DNA induces folding in α-synuclein: Understanding the mechanism using chaperone property of osmolytes

ABB Archives of Biochemistry and Biophysics 464 (2007) 57–69 www.elsevier.com/locate/yabbi

DNA induces folding in a-synuclein: Understanding the mechanism using chaperone property of osmolytes Muralidhar L. Hegde, K.S.J. Rao

*

Department of Biochemistry and Nutrition, Central Food Technological Research Institute, Mysore 570020, India Received 20 February 2007, and in revised form 24 March 2007 Available online 26 April 2007

Abstract a-Synuclein conformational modulation leading to fibrillation has been centrally implicated in Parkinson’s disease. Previously, we have shown that a-synuclein has DNA binding property. In the present study, we have characterized the effect of DNA binding on the conformation and fibrillation kinetics of a-synuclein. It was observed that single-stranded circular DNA induce a-helix conformation in a-synuclein while plasmid supercoiled DNA has dual effect inducing a partially folded conformation and a-helix under different experimental conditions. Interestingly, a-synuclein showed a specificity for GC* nucleotide sequence in its binding ability to DNA. The aggregation kinetics data showed that DNA which induced partially folded conformation in a-synuclein promoted the fibrillation while DNA which induced a-helix delayed the fibrillation, indicating that the partially folded intermediate conformation is critical in the aggregation process. Further, the mechanism of DNA-induced folding/aggregation of a-synuclein was studied using effect of osmolytes on a-synuclein as a model system. Among the five osmolytes used, Glycerol, trimethylamine-N-oxide, Betaine, and Taurine induced partially folded conformation and in turn enhanced the aggregation of a-synuclein. The ability of DNA and osmolytes in inducing conformational transition in a-synuclein, indicates that two factors are critical in modulating a-synuclein folding: (i) electrostatic interaction as in the case of DNA, and (ii) hydrophobic interactions as in the case of osmolytes. The property of DNA inducing a-helical conformation in a-synuclein and inhibiting the fibrillation may be of significance in engineering DNA-chip based therapeutic approaches to PD and other amyloid disorders.  2007 Elsevier Inc. All rights reserved. Keywords: Parkinson’s disease; a-Synuclein; Protein folding; a-Synuclein aggregation; Omolytes; Neurodegeneration

a-Synuclein is a small (14 kDa), soluble, intracellular, highly conserved protein, of unknown function and it is abundant in various regions of the brain [1–4]. a-Synuclein was characterized by the presence of acidic stretches within the C-terminal region and a repetitive motif, KTKEGV, in the first 93 residues [4,5]. Such a periodicity is characteristic of the amphipathic helices of apolipoproteins [6,7]. Structurally, purified a-synuclein is a natively unfolded protein [8–11]. This lack of folding has been shown to correlate with the specific combinations of low overall hydrophobicity and large net charge [12–14]. Few studies have highlighted the importance on being unfolded/disordered and it has been proposed that the increased intrinsic plasticity represents an important pre-requisite for effective molec*

Corresponding author. Fax: +91 821 2517233. E-mail address: [email protected] (K.S.J. Rao).

0003-9861/$ - see front matter  2007 Elsevier Inc. All rights reserved. doi:10.1016/j.abb.2007.03.042

ular recognition [15–19]. Proteins such as a-synuclein, which lack rigid 3D structure under physiological conditions in vitro, exist instead as dynamic ensembles of interconverting structures. These intrinsically disordered regions serve as a valid identification for recognition, regulation and cell signaling [20]. The role of intrinsic disorder in protein interaction networks of hub proteins has been reviewed [21]. They propose a-synuclein as a hub protein interacting with parkin, s, Ab etc., through intrinsic disordered regions. Deposition of aggregated forms of a-synuclein in neuronal or glial cytoplasm is pathological hallmark of Parkinson’s disease (PD)1 and other neurological disorders like 1 Abbreviations used: PD, Parkinson’s disease; TMAO, trimethylamineN-oxide; sscDNA, single strand circular DNA; dscDNA, double strand circular DNA; scDNA, supercoiled DNA; ANS, 1-anilinonaphthaleine-8sulphonic acid.

58

M.L. Hegde, K.S.J. Rao / Archives of Biochemistry and Biophysics 464 (2007) 57–69

dementia with Lewy bodies, Lewy body variant of Alzheimer’s disease, and multiple system atrophy [22,23]. In vitro a-synuclein readily assembles into fibrils, with morphologies and staining characteristics similar to those of fibrils extracted from PD affected brain [10,24–33]. The kinetics of fibrillation occurs via a nucleation dependent mechanism [26,34], with the critical primary stage being formation of a partially folded intermediate [10]. a-Synuclein folding and fibrillation have been found to be promoted on binding to long chain fatty acids [35] and also upon its interaction with lipid droplets [36]. It was also shown that membrane interactions induce a large conformational change from random coil to a-helix in a-synuclein and these interactions may be physiologically important [37]. On the basis of these observations, it has been assumed that a-synuclein may exist in two structurally different isoforms in vivo: a helix-rich, membrane-bound form and a disordered, cytosolic form, with the membrane-bound a-synuclein generating nuclei that seed the aggregation of the more abundant cytosolic form [33,38]. In the presence of several divalent and trivalent metal ions [39] or several common pesticides [40,41] a-synuclein has been shown to adopt a partially folded conformation, which is critical for fibrillation. These studies indicate that electrostatic interactions with a-synuclein resulting in neutralization of net negative charge would favor folding of the protein. Recently, it was shown that double-stranded DNA promotes aggregation of a-synuclein [42]. We have recently shown that a-synuclein has DNA binding property [8,43]. Further, it was shown that the binding of a-synuclein to synthetic and natural membranes is accompanied by a dramatic increase in a-helical content [37,44]. It was also observed that naturally occurring osmolytes-like trimethylamine-N-oxide (TMAO) causes natively unfolded a-synuclein to fold in a manner described for this protein at high temperatures, and low pH [45]. These results indicate that the osmophobic effect must be added to the previously described factors that induce the transformation of natively unfolded a-synuclein in to partially folded intermediate [45]. In a quest to understand the DNA-induced conformational transition in a-synuclein, we have used chaperon properties of five osmolytes viz. glycerol, Betaine, Taurine, TMAO, Sarcosine for their ability to induce folding in asynuclein, as a model system. Experimental procedures

Glycerol, and Copper grids (300 mesh size) for electron microscopy, were purchased from Sigma Chemical Company USA. Uranyl acetate was purchased from B.D.H. Laboratory Chemicals Division. The stock solutions of 7 M Sarcosine, 5 M each of Betaine, and TMAO, and Taurine were prepared in distilled water depending on the solubility of respective osmolytes.

Circular dichroism studies The CD spectra (190–260 nm) were recorded for a-synuclein in the presence/absence of increasing concentrations of various DNA (supercoiled DNA, kDNA, single and double strand circular DNA etc.) in 0.01 M Hepes buffer (pH 7.4) on a JASCO J-700 Spectropolarimeter. Each spectrum was the average of four repetitions. The CD contributions from DNA alone was corrected in the a-synuclein–DNA complex spectra. Five to ten micromolar a-synuclein was used for each measurement. The effect of osmolytes, Sarcosine, TMAO, Taurine, Betaine, and Glycerol on a-synuclein secondary conformation were also determined by CD. The protein was diluted to a final concentration of 5 lM in Hepes buffer, pH 7.0. The protein solutions were combined with 1 · 106 to 1 M Sarcosine, TMAO, Taurine, and Betaine and 1–50% by volume of Glycerol. Spectra were collected after 15 min of equilibrium period at room temperature. Spectra were acquired on a JASCO J-700 spectropolarimeter in a 0.1-cm path length cell over a wavelength range of 200–260 and 190–260 nm wherever possible, with a 1.0-nm bandwidth, 0.1-nm resolution, 4-s response time and 20-nm/min scan rate. All spectra were corrected by substraction from buffer and respective osmolyte. The deconvolution or secondary structure analysis of CD spectra of asynuclein was carried out using j-fit software.

