 181 (2009) 282–291. [7] N. Singh, D. Das, A. Singh, M.L. Mohan, Prion protein and metal interaction: physiological and pathological implications, Curr. Issues Mol. Biol. 12 (2010) 99–107. [8] C. Treiber, A.R. Thompsett, R. Pipkorn, D.R. Brown, G. Multhaup, Real-time kinetics of discontinuous and highly conformational metal-ion binding sites of prion protein, J. Biol. Inorg. Chem. 12 (2007) 711–720. [9] M.W. Brazier, P. Davies, E. Player, F. Marken, J.H. Viles, D.R. Brown, Manganese binding to the prion protein, J. Biol. Chem. 283 (2008) 12831–12839. [10] A.R. Thompsett, S.R. Abdelraheim, M. Daniels, D.R. Brown, High affinity binding between copper and full-length prion protein identified by two different techniques, J. Biol. Chem. 280 (2005) 42750–42758. [11] G.S. Jackson, I. Murray, L.L. Hosszu, N. Gibbs, J.P. Waltho, A.R. Clarke, J. Collinge, Location and properties of metal-binding sites on the human prion protein, Proc. Natl. Acad. Sci. U. S. A. 98 (2001) 8531–8535. [12] D. McKenzie, J. Bartz, J. Mirwald, D. Olander, R. Marsh, J. Aiken, Reversibility of scrapie inactivation is enhanced by copper, J. Biol. Chem. 273 (1998) 25545–25547. [13] K. Qin, D.S. Yang, Y. Yang, M.A. Chishti, L.J. Meng, H.A. Kretzschmar, C.M. Yip, P.E. Fraser, D. Westaway, Copper(II)-induced conformational changes and protease resistance in recombinant and cellular PrP. Effect of protein age and deamidation, J. Biol. Chem. 275 (2000) 19121–19131. [14] E. Quaglio, R. Chiesa, D.A. Harris, Copper converts the cellular prion protein into a protease-resistant species that is distinct from the scrapie isoform, J. Biol. Chem. 276 (2001) 11432–11438.

