Relative and regional stabilities of the hamster, mouse, rabbit, and bovine prion proteins toward urea unfolding assessed by nuclear magnetic resonance and circular dichroism spectroscopies.

The residue-specific urea-induced unfolding patterns of recombinant prion proteins from different species (bovine, rabbit, mouse, and Syrian hamster) were monitored using high-resolution (1)H nuclear magnetic resonance (NMR) spectroscopy. Protein constructs of different lengths, and with and without a His tag attached at the N-terminus, were studied. The various species showed different overall sensitivities toward urea denaturation with stabilities in the following order: hamster ≤ mouse < rabbit < bovine protein. This order is in agreement with recent circular dichroism (CD) spectroscopic measurements for several species [Khan, M. Q. (2010) Proc. Natl. Acad. Sci. U.S.A.107, 19808-19813] and for the bovine protein presented herein. The [urea](1/2) values determined by CD spectroscopy parallel those of the most stable residues observed by NMR spectroscopy. Neither the longer constructs containing an additional hydrophobic region nor the His tag influenced the stability of the structured domain of the constructs studied. The effect of the S174N mutation in rabbit PrP(C) was also investigated. The rank order of the regional stabilities within each protein remained the same for all species. In particular, the residues in the ß-sheet region in all four species were more sensitive to urea-induced unfolding than residues in the α2 and α3 helical regions. These observations indicate that the regional specific unfolding pattern is the same for the four mammalian prion proteins studied but militate against the idea that PrP(Sc) formation is linked with the global stability of PrP(C).

Detailed biophysical characterization of the acid-induced PrP(c) to PrP(ß) conversion process.

Prions are believed to spontaneously convert from a native, monomeric highly helical form (called PrP(c)) to a largely ß-sheet-rich, multimeric and insoluble aggregate (called PrP(sc)). Because of its large size and insolubility, biophysical characterization of PrP(sc) has been difficult, and there are several contradictory or incomplete models of the PrP(sc) structure. A ß-sheet-rich, soluble intermediate, called PrP(ß), exhibits many of the same features as PrP(sc) and can be generated using a combination of low pH and/or mild denaturing conditions. Studies of the PrP(c) to PrP(ß) conversion process and of PrP(ß) folding intermediates may provide insights into the structure of PrP(sc). Using a truncated, recombinant version of Syrian hamster PrP(ß) (shPrP(90-232)), we used NMR spectroscopy, in combination with other biophysical techniques (circular dichroism, dynamic light scattering, electron microscopy, fluorescence spectroscopy, mass spectrometry, and proteinase K digestion), to characterize the pH-driven PrP(c) to PrP(ß) conversion process in detail. Our results show that below pH 2.8 the protein oligomerizes and conversion to the ß-rich structure is initiated. At pH 1.7 and above, the oligomeric protein can recover its native monomeric state through dialysis to pH 5.2. However, when conversion is completed at pH 1.0, the large oligomer "locks down" irreversibly into a stable, ß-rich form. At pH values above 3.0, the protein is amenable to NMR investigation. Chemical shift perturbations, NOE, amide line width, and T(2) measurements implicate the putative "amylome motif" region, "NNQNNF" as the region most involved in the initial helix-to-ß conversion phase. We also found that acid-induced PrP(ß) oligomers could be converted to fibrils without the use of chaotropic denaturants. The latter finding represents one of the first examples wherein physiologically accessible conditions (i.e., only low pH) were used to achieve PrP conversion and fibril formation.

Non-identical electronic characters of the internucleotidic phosphates in RNA modulate the chemical reactivity of the phosphodiester bonds.

We here show that the electronic properties and the chemical reactivities of the internucleotidic phosphates in the heptameric ssRNAs are dissimilar in a sequence-specific manner because of their non-identical microenvironments, in contrast with the corresponding isosequential ssDNAs. This has been evidenced by monitoring the delta H8(G) shifts upon pH-dependent ionization (pK