Intrinsic fluorescence Tyrosine intrinsic fluorescence spectra were collected on a HITACHI 2000 spectrofluorimeter in semimicro quartz cuvette with a 1-cm excitation light path. The light source was a 150 W xenon lamp. For tyrosine intrinsic fluorescence a-synuclein containing solution was excited at 275 nm and emission monitored in the range from 290 to 350 nm. The maximum emission was observed at 306 nm. a-Synuclein concentrations for intrinsic fluorescence measurements were kept at 2 lM. a-Synuclein contains four tyrosine residues and there are no tryptophan. Hence the tyrosine fluorescence has been used to monitor folding of a-synuclein [39]. In the present investigation, the tyrosine intrinsic fluorescence was used to understand the osmolyte-induced folding in a-synuclein.

Acrylamide fluorescence quenching Acrylamide quenching studies of the intrinsic fluorescence of a-synuclein were performed by adding aliquots from a stock solution of the quencher (acrylamide) into a cuvette containing the protein (a-synuclein) solution. Fluorescence intensities were corrected for dilution effects. Fluorescence quenching data were analyzed using the general form of the Stern–Volmer equation, taking in to account not only the dynamic but also static quenching [46] as shown in the equation below: I o =I ¼ ð1 þ K sv ½QÞev½Q where, Io and I are the fluoroscence intensities in the absence and presence of quencher; Ksv is the dynamic quenching constant; V is a static quenching constant; and [Q] is the quencher concentration.

Materials ANS (1-anilinonaphthaleine-8-sulphonic acid) binding a-Synuclein was purchased from rPeptides, USA, the purity of asynuclein was confirmed by Gel filteration and mass-spectrometric analysis. Supercoiled DNA pUC 18 plasmid (Cesium chloride purified, 90% supercoiled structure), kDNA, Calf-thymus DNA, single-stranded circular (ssc)DNA (MP3), double-stranded circular (dsc)DNA (MP3), Tris and Hepes buffers were purchased from Bangalore Genei, India. Poly d(GC)Æd(GC) and poly d(AT)Æd(AT), Sarcosine, TMAO, Taurine, Betaine,

ANS is frequently used to demonstrate the presence of partially folded conformations of proteins, characterized by the presence of solventexposed hydrophobic clusters. This is because ANS binds to solventexposed hydrophobic clusters, resulting in a considerable increase in the ANS fluorescence intensity and in a pronounced blue shift of the fluorescence emission maximum.

M.L. Hegde, K.S.J. Rao / Archives of Biochemistry and Biophysics 464 (2007) 57–69 We measured the change of ANS fluorescence to monitor the gain of structure in a-synuclein in terms of solvent-exposed hydrophobic clusters upon binding with the osmolytes: Sarcosine, TMAO, Betaine, Taurine, and Glycerol. A stock solution of 1 mM ANS was prepared in distilled water. The [ANS]/[a-synuclein] ratio in all experiments was kept equal to 5. Fluorescence measurements were performed at 25 C in 1 mL semimicro quartz cuvette. Emission spectra were recorded from 400 to 600 nm with excitation at 350 nm. The protein concentration was 2 lM.

Trypsin cleavage and Tricine-SDS–PAGE Proteolytic cleavage of proteins has been used as a probe of protein conformation and stability [47,48]. Limited proteolysis of a-synuclein was implemented earlier for analysis of interaction of a-synuclein with SDS micelles [49]. In the present study, trypsin was used at a final concentration of 0.1 mg/mL. The effect of the presence of osmolytes on tryptic digestion was studied by 15% Tricine-SDS-PAGE.

Thioflavin T fluorescence for fibril formation assay Preparation of a-synuclein incubations for aggregation studies a-Synuclein at 50 lM concentration was dissolved/diluted in 10 mM Tris–HCl, pH 7.4 and incubated at 37 C with agitation/stirring using micro-magnetic bars (teflon coated) in glass vials.

Thio T fluorescence A 100 lM aqueous solution of Thio T was prepared and filtered through 0.2-lm polyether sulfone filter. At various time intervals, aliquots of the a-synuclein incubations were diluted to 10 lM in 50 mM glycine–NaOH buffer, pH 9.0. Assay solutions contained 10 lM Thio T and a-synuclein at a concentration of 5 lM in 50 mM glycine–NaOH buffer, pH 9.0.

59

Transmission electron microscopy of a-synuclein aggregates a-Synuclein aggregates were observed under JEOL 1010 TEM. A drop (5 lL) of a-synuclein solution was placed on the carbon-coated copper grid and was allowed to dry in air for 30 min. A second drop was applied after blotting the first drop with filter paper. This process was repeated 4–6 times after which the grids were negative stained by adding a drop of 1% uranyl acetate (pH 5.1) on the grid and blotted with a filter paper after 10 s. The grids were completely dried so as to avoid moisture for TEM examination [50]. The samples were examined with a JEOL 1010 Transmission Electron Microscope and the images were photographed.

Results scDNA induces ordered conformation in a-synuclein Circular dichroism spectroscopy (CD) was used to determine the effects of DNA binding on the secondary structure of a-synuclein. It was observed that scDNA caused a biphasic conformational transition in a-synuclein. Natively, a-synuclein is in random-coil conformation. On immediate mixing of the DNA and a-synuclein at room temperature a partial folding was induced in a-synuclein (Fig. 1A) while a-helix conformation was formed on long term incubation at 4 C (Fig. 1B). The CD spectrum of native a-synuclein was characterized by a strong negative CD band in the 195- to 200-nm region, indicative of a disordered or random-coil

Fig. 1. DNA-induced a-helix in a-synuclein. The CD measurements were carried out at 25 C. (A) Supercoiled (sc) DNA-induced partially folded conformation in a-synuclein on 15-min incubation. The CD spectra of 5 lM a-synuclein was measured in the absence or presence of scDNA (1–10 lg) and allowed to interact for 15 min. (—), a-synuclein; (.....), a-synuclein + 1 lg scDNA; (..—..), a-synuclein + 5 lg scDNA; (- - -), a-synuclein+10 lg scDNA. (B) scDNA-induced a-helix in a-synuclein on long term incubation. (...), a-synuclein (5 lM) in 1 mM Hepes, pH 7.4; (—), a-synuclein + scDNA (10 lg) incubated for 30 days at 4 C. (C) Single strand circular (ssc) DNA-induced coil to helix conformational changes in a-synuclein. (—), a-synuclein alone (10 lM); (...), a-synuclein + sscDNA (1, 2.5, 5, 7.5 lM); (- -), a-synuclein + sscDNA (10 lM). (D) Dependence of fractional CD change with ratio of concentration of sscDNA and a-synuclein. Open circles—[h]197 and closed circles—[h]222.