917

[15] E.M. Sigurdsson, D.R. Brown, M.A. Alim, H. Scholtzova, R. Carp, H.C. Meeker, F. Prelli, B. Frangione, T. Wisniewski, Copper chelation delays the onset of prion disease, J. Biol. Chem. 278 (2003) 46199–46202. [16] O.V. Bocharova, L. Breydo, V.V. Salnikov, I.V. Baskakov, Copper(II) inhibits in vitro conversion of prion protein into amyloid fibrils, Biochemistry 44 (2005) 6776–6787. [17] N. Hijazi, Y. Shaked, H. Rosenmann, T. Ben-Hur, R. Gabizon, Copper binding to PrPC may inhibit prion disease propagation, Brain Res. 993 (2003) 192–200. [18] E.D. Walter, D.J. Stevens, M.P. Visconte, G.L. Millhauser, The prion protein is a combined zinc and copper binding protein: Zn2+ alters the distribution of Cu2+ coordination modes, J. Am. Chem. Soc. 129 (2007) 15440–15441. [19] P. Davies, F. Marken, S. Salter, D.R. Brown, Thermodynamic and voltammetric characterization of the metal binding to the prion protein: insights into pH dependence and redox chemistry, Biochemistry 48 (2009) 2610–2619. [20] J.H. Weiss, S.L. Sensi, J.Y. Koh, Zn2+: a novel ionic mediator of neural injury in brain disease, Trends Pharmacol. Sci. 21 (2000) 395–401. [21] N.T. Watt, N.M. Hooper, The prion protein and neuronal zinc homeostasis, Trends Biochem. Sci. 28 (2003) 406–410. [22] W.S. Perera, N.M. Hooper, Ablation of the metal ion-induced endocytosis of the prion protein by disease-associated mutation of the octarepeat region, Curr. Biol. 11 (2001) 519–523. [23] L.R. Brown, D.A. Harris, Copper and zinc cause delivery of the prion protein from the plasma membrane to a subset of early endosomes and the Golgi, J. Neurochem. 87 (2003) 353–363. [24] N.T. Watt, D.R. Taylor, T.L. Kerrigan, H.H. Griffiths, J.V. Rushworth, I.J. Whitehouse, N.M. Hooper, Prion protein facilitates uptake of zinc into neuronal cells, Nat. Commun. 3 (2012) 1134. [25] J.D. Wadsworth, A.F. Hill, S. Joiner, G.S. Jackson, A.R. Clarke, J. Collinge, Strain-specific prion-protein conformation determined by metal ions, Nat. Cell Biol. 1 (1999) 55–59. [26] 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, Copper and zinc binding modulates the aggregation and neurotoxic properties of the prion peptide PrP106-126, Biochemistry 40 (2001) 8073–8084. [27] J.R. Brender, K. Hartman, R.P.R. Nanga, N. Popovych, R. de la Salud Bea, S. Vivekanandan, E.N.G. Marsh, A. Ramamoorthy, Role of zinc in human islet amyloid polypeptide aggregation, J. Am. Chem. Soc. 132 (2010) 8973–8983. [28] S. Salamekh, J.R. Brender, S.J. Hyung, R.P.R. Nanga, S. Vivekanandan, B.T. Ruotolo, A. Ramamoorthy, A two-site mechanism for the inhibition of IAPP amyloidogenesis by zinc, J. Mol. Biol. 410 (2011) 294–306. [29] J.R. Brender, J. Krishnamoorthy, G.M.L. Messina, A. Deb, S. Vivekanandan, C. La Rosa, J.E. Penner-Hahn, A. Ramamoorthy, Zinc stabilization of prefibrillar oligomers of human islet amyloid polypeptide, Chem. Commun. 49 (2013) 3339–3341. [30] J.R. Brender, S. Salamekh, A. Ramamoorthy, Membrane disruption and early events in the aggregation of the diabetes related peptide IAPP from a molecular perspective, Acc. Chem. Res. 45 (2012) 454–462. [31] A.S. DeToma, S. Salamekh, A. Ramamoorthy, M.H. Lim, Misfolded proteins in Alzheimer's disease and type II diabetes, Chem. Soc. Rev. 41 (2012) 608–621. [32] H.R. Patel, A.S. Pithadia, J.R. Brender, C.A. Fierke, A. Ramamoorthy, In search of aggregation pathways of IAPP and other amyloidogenic proteins: finding answers through NMR spectroscopy, J. Phys. Chem. Lett. 5 (2014) 1864–1870. [33] S.A. Kotler, P. Walsh, J.R. Brender, A. Ramamoorthy, Differences between amyloid-β aggregation in solution and on the membrane: insights into elucidation of the mechanistic details of Alzheimer's disease, Chem. Soc. Rev. 43 (2014) 6692–6700. [34] O.V. Bocharova, L. Breydo, A.S. Parfenov, V.V. Salnikov, I.V. Baskakov, In vitro conversion of full-length mammalian prion protein produces amyloid form with physical properties of PrPSc, J. Mol. Biol. 346 (2005) 645–659. [35] M. Chattopadhyay, A. Durazo, S.H. Sohn, C.D. Strong, E.B. Gralla, J.P. Whitelegge, J.S. Valentine, Initiation and elongation in fibrillation of ALS-linked superoxide dismutase, Proc. Natl. Acad. Sci. U. S. A. 105 (2008) 18663–18668. [36] Z. Zhou, J.B. Fan, H.L. Zhu, F. Shewmaker, X. Yan, X. Chen, J. Chen, G.F. Xiao, L. Guo, Y. Liang, Crowded cell-like environment accelerates the nucleation step of amyloidogenic protein misfolding, J. Biol. Chem. 284 (2009) 30148–30158. [37] Z.Y. Mo, Y.Z. Zhu, H.L. Zhu, J.B. Fan, J. Chen, Y. Liang, Low micromolar zinc accelerates the fibrillization of human Tau via bridging of Cys-291 and Cys-322, J. Biol. Chem. 284 (2009) 34648–34657. [38] Z. Zhou, X. Yan, K. Pan, J. Chen, Z.S. Xie, G.F. Xiao, F.Q. Yang, Y. Liang, Fibril formation of the rabbit/human/bovine prion proteins, Biophys. J. 101 (2011) 1483–1492. [39] K.E. van Holde, W.E. Johnson, P.S. Ho, Principles of Physical Biochemistry, Pearson Prentice Hall, Upper Saddle River, New Jersey, 2006. 91–93. [40] H. Naiki, K. Higuchi, M. Hosokawa, T. Takeda, Fluorometric determination of amyloid fibrils in vitro using the fluorescent dye, thioflavin T1, Anal. Biochem. 177 (1989) 244–249. [41] Y. Liang, F. Du, S. Sanglier, B.R. Zhou, Y. Xia, A. Van Dorsselaer, C. Maechling, M.C. Kilhoffer, J. Haiech, Unfolding of rabbit muscle creatine kinase induced by acid. A study using electrospray ionization mass spectrometry, isothermal titration calorimetry, and fluorescence spectroscopy, J. Biol. Chem. 278 (2003) 30098–30105. [42] G. Zandomeneghi, M.R. Krebs, M.G. McCammon, M. Fandrich, FTIR reveals structural differences between native beta-sheet proteins and amyloid fibrils, Protein Sci. 13 (2004) 3314–3321. [43] J.S. Griffith, Self-replication and scrapie, Nature 215 (1967) 1043–1044. [44] S.B. Prusiner, Novel proteinaceous infectious particles cause scrapie, Science 216 (1982) 136–144. [45] C. Soto, Prion hypothesis: the end of the controversy? Trends Biochem. Sci. 36 (2011) 151–158. [46] G. Legname, I.V. Baskakov, H.O. Nguyen, D. Riesner, F.E. Cohen, S.J. DeArmond, S.B. Prusiner, Synthetic mammalian prions, Science 305 (2004) 673–676.