60

M.L. Hegde, K.S.J. Rao / Archives of Biochemistry and Biophysics 464 (2007) 57–69

conformation (Fig. 1A, solid line). On interaction of a-synuclein with scDNA (for 15 min at room temperature) the CD spectrum of a-synuclein revealed a partially folded conformation with a subtle shift toward a-helical conformation. On incubation of a-synuclein with scDNA for long time (one month) at 4 C, the CD spectrum of a-synuclein clearly showed a typical a-helical secondary structure with characteristic minima at 208 and 222 nm (Fig. 1B). The CD contribution from scDNA was very small and it was subtracted from the complex spectra. Single strand circular (ssc)DNA-induced a-helix in asynuclein The effect of sscDNA (non-supercoiled) binding to a-synuclein was observed by CD spectroscopy. The results revealed that sscDNA converts random-coiled a-synuclein to a typical a-helix (Fig. 1C and D). The percentage secondary structure content was calculated for the CD signal of a-synuclein–sscDNA complex using j-fit software (Table 1). It was observed that on binding to sscDNA (1:1, synuclein–sscDNA molar ratio) the helix content increases from 5% to 64%, while the random coil decreases from 95% to 33%. We also studied the effect of double strand circular (dsc) DNA binding on the CD spectra of a-synuclein. The dscDNA–a-synuclein was incubated in 0.01 M Hepes, pH 7.4, and incubated for 15 min at room temperature. However, no conformational change was induced by dscDNA (10 lg) on a-synuclein (Fig. 2A). GC* and AT* specificity of DNA-induced conformational transition of a-synuclein In order to understand the GC and AT nucleotide sequence specificity of DNA binding to a-synuclein, we studied the interaction of a-synuclein with poly d(AT)Æ(AT) and poly d(GC)Æd(GC). The study provided interesting insight on sequence specific binding affinity of DNA to asynuclein. In case of poly d(GC)Æd(GC) a partially folded conformation was formed in natively random-coiled a-synuclein (Fig. 2B), where as no such conformational transition was observed in case of poly d(AT)Æd(AT) binding to Table 1 The percent secondary structural content of a-synuclein on binding to various concentrations of sscDNA sscDNA in lg

[sscDNA]/[a-synuclein]

Coil

b-Sheet

Helix

0 1 2.5 5.0 7.5 10

0 0.17 0.42 0.84 1.25 1.67

94.60 76.89 54.24 38.52 33.45 33.10

0.06 2.97 2.02 3.46 4.20 3.15

4.80 12.73 44.74 58.01 62.10 63.75

Deconvolution of CD spectra of a-synuclein in the presence of DNA was carried out using J-fit software.

a-synuclein (Fig. 2C). Both the DNA was allowed to interact with a-synuclein at room temperature for 15 min in 0.01 M Hepes buffer, pH 7.4. Further, the ability of GC and AT specific 8-mer oligonucleotides to induce conformational transition in a-synuclein was also studied using CD. It was observed that d(GCGCGCGC) induced a partial folding in a-synuclein (Fig. 2D), while d(ATATATAT) showed only binding but did not bring about any such conformational change (Fig. 2E). Interestingly, d(GCATGCAT) also induced a partial folding in a-synuclein (Fig. 2F). Closer examination of the CD data indicated that the folding induced by d(GCGCGCGC) was more in magnitude compared to d(GCATGCAT). Interaction of genomic DNA with a-synuclein The effect of binding of k and Calf-thymus DNA on a-synuclein conformations was studied by CD experiments. It was observed that both these genomic DNA caused the formation of a partially folded structure in a-synuclein. However, the amount of folding induced by kDNA was more when compared to Calf-thymus DNA (Fig. 2G and H). To understand the differential ability of Calf-thymus and kDNA in inducing conformational transition in a-synuclein, the GC and AT nucleotide content of these DNA was correlated with their ability to induce folding. The GC content of Calf-thymus DNA is 70%, while for kDNA it is 42%. We have shown previously that GC* specific DNA is more efficient in inducing conformational transition in a-synuclein compared to AT-rich DNA. These results showed that DNA binding, results in a conformational transition in a-synuclein. sscDNA and scDNA-induced a-helix conformation, while genomic DNA and GC* specific DNA formed partially folded conformation, indicating a specificity for single-stranded DNA and GC sequence in inducing folding in a-synuclein. It appears that the DNA binding to a-synuclein is mediated through electrostatic interaction between negatively charged phosphate groups of DNA and the eamino group of lysine aminoacids in a-synuclein. The DNA molecule is richly negatively charged on its surface as it is laced with phosphate groups, where as a-synuclein has 15 basic lysine residues which are mostly clustered in the N-terminal of its sequence. The neutralization of basic charge on e amino group side chain of lysine residues will reduce the repulsion between the like charges in the N-terminal end of a-synuclein and this appears to be the driving force in inducing DNA mediated folding in the protein. Studies have shown that the N-terminal half of a-synuclein sequence has a very high propensity to form ordered conformation [51]. To understand the mechanism of DNA binding and DNA-induced folding in a-synuclein, we studied the effect of naturally occurring osmolytes on a-synuclein folding dynamics.

M.L. Hegde, K.S.J. Rao / Archives of Biochemistry and Biophysics 464 (2007) 57–69

61

Fig. 2. Linear DNA-induced partially folded conformation in a-synuclein. (A) CD spectra of double strand circular (dsc)DNA binding to a-synuclein. The dscDNA showed only binding to a-synuclein but no conformational folding was induced unlike sscDNA. (—), a-synuclein (10 lg); (. . .), asynuclein + dscDNA (5 lg); (- - -), a-synuclein + dscDNA (10 lg). (B and C) The interaction of poly d(GC)Æd(GC) and poly d(AT)Æd(AT) with a-synuclein. (B) CD spectra of a-synuclein-poly d(GC)Æd(GC) complex. (—), a-synuclein, 10 lM; (....), a-synuclein + 5 lg poly d(GC)Æd(GC); (- - -), a-synuclein + 10 lg poly d(GC)Æd(GC). (C) CD spectra of a-synuclein-poly d(AT)Æd(AT) complex. (—), a-synuclein, 10 lM; (- - -), a-synuclein + 10 lg poly d(AT)Æd(AT). (D–F) Interactions of a-synuclein with GC, AT oligonucleotides. Complex spectra of a-synuclein with different concentrations (1.0, 2.5, 5.0, 7.5 and 10 lM) d(GCGCGCGC), d(ATATATAT) and (GCATGCAT), respectively. (—), a-synuclein alone, 10 lM; (- - -), a-synuclein + 10 lM DNA oligonucleotide. (D) CD spectra of d(GCGCGCGC) interactions with a-synuclein; (E) CD spectra of d(ATATATAT) interactions with a-synuclein; (F) CD spectra of d(GCATGCAT) interactions with a-synuclein. (G and H) Interaction a-synuclein with kDNA and Calf-thymus DNA. Both lamda (G) and Calf-thymus (H) DNA-induced partial folding in a-synuclein, however, the amount of ordered conformation induced by kDNA was higher compared to Calf-thymus DNA. (G) (—), a-synuclein, 10 lM; (....), a-synuclein + kDNA (1.0, 2.5 and 5.0 lg, respectively); (- - -), a-synuclein + 10 lg kDNA. (H) (—), a-synuclein, 10 lM; (....), a-synuclein + Calf-thymus DNA (1.0 and 5.0 lg, respectively); (- - -), a-synuclein + 10 lg Calf-thymus DNA. The DNAsynuclein complexes are allowed to interact in 0.01 M Hepes, pH 7.4, for 15 min at room temperature and CD spectra were measured.

Role of naturally occurring osmolytes in inducing partial folding in a-synuclein Osmolytes are molecules used in nature to protect organisms against stresses at high osmotic pressure. These compounds have also been found to stabilize the native

state of proteins relative to the unfolded state [52]. The mechanism of osmolytes triggered stabilization of protein has been studied in some detail [53–55] and it is believed to result primarily from an unfavorable free-energy of interaction between the osmolytes and the unfolded state of the protein [56]. Proteins retain activity in the presence

62

M.L. Hegde, K.S.J. Rao / Archives of Biochemistry and Biophysics 464 (2007) 57–69

of osmolytes suggesting that the native state structure and dynamics are not greatly perturbed. The main classes of osmolytes are sugars, methyl ammonium derivatives, polyhydric alcohols, and amino acids and their derivatives [52]. Molar concentrations of all the above classes of molecules have been shown to stabilize proteins. The observation that a-synuclein and other amyloid peptides are generated within intracellular compartments, including endoplasmic reticulum, which is the quality control site for protein folding, has prompted us to investigate the role of organic chaperones in a-synuclein folding pathway [57]. These molecules have been shown to induce protein folding through preferential hydration of exposed polypeptide backbone and side chains of partially unfolded structures [58,59]. Hydration effects are equally important in protein polymerization or aggregation where osmolytes are excluded through increased protein–protein interactions [60]. We have used five osmolytes namely Glycerol, Betaine, Taurine, Sarcosine, TMAO to understand the initiation of a-synuclein folding from random coil to b-sheet conformation resulting in aggregation and also to understand the DNA-induced folding of a-synuclein. CD spectroscopic studies on the osmolyte-induced conformational changes in a-synuclein The effect of Osmolytes, Glycerol, Betaine, Taurine, Sarcosine, and TMAO, (all from Sigma) on a-synuclein secondary conformation were determined by CD. The protein was diluted to a final concentration of 5 lM in Hepes, pH 7.4. The protein solutions were combined with 1 · 106 to 1.0 M Betaine, Taurine, Sarcosine, and TMAO and 1.0–50% by volume of Glycerol. Spectra were collected after 15 min of equilibrium period at room temperature. The a-synuclein dissolved in Hepes, pH 7.4, exhibited a random-coil conformation indicative of an unordered secondary structure. The self-aggregation of a-synuclein involving conformational changes from random coil to bsheet and subsequent fibril formation takes from 25– 30 days, depending on the incubation conditions. In contrast, adjusting the protein solution to 25–50% v/v Glycerol, and 0.1–1.0 M Betaine, Taurine, and TMAO resulted in an immediate partial folding of the protein towards an ordered structure (Fig. 3). However, Sarcosine did not affect the CD signal of a-synuclein. a-Synuclein folding analysis using intrinsic fluorescence measurements The effect of osmolytes Glycerol, Betaine, TMAO, Taurine, and Sarcosine, on a-synuclein conformation were also studied using Tyrosine intrinsic fluorescence of a-synuclein. Tyrosine fluorescence was excited at 280 nm and emission monitored in the range from 290 to 450 nm. Protein concentrations for intrinsic fluorescence were kept at 2 lM at pH 7.4, in 0.01 M Hepes buffer.

Fig. 3. CD spectra of a-synuclein in presence of osmolytes: (A) Glycerol; (B) Betaine; and (C) Taurine; (D) TMAO. (—), a-synuclein, 5 lM; (....), asynuclein + 25% Glycerol or 0.1 M Betaine/TMAO; and (- - -), a-synuclein + 50% Glycerol or 1.0 M Betaine/Taurine/TMAO. The results demonstrate immediate conversion of a-synuclein conformation from random coil to partially folded conformation by Glycerol, Betaine, Taurine, and TMAO.

The effects of different osmolytes on the intrinsic tyrosine fluorescence of a-synuclein varied significantly. Among the five osmolytes studied, four osmolytes (Betaine, Taurine, TMAO, and Sarcosine) quench the protein fluorescence emission, whereas Glycerol enhanced the tyrosine fluorescence (Fig. 4). It is important to note that there are no tryptophan residues in a-synuclein. The fluorescence emission spectra also show that the osmolytes differ both in the amplitude of the induced spectral changes and in the efficiency in inducing protein folding. Interestingly, the pattern of fluorescence changes with Betaine, Taurine, TMAO, and Glycerol was exponential while Sarcosine decreased the tyrosine fluorescence of a-synuclein in a linear manner against the concentration of osmolyte.

M.L. Hegde, K.S.J. Rao / Archives of Biochemistry and Biophysics 464 (2007) 57–69

To confirm that the osmolyte-induced fluorescence change in a-synuclein is associated with changes in the protein conformation or folding and not just direct fluorescence quenching of tyrosine, we investigated the effect of these osmolytes on the fluorescence of free-L-tyrosine. It was found that all the osmolytes studied have much less pronounced effect on the fluorescence of free-L-tyrosine compared to the intrinsic fluorescence of a-synuclein. In fact, the fluorescence intensity of L-tyrosine was not affected at all in the presence of Taurine, which decreased the intrinsic fluorescence of a- synuclein quite significantly. Sarcosine, Betaine, Glycerol, and TMAO quenched L-tyrosine fluorescence to a much smaller extent than the intrinsic fluorescence of a-synuclein. It is also interesting to note that the quenching of free-tyrosine fluorescence by osmolytes shows linear relationship, indicating that the process involves only collisional or dynamic quenching. However, major deviations from linear dependence are observed in the case of intrinsic fluorescence of a-synuclein.

63

These observations indicate that the osmolytes under study induce structural reorganization of a-synuclein in the form of ordered folding. ANS binding Changes in ANS fluorescence are characteristic of the interaction of this dye with solvent-exposed hydrophobic surfaces of partially folded proteins [61]. Fig. 5A represents the ANS fluorescence of a-synuclein in the absence/presence of osmolytes. The results revealed that among the osmolytes used, Glycerol, TMAO, Betaine, and Taurine enhanced the ANS fluorescence in the same order (Fig. 5A). Interestingly, all these four osmolytes induced a blue shift (30 to 65 nm) of ANS kmax. This means that in the presence of Glycerol, TMAO, Betaine, and Taurine, natively unfolded a-synuclein is converted into a partially folded conformation with solvent-exposed hydrophobic patches. Glycerol, Taurine, and Betaine did not affect the

Fig. 4. The effect of osmolytes on a-synuclein intrinsic fluorescence. (d), free-L-tyrosine; (s), a-synuclein. (A) Effect of increasing concentrations of TMAO on tyrosine intrinsic fluorescence of a-synuclein, excitation at 280 nm, emission at 306 nm (295–375 nm). (B–F) Effect of TMAO, Taurine, Betaine, Glycerol, and Sarcosine on a-synuclein intrinsic fluorescence and their comparison to free-L-tyrosine. a-Synuclein concentrations were kept at 2.0 lM.

64

M.L. Hegde, K.S.J. Rao / Archives of Biochemistry and Biophysics 464 (2007) 57–69

Glycerol-induced folding exposes tyrosine residues as explained earlier. Effect of osmolytes on tryptic digestion resistance of a-synuclein a-Synuclein appears to be sensitive to trypsin digestion in its native conformation as it contains 15 lysine residues in its sequence. We have used proteolytic (tryptic) cleavage of a-synuclein as a probe of protein conformation by looking at the sensitivity of a-synuclein in the presence/absence of osmolytes. It was observed that a partial protection to tryptic action was shown by a-synuclein in the presence of osmolytes when compared to a-synuclein alone (Fig. 6). The restricted trypsin digestion of a-synuclein in the presence of osmolytes, along with CD and fluorescence data confirms that a-synuclein adopts a folded conformation in the presence of osmolytes. Effect of presence of DNA and osmolytes on a-synuclein fibril formation Fig. 5. (A) Effect of osmolytes on ANS fluorescence of a-synuclein. ANS fluorescence spectra measured for free-dye (.......) and in the presence of 5 lM a-synuclein and 50% v/v Glycerol, 0.5 M Taurine, 1 M each of TMAO, Betaine, and Sarcosine. (b) Stern–Volmer plots for acrylamide mediated tyrosine fluorescence quenching of a-synuclein in presence of osmolytes. Synuclein (s), free-L-tyrosine (D), Glycerol (d), Taurine (j), Betaine (h), TMAO (m), Sarcosine ( ). The conditions were as described under ‘Experimental procedures’.

ANS fluorescence emission in the absence of protein. Sarcosine did not show any change in ANS kmax even in the absence of a-synuclein, indicating no conformational change in a-synuclein.