918

K. Pan et al. / Biochimica et Biophysica Acta 1854 (2015) 907–918

[47] F. Wang, X. Wang, C.G. Yuan, J. Ma, Generating a prion with bacterially expressed recombinant prion protein, Science 327 (2010) 1132–1135. [48] M. Klingeborn, B. Race, K.D. Meade-White, B. Chesebro, Lower specific infectivity of protease-resistant prion protein generated in cell-free reactions, Proc. Natl. Acad. Sci. U. S. A. 108 (2011) E1244–E1253. [49] R. Atarashi, R.A. Moore, V.L. Sim, A.G. Hughson, D.W. Dorward, H.A. Onwubiko, S.A. Priola, B. Caughey, Ultrasensitive detection of scrapie prion protein using seeded conversion of recombinant prion protein, Nat. Methods 4 (2007) 645–650. [50] J. Bieschke, P. Weber, N. Sarafoff, M. Beekes, A. Giese, H. Kretzschmar, Autocatalytic self-propagation of misfolded prion protein, Proc. Natl. Acad. Sci. U. S. A. 101 (2004) 12207–12211. [51] J.I. Kim, I. Cali, K. Surewicz, Q. Kong, G.J. Raymond, R. Atarashi, B. Race, L. Qing, P. Gambetti, B. Caughey, W.K. Surewicz, Mammalian prions generated from bacterially expressed prion protein in the absence of any mammalian cofactors, J. Biol. Chem. 285 (2010) 14083–14087. [52] N. Makarava, G.G. Kovacs, O. Bocharova, R. Savtchenko, I. Alexeeva, H. Budka, R.G. Rohwer, I.V. Baskakov, Recombinant prion protein induces a new transmissible prion disease in wild-type animals, Acta Neuropathol. 119 (2010) 177–187. [53] N.R. Deleault, R.W. Lucassen, S. Supattapone, RNA molecules stimulate prion protein conversion, Nature 425 (2003) 717–720. [54] A.R. Spevacek, E.G. Evans, J.L. Miller, H.C. Meyer, J.G. Pelton, G.L. Millhauser, Zinc drives a tertiary fold in the prion protein with familial disease mutation sites at the interface, Structure 21 (2013) 236–246.

[55] F. Stellato, A. Spevacek, O. Proux, V. Minicozzi, G. Millhauser, S. Morante, Zinc modulates copper coordination mode in prion protein octa-repeat subdomains, Eur. Biophys. J. 40 (2011) 1259–1270. [56] E. Gaggelli, F. Bernardi, E. Molteni, R. Pogni, D. Valensin, G. Valensin, M. Remelli, M. Luczkowski, H. Kozlowski, Interaction of the human prion PrP(106–126) sequence with copper(II), manganese(II), and zinc(II): NMR and EPR studies, J. Am. Chem. Soc. 127 (2005) 996–1006. [57] M.G. Savelieff, Y. Liu, R.R. Senthamarai, K.J. Korshavn, H.J. Lee, A. Ramamoorthy, M.H. Lim, A small molecule that displays marked reactivity toward copper- versus zincamyloid-β implicated in Alzheimer's disease, Chem. Commun. 50 (2014) 5301–5303. [58] D. Noy, I. Solomonov, O. Sinkevich, T. Arad, K. Kjaer, I. Sagi, Zinc–amyloid β interactions on a millisecond time-scale stabilize non-fibrillar Alzheimer-related species, J. Am. Chem. Soc. 130 (2008) 1376–1383. [59] S.S. Leal, H.M. Botelho, C.M. Gomes, Metal ions as modulators of protein conformation and misfolding in neurodegeneration, Coord. Chem. Rev. 256 (2012) 2253–2270. [60] S.S. Leal, J.S. Cristóvão, A. Biesemeier, I. Cardoso, C.M. Gomes, Aberrant zinc binding to immature conformers of metal-free copper–zinc superoxide dismutase triggers amorphous aggregation, Metallomics 7 (2015) 333–346.