Previous studies have shown that the transformation of a-synuclein into a partially folded conformation (induced by pH or temperature or metal ions) is strongly correlated with the enhanced formation of a-synuclein fibrils [10,39]. In this context, we analyzed the aggregation propensity of a-synuclein in the presence of osmolytes which induce partially folded conformation and also in the presence of DNA which induce a-helix. In agreement with the results of Uversky et al. [10,39], Fig. 7A showed that osmolytes which induced partial folding in a-synuclein resulted in a very substantial acceleration of the kinetics of aggregation. A shorter lag-time and a larger rate of fibril formation were

Acrylamide mediated fluorescence quenching studies Further evidence for the osmolytes induced changes in the environment of tyrosine residues in a-synuclein has been studied by acrylamide-induced tyrosine fluorescence quenching. Information on the relative solvent exposure of tyrosine residues (there are no tryptophans in a-synuclein) can be obtained from analysis of the effect of quencher molecules such as acrylamide [46]. Fig. 5B represents the Stern–Volmer plots for a-synuclein and a-synuclein–osmolyte complexes. The upward curvature indicates the presence of both dynamic and static quenching [46]. The results showed that in the absence of osmolytes, the Ksv for a-synuclein is smaller than the value for free-L-tyrosine. This indicates that the tyrosine residues of natively unfolded a-synuclein are partially protected from quencher molecules. In the presence of osmolytes the Ksv value decreases and in turn, the degree of protection increases, in the following order: a-synuclein + Glycerol; fia-synuclein + Sarcosine; fia-synuclein + Taurine; fia-synuclein + Betaine; fia-synuclein + TMAO. The Ksv for synuclein is less than that in presence of Glycerol because

Fig. 6. Effect of osmolytes on tryptic cleavage of a-synuclein. (a) a-Synuclein alone (4 lg); (b–f) a-synuclein subjected to limited trypsin treatment (detailed methodology given in ‘Experimental procedures’) in the absence (b) and presence of 50% v/v Glycerol (c); 1 M TMAO (d); 1 M Betaine (e), and 0.5 M Taurine (f). The samples were electrophoresed on Tricine-SDS–polyacrylamide gel and photographed after silver staining. a-Synuclein showed restricted tryptic digestion in the presence of osmolytes.

M.L. Hegde, K.S.J. Rao / Archives of Biochemistry and Biophysics 464 (2007) 57–69

observed in presence of osmolytes when compared to asynuclein alone. However, sscDNA which formed a-helix conformation in a-synuclein (Fig. 3), delayed the aggregation by nearly 25 hrs (Fig. 7B). For a-synuclein alone, a lag-time of 18–20 hrs was observed before the aggregation started as indicated by the increase in the Thio T fluorescence emission which it reached a limiting value after 37 h. In the presence of sscDNA (0.13 lM), the kinetics of fibrillation was much slower, reflected in the increase in lag-time for the Thio T intensity signal. Also the time it took for

65

reaching the limiting value (complete fibril formation was much greater compared to synuclein alone). This indicates that sscDNA inhibits synuclein fibril formation to a considerable extent. The observations were confirmed repeating the experiment three times, where similar trend was observed. scDNA also delayed the a-synuclein aggregation by 10 h. However, the results obtained with scDNA were not very consistent, the lag-time varying between 20 and 30 h. The d(GCGCGCGC) DNA which induced a partially folded intermediate conformation in a-synuclein (Fig. 2D), caused significant enhancement in a-synuclein aggregation kinetics (Fig. 7B). Similar observations were made by Uversky et al. [45], where they showed that TMAO induces a partial folding and acceleration of fibrillization in a-synuclein at low concentrations, where as, at high concentrations causes the formation of a-helix conformation and inhibits aggregation to a considerable extent. Our results are in agreement with Uversky et al. [45]. Hence, it appears that a partially folded intermediate conformation is a very critical step in a-synuclein aggregation pathway. The a-synuclein aggregation is associated with a conformational transition from random coil to b-sheet secondary structure. Hence, we confirmed the formation of b-sheet conformation in a-synuclein in the presence of osmolytes/ DNA once the Thio T fluorescence reached a limiting value. The CD spectra of representative aggregated a-synuclein showing b-sheet secondary structure is represented in Fig. 7C. TEM study of a-synuclein aggregates

Fig. 7. Modulation of a-synuclein aggregation by osmolytes and DNA. The kinetics of a-synuclein fibrillation was monitored by Thio T fluorescence. The symbols represent Thio T fluorescence intensities at 482 nm determined experimentally. (A) Effect of osmolytes on fibrillation of a-synuclein. The a-synuclein concentrations were 50 lM in 0.01 M Tris–Cl buffer, pH 7.4. (B) Modulation of a-synuclein aggregation by DNA. Variation of flourescence emission intensity of Thio T at 482 nm for a-synuclein incubated in the absence (a) and presence of pUC 18 scDNA (b); single-stranded circular DNA (c); d(GCGCGCGC) (d). The DNA: a-synuclein molar ratio was 0.13:50 lM. The excitation wavelength was 450 nm. A significant increase in the lag-time of aggregation of 10 h was observed for synuclein in the presence of DNA. (C) CD spectra of aggregated a-synuclein indicating b-sheet secondary conformation.

The structures of a-synuclein aggregates formed upon incubation at 37 C with stirring using micro-magnetic bars were examined by TEM. The aggregates were seen as long, often forming large networks, and distributed rather uniformly on the surface of the carbon film (Fig. 8). Some small aggregates of variable size (more or like amorphous) were also seen in the image (Fig. 8, see arrow), however, the aggregates were predominantly fibrillar. The structure of asynuclein aggregates/ fibrils were qualitatively similar in the presence or absence of DNA or osmolytes. This could be because, the images were taken when the fibrillization was complete as indicated by the maximum Thio T fluorescence. However, it is important to mention that fibrillation of ordered and natively unfolded proteins involves different mechanisms: partial unfolding of ordered proteins and partial folding of natively unfolded proteins as discussed in detail by Uversky and Fink [62]. The mechanism of fibrillation of a-synuclein with helix inducing DNA or partial folding inducing osmolytes may involve similar mechanisms. Discussion a-Synuclein is a natively unfolded protein with little or no ordered structure under physiological conditions. At

66

M.L. Hegde, K.S.J. Rao / Archives of Biochemistry and Biophysics 464 (2007) 57–69

Fig. 8. Negatively stained transmission electron micrographs of fibrils formed from a-synuclein in the presence of DNA and osmolytes. (A) aSynuclein alone; (B) a-synuclein + d(GC)4 oligonucleotide; (C) a-synuclein + single-stranded circular DNA; (D) a-synuclein + supercoiled DNA; (E) a-synuclein + Glycerol; (F) a-synuclein + Taurine. The protein samples were applied on carbon-coated Formver grids, and allowed to dry in air. After staining for 2 min with 2% uranyl acetate, samples were observed under a JEOL 1010 transmission electron microscope. Magnification: scale bar, 100 nm.

neutral pH, it is calculated to have 24 negative charges (15 of which are localized in the last third of the protein sequence), leading to a strong electrostatic repulsion, which hinders the folding of a-synuclein [39]. As previously reported [9,10], a-synuclein at neutral pH has a far-UVCD spectrum typical of an unfolded polypeptide chain, and reflecting the lack of ordered secondary conformation under these conditions. This raises the question of how an essentially disordered protein is transformed into highly organized fibrils in Parkinson’s brain?. In the search for an answer to this question, we have investigated the effects of DNA and osmolyte binding on the conformational properties of a-synuclein. Aggregation or self-association is a characteristic property of a partially folded (denatured) proteins and most aggregating protein systems probably involve a transient partially folded intermediate as the key precursor of fibrillation [62–64]. It has also been shown

that in some cases the self-association induces additional structure and stability in the partially folded intermediates. The natively unfolded character of a-synuclein arises from its low intrinsic hydrophobicity and high net charge at neutral pH (pI 4.7) [12]. Thus the conditions that decrease the net charge and that increase the hydrophobicity would be expected to result in partial folding. Osmolytes provide such conditions by affecting the hydration properties at the protein surface [55,65]. In the present study, the shape and intensity of the far-UV-CD and intrinsic fluorescence spectra change significantly with an increase in the osmolyte concentration, indicating that the conformation of natively unfolded a-synuclein is significantly affected by interaction with the osmolytes, which induce secondary conformation in synuclein. This process is accompanied by an increase in [h]220 (negative CD intensity) and change in the intrinsic fluorescence at 305 nm. These changes are attributed to the osmolyte-induced changes in the environment of tyrosine residues, possibly due to their solvent accessibility because of partial folding of the protein. It has been shown previously that [10] either an increase in temperature or a decrease in pH level rapidly transformed natively unfolded a-synuclein into a partially folded conformation. Moreover, the same study showed a tight correlation between the increase of this intermediate and the enhanced formation of a-synuclein fibrils. Interestingly, in the present study, the magnitude of the osmolyte mediated structural perturbations, as monitored by CD, is comparable with that induced by low pH or high temperature [10]. Thus, natively unfolded a-synuclein adopts a partially folded conformation even at neutral pH in the presence of osmolytes. This partially folded conformation is characterized by an increased amount of ordered secondary structure and by the appearance of solvent accessible hydrophobic clusters. In agreement with Uversky et al. [10], our results showed that the osmolytes induced partial folding in a-synuclein is associated with enhanced assembly of a-synuclein into fibrils. Munishkina et al. [33] observed that organic solvents (simple alcohols, trifluoroethanol (TFE) etc.) cause natively unfolded a-synuclein to fold in a multiphasic manner, the mechanism of such folding differing for different solvents. However, all the solvents lead to the formation of a previously described partially folded intermediate [10,39,41,45]. The subsequent fate of the partially folded intermediate was dependent on the nature of the solvent: simple alcohols and moderate concentrations of TFEinduced b-structure-enriched oligomers, where as high concentrations of fluoro alcohols gave rise to a-helical conformation. It was further shown that under conditions where this partially folded intermediate is populated, a-synuclein fibrillation is significantly faster, indicating that the partially folded intermediate is a critical species in the a-synuclein fibrillation pathway [33]. Our results on DNA- and osmolytes-induced conformations in a-synuclein is in agreement with Munishkina et al. [33] as far as the fibrilla-

M.L. Hegde, K.S.J. Rao / Archives of Biochemistry and Biophysics 464 (2007) 57–69

tion propensity of partially folded intermediate is concerned. Importantly, the strong correlation observed between the degree of protein folding and its efficiency of fibril formation suggests that this intermediate partially folded form can be a precursor of fibrils [10]. This is because, in contrast to an unfolded protein, a partially folded intermediate is anticipated to have contiguous hydrophobic patches on its surface (also indicated by ANS binding studies), which are likely to foster self-association and hence potentially fibrillation. Since hydrophobic interactions increase with increase in osmolyte concentration, the simplest explanation for the osmolyte-induced formation of aggregation competent intermediate (confirmed by Thio T binding) is due to increased hydrophobic interactions. There are two interesting features of the a-synuclein partially folded intermediate: it has some b-structure, which is the major type of secondary structure in a-synuclein fibrils, and it is relatively unfolded (i.e. more similar to randomcoil conformation than a native tightly folded globular conformation). This partially folded conformation shares many structural properties with the pre-molten globule state [62]. Khurana et al. [66] have recently observed that a partially folded intermediate of IgG light chains that form amyloid fibrils is relatively unfolded and they have also predicted that it could be a common feature in amyloid fibril formation that the critical monomeric partially folded intermediate is relatively unfolded. The question arises as to the physiological significance of the observation of a partially folded intermediate of a-synuclein formed with osmolytes. Clearly, these particular conditions will not be found in the dopaminergic cells of potential PD patients. However, the existence of such an intermediate on the pathway to fibrils means that in vivo conditions that lead to population of the intermediate will lead to increased risk of the disease. Thus, intracellular factors that lead to a shift in the equilibrium position between the natively unfolded state and the partially folded intermediate will increase the likelihood of a-synuclein fibril formation. Such factors could include relatively non-polar molecules that would preferentially bind to the partially folded intermediate. The ability of DNA and osmolytes in inducing conformational transition in a-synuclein, indicates that two factors are critical in modulating a-synuclein folding: (i) electrostatic interaction as in the case of DNA, and (ii) hydrophobic interactions as in the case of osmolytes. It was previously shown that conformational and fibrillation behavior of a-synuclein was modulated by different polyions [67–69]. The formation of a-synuclein–polycation complex dramatically enhanced a-synuclein fibrillation [67]. The authors observed that the magnitude of the effect depended on the nature of the polymer, its length and concentration [67]. Similarly, different DNAs induced different conformational changes in a-synuclein in the present study. The single-stranded circular DNA-induced a-helix conformation, while smaller oligos and linear duplex DNAinduced partial folding which are prone to aggregation.

67

The physiological significance of DNA-induced a-synuclein conformation and modulation of its assembly/fibrillation is unclear at the present time. However, several studies have shown the presence of a-synuclein into neuronal nuclei [1,70–76]. Recently, it was shown that a-synuclein is present in neuronal nuclear inclusions and neuritis in multiple system atrophy [76]. Goers et al. [75] provided evidence for the co-localization of a-synuclein with histones in the nuclei of nigral neurons from mice exposed to a toxic insult. It was further shown that the stoichiometry of the histone: a-synuclein complex was 2:1 and histones stimulate a-synuclein fibrillation in vitro [75]. A more recent study on interaction of double-stranded DNA with a-synuclein and its ability to enhance the fibrillation of a-synuclein revealed that both the components of chromatin, histone and DNA may have a significant effect on nuclear translocated a-synuclein functioning [42]. The authors proposed that a-synuclein may interact with histone-free, transcriptionally active DNA segments and hence may lead to a decreased transcriptional activity of some genes responding to environmental stimuli. These results along with our present study reveal that the interactions of a-synuclein with DNA and histones may function to regulate gene expressions. The possible mechanisms of double-stranded DNA promoting a-synuclein fibrillation has been proposed recently by Cherny et al. [42]. The authors also observed that neuronal nuclear inclusions potentially account for a significant fraction of the total amount of a-synuclein in a cell. Hence, minute variations in local a-synuclein concentrations or the presence of factors enhancing its fibrillation, e.g. DNA or histones, may stimulate significantly the aggregation of a-synuclein into fibrillar structures. It was further proposed that effective mechanisms preventing unnecessary or occasional conversion of a soluble a-synuclein into insoluble isoforms must exist in both cytoplasm and nucleus [42]. However, in the present study, we provide a comprehensive picture of DNA binding effect on a-synuclein fibrillation using different DNAs such as double and single-stranded DNA, AT, and GC sequence specific DNA of different sizes etc and showed that only those DNA which induce a partial folding in a-synuclein (GC*rich DNA) promote its aggregation, while, sscDNA forms a-helix conformation in a-synuclein and also inhibit aggregation to a considerable extent (Fig. 9). Hence, we feel that extrapolation of in vitro results on DNA binding property of a-synuclein to in vivo system in PD has to be more cautiously done. However, we cannot ignore the significance of DNA binding property of a-synuclein as a non-specific event. It is because, amyloid b peptides (Ab), tau and prion proteins implicated in AD and prion diseases have also been shown to have DNA binding property [77–84]. Nucleic acids was shown to induce structural changes in prion peptides by forming stable complexes, which catalyzed further polymerization [82,85]. Association of Ab(1–40) and Ab(25–35) with double-stranded DNA was detected [77]. Ab(25–35)

68

M.L. Hegde, K.S.J. Rao / Archives of Biochemistry and Biophysics 464 (2007) 57–69

tions. We also thank Chairman, Molecular Biophysics Unit, Indian Institute of Science, India for allowing us to use the Circular Dichroism facility. We thank Prof. H.S. Prakash, Department of Applied Botany and Biotechnology, University of Mysore, India for providing TEM facility. References

Fig. 9. Schematic representation of DNA- and osmolytes-induced conformational changes and fibrillation of a-synuclein. sscDNA, single strand circular DNA; scDNA, supercoiled DNA.

was shown to cause formation of open circular DNA from scDNA in presence of ferrous ions [78]. We have recently observed binding of Ab(1–42) and Ab(1–16) peptides with scDNA and their ability to convert scDNA into open circular form [86–88]. It is also interesting to understand the role of metals in synuclein/amyloid interactions with DNA in relevance to neurodegeneration [89–95]. From the above evidences, it appears that the DNA binding is a unifying property of amyloidogenic peptides (a-synuclein, Ab, s, prions) implicated in various neurodegenerative disorders. Hence, we feel that further multiple systematic approaches including in vivo studies using animal models and cell cultures are required to understand the implications of these in vitro evidences on DNA binding properties of a-synuclein and other amyloidogenic peptides. Further, we feel that the property of sscDNA in inducing a-helical conformation in a-synuclein and inhibiting the fibrillation may be of significance in engineering DNA-chip based therapeutic approaches to PD and other amyloid disorders. Acknowledgments The authors thank Director, CFTRI for his encouragement. The financial assistance by Department of Atomic Energy, India through BNRS project is gratefully acknowledged. M.L.H. is grateful to Council for Scientific and Industrial Research, India for Senior Research Fellowships. We thank Prof. R. Varadarajan, Molecular Biophysics Unit, Indian Institute of Science, Bangalore for reviewing the manuscript and providing valuable sugges-

[1] L. Maroteaux, J.T. Campanelli, R.H. Scheeller, J. Neurosci. 8 (1988) 2804–2815. [2] R. Jakes, M.G. Spillantini, M. Goedert, FEBS Lett. 345 (1994) 27–32. [3] A. Iwai, E. Masliah, M. Yoshimoto, N. Ge, L. Flanagan, H.A. De Silva, A. Kittel, T. Saitoh, Neuron 14 (1996) 461–475. [4] J.M. George, Genome Biol. 3 (2002) 3002.1–3002.6. [5] J.M. George, H. Jin, W.S. Woods, D.F. Clayton, Neuron 15 (1995) 361–372. [6] J.P. Segrest, H. De Loof, J.G. Dohlman, C.G. Brouillette, G.M. Anantharamaiah, Proteins 8 (1990) 103–117. [7] J. Segrest, M. Jones, H. De Loof, C.G. Trouillette, Y.V. Venkatachalapathi, G.M. Ananthramaih, J. Lipid Res. 33 (1992) 141–166. [8] M.L. Hegde, K.S.J. Rao, Arch. Biochem. Biophys. 418 (2003) 169– 178. [9] P.H. Weinreb, W. Zhen, A.W. Poon, K.A. Conway, P.T. Lansbury Jr., Biochemistry 35 (1996) 13709–13715. [10] V.N. Uversky, J. Li, A.L. Fink, J. Biol. Chem. 276 (2001) 10737– 10744. [11] D. Eliezer, E. Kutluay, R. Bussell, G.J. Browne, J. Mol. Biol. 307 (2001) 1061–1073. [12] V.N. Uversky, J.R. Gillespie, A.L. Fink, Proteins Struct. Funct. Genet. 41 (2000) 415–427. [13] V.N. Uversky, Protein Sci. 11 (2002) 739–756. [14] V.N. Uversky, Eur. J. Biochem. 269 (2002) 2–12. [15] K.W. Plaxco, C. Spitzfaden, I.D. Campbell, C.M. Dobson, J. Mol. Biol. 270 (1997) 763–770. [16] A.K. Dunker, J.D. Lawson, .J. Brown, R.M. Williams, P. Romero, J.S. Oh, C.J. Oldfield, A.M. Campen, C.M. Ratliff, K.W. Hipps, J. Ausio, M.S. Nissen, R. Reeves, C. Kang, C.R. Kissinger, R.W. Bailey, M.D. Griswold, W. Chiu, E.C. Garner, Z. Obradovic, J. Mol. Graph. Model. 19 (2001) 26–59. [17] V.N. Uversky, Cell. Mol. Life Sci. 60 (2003) 1852–1871. [18] V.N. Uversky, J. Biomol. Struct. Dyn. 21 (2003) 211–234. [19] P. Radivojac, L.M. Iakoucheva, C.J. Oldfield, Z. Obradovic, V.N. Uversky, A.K. Dunker, Biophys. J. 2007 92 (2007) 1439–1456. [20] V.N. Uversky, C.J. Oldfield, A.K. Dunker, J. Mol. Recognit. 18 (2005) 343–384. [21] A.K. Dunker, M.S. Cortese, P. Romero, L.M. Iakoucheva, V.N. Uversky, FEBS J. 272 (2005) 5129–5148. [22] C.B. Lucking, A. Brice, Cell. Mol. Life Sci. 57 (2000) 1894–1908. [23] J.Q. Trojanowsky, V.M. Lee, Arch. Neurol. 55 (1998) 151–152. [24] R.A. Crowther, R. Jakes, M.G. Spillantini, M. Goedert, FEBS Lett. 436 (1998) 309–312. [25] O.M. El-Agnaf, S.R. Jake, M.D. Curran, A. Wallace, FEBS Lett. 440 (1998) 67–70. [26] S.J. Wood, J. Wypych, S. Steavenson, J.C. Louis, M. Citron, A.L. Biere, J. Biol. Chem. 274 (1999) 19509–19512. [27] B.I. Giasson, K. Uryu, J.Q. Trojanowski, V.M. Lee, J. Biol. Chem. 274 (1999) 7619–7622. [28] L. Narhi, S.J. Wood, S. Steavenson, Y. Jiang, G.M. Wu, D. Anafi, S.A. Kaufman, F. Martin, K. Sitney, P. Denis, J.C. Louis, J. Wypych, A.L. Biere, M. Citron, J. Biol. Chem. 274 (1999) 9843–9846. [29] L.C. Serpel, J. Berriman, R. Jakes, M. Goedert, R.A. Crowther, Proc. Natl. Acad. Sci. USA 97 (2000) 4902–5897. [30] K.A. Conway, J.D. Harper, P.T. Lansbury Jr., Biochemistry 39 (2000) 2552–2563. [31] J. Li, V.N. Uversky, A.L. Fink, Biochemistry 40 (2001) 11604–11613.

M.L. Hegde, K.S.J. Rao / Archives of Biochemistry and Biophysics 464 (2007) 57–69 [32] W. Hoyer, T. Antony, D. Cherny, G. Heim, T.M. Jovin, V. Subramaniam, J. Mol. Biol. 322 (2002) 383–393. [33] L.A. Munishkina, C. Phelan, V.N. Uversky, A.L. Fink, Biochemistry 42 (2003) 2720–2730. [34] K.A. Conway, S.J. Lee, J.C. Rochet, T.T. Ding, R.E. Williamson, P.T. Lansbury, Proc. Natl. Acad. Sci. USA 97 (2000) 571–576. [35] R.J. Perrin, W.S. Woods, D.F. Clayton, J.M. George, J. Biol. Chem. 276 (2001) 41958–41962. [36] B.N. Cole, D.D. Murphy, T. Grider, S. Rueter, D. Brasaemle, R.L. Nussbaum, J. Biol. Chem. 277 (2002) 6344–6352. [37] E. Jo, A.A. Darabie, K. Han, A. Tandon, P.E. Fraser, J. McLaurin, Eur. J. Biochem. 271 (2004) 3180–3189. [38] M.K. Lee, W. Stirling, Y.Q. Xu, X.Y. Xu, D. Qui, A.S. Mandir, T.M. Dawson, M.G. Copeland, N.A. Jenkins, D.L. Price, Proc. Natl. Acad. Sci. USA 99 (2002) 8968–8973. [39] V.N. Uversky, J. Li, A.L. Fink, J. Biol. Chem 276 (2001) 44284– 44296. [40] V.N. Uversky, J. Li, A.L. Fink, FEBS Lett. 500 (2001) 105–108. [41] A.B. Manning-Bog, A.L. McCormack, J. Li, V.N. Uversky, A.L. Fink, D.A. Di Monte, J. Biol. Chem. 277 (2002) 1641–1644. [42] D. Cherny, W. Hoyer, V. Subramaniam, T.M. Jovin, J. Mol. Biol. 344 (2004) 929–938. [43] M.L. Hegde, R.K.S. Jagannatha, J. Neurochem. (Abstract) 98 (Suppl.) (2006) 114–115. [44] R.J. Perrin, W.S. Woods, D.F. Clayton, J.M. George, J. Biol. Chem. 275 (2000) 34393–34398. [45] V.N. Uversky, J. Li, A.L. Fink, FEBS Lett. 509 (2001) 31–35. [46] M.R. Eftink, C.A. Ghiron, Anal. Biochem. 114 (1981) 199–227. [47] G. Nadig, G.S. Ratnaparkhi, R. Varadarajan, S. Vishveshwara, Protein Sci. 5 (1996) 2104–2114. [48] A. Fontana, M. Zambonin, P. Polverino de Laureto, V. De Filippis, A. Clementi, E. Scaramella, J. Mol. Biol. 266 (1997) 223–230. [49] P.P. de Laureto, L. Tosatto, E. Frare, O. Marin, V.N. Uversky, A. Fontana, Biochemistry 45 (2006) 11523–11531. [50] T.J. Thomas, V.A. Bloomfield, Biopolymers 24 (1985) 2185–2194. [51] S. Chandra, X. Chen, J. Rizo, R. Jahn, T.C. Sudhof, J. Biol. Chem. 278 (2003) 15313–15318. [52] G.S. Ratnaparkhi, R. Varadarajan, J. Biol. Chem. 276 (2001) 28789– 28798. [53] T. Arakawa, S.N. Timasheff, Biophys. J. 47 (1985) 411–414. [54] P.I.M. Pino, J.M. Sanchez-Ruiz, Biochemistry 34 (1995) 8621–8630. [55] Y. Liu, D.W. Bolen, Biochemistry 34 (1995) 12884–12891. [56] G. Xie, S.N. Timasheff, Protein Sci. 6 (1997) 211–221. [57] D.S. Yang, C.M. Yip, T.H. Huang, A. Chakrabartty, P.E. Fraser, J. Biol. Chem. 274 (1999) 32970–32974. [58] M.B. Burg, Am. J. Physiol. 268 (1995) F983–F996. [59] A. Wang, D.W. Bolen, Biochemistry 36 (1997) 9101–9108. [60] D.L. Sackett, Am. J. Physiol. 273 (1997) R669–R676. [61] G.V. Semisotnov, N.A. Rodionova, O.I. Razgulyaev, V.N. Uversky, A.F. Gripas, R.I. Gilmanshin, Biopolymers 31 (1991) 119–128. [62] V.N. Uversky, A.L. Fink, Biochim. Biophys. Acta 1698 (2004) 131– 153. [63] R. Wetzel, Cell 86 (1996) 699–702. [64] A.L. Fink, Annu. Rev. Biophys. Biomol. Struct. 24 (1995) 495–522. [65] M.M. Santoro, Y. Liu, M.A. Saber, H. Li-Xiang, W. Bolen, Biochemistry 31 (1992) 5278–5283. [66] R. Khurana, J.R. Gillespie, A. Talapatra, L.J. Minert, C. IonescuZanetti, I. Millett, A.L. Fink, Biochemistry 40 (2001) 3525–3535.

69

[67] J. Goers, V.N. Uversky, A.L. Fink, Protein Sci. 12 (2003) 702–707. [68] J.A. Cohlberg, J. Li, V.N. Uversky, A.L. Fink, Biochemistry 41 (2002) 1502–1511. [69] I.H. Liu, V.N. Uversky, L.A. Munishkina, A.L. Fink, W. Halfter, G.J. Cole, Glycobiology 15 (2005) 1320–1331. [70] E. Gomez-Tortosa, K. Newell, M.C. Irizarry, J.L. Sanders, B.T. Hyman, Acta Neuropathol. (Berl.) 99 (2000) 352–357. [71] P.J. McLean, H. Kawamata, S. Ribich, B.T. Hyman, J. Biol. Chem. 275 (2000) 8812–8816. [72] K. Tanji, F. Mori, T. Imaizumi, H. Yoshida, T. Matsumiya, W. Tamo, M. Yoshimoto, H. Odagiri, M. Sasaki, H. Takahashi, K. Satoh, K. Wakabayashi, Pathol. Int. 52 (2002) 572–577. [73] Y. Leng, T.N. Chase, M.C. Bennet, J. Biol. Chem. 30 (2001) 28212– 28218. [74] P. Sangchot, S. Sharma, B. Chetsawang, J. Porter, P. Govitrapong, M. Ebadi, Dev. Neurosci. 24 (2002) 143–153. [75] J. Goers, A.B. Manning-Bog, A.L. McCormack, I.S. Millett, S. Doniach, D. Di Monte, V.N. Uversky, A.L. Fink, Biochemistry 42 (2003) 8465–8471. [76] W.L. Lin, M.W. DeLucia, D.W. Dickson, Neurosci. Lett. 354 (2004) 99–102. [77] B.W. Ahn, D.U. Song, Y.D. Jung, K.O. Chay, M.A. Chung, S.Y. Yang, B.A. Shin, Anal. Biochem. 284 (2000) 401–405. [78] J.H. Jang, Y.J. Surh, Ann. N. Y. Acad. Sci. 973 (2002) 228–236. [79] T. Kampers, P. Friedhoff, J. Biernat, E.M. Mandelkow, E. Mandelkow, FEBS Lett. 399 (1996) 344–349. [80] P.K. Nandi, E. Leclerc, Arch. Virol. 144 (1999) 1751–1763. [81] P.K. Nandi, Arch. Virol. 143 (1998) 1251–1263. [82] Y. Cordeiro, F. machado, L. Juliano, M.A. Juliano, R.R. Brentani, D. Foguel, J.L. Silva, J. Biol. Chem. 276 (2001) 49400–49409. [83] C. Gabus, S. Auxilien, C. Pechoux, D. Dormont, W. Swietnicki, M. Morillas, J. Mol. Biol. 307 (2001) 1011–1021. [84] C. Gabus, E. Derrington, P. Leblanc, J. Chnaiderman, D. Dormant, W. Swietnicki, J. Biol. Chem. 276 (2001) 19301–19309. [85] P.K. Nandi, E. Leclerc, J.C. Nicol, M. Takahashi, J. Mol. Biol. 322 (2002) 153–161. [86] M.L. Hegde, S. Anitha, K.S. Latha, M.S. Mustak, Reuven Stein, Rivka Ravid, K.S.J. Rao, J. Mol. Neurosci. 22 (2004) 19–31. [87] M.L. Hegde, S. Anitha, K.S.J. Rao, Neurobiol. Aging (Abstract) 25 (Suppl. 2) (2004) 170. [88] H. Yu, J. Ren, X. Qu. Biophys. J. 92 (2007) 185–191. [89] K.S.J. Rao, M.L. Hegde, S. Anitha, M. Musicco, F.A. Zucca, N.J. Turro, L. Zecca, Prog. Neurobiol. 78 (2006) 364–373. [90] M.L. Hegde, V.B. Gupta, M. Anitha, T. Hankrishna, S.K. Shankar, U. Muthane, K.S. Rao, K.S.J. Rao, Arch. Biochem. Biophys. 449 (2006) 143–156. [91] M.B.S. Pande, P. Nagabhushan, M.L. Hegde, T.S.S. Rao, K.S.J. Rao, comput. Biol. Med. 35 (2005) 475–493. [92] M.L. Hegde, P. Shanmugavelu, B. Vengamma, T.S.S. Rao, R.B. Menon, R.V. Rao, K.S.J. Rao, J. Trace Elem. Med. Biol. 18 (2004) 163–171. [93] V.B. Gupta, S. Anitha, M.L. Hegde, L. Zecca, R.M. Garruto, R. Ravid, S.K. Shankar, R. Stein, P. Shanmugavelu, K.S.J. Rao, Cell. Mol. Life Sci. 62 (2005) 143–158. [94] P. Shanmugavelu, T.S.S. Rao, M.L. Hegde, M.S. Mustak, R.B. Menon, R.V. Rao, K.S.J. Rao, J. Neurochem. (Suppl. 8) (2004) 49. [95] V.B. Gupta, M.L. Hegde, K.S. Rao, Curr. Alzheimer Res. 3 (2006) 297–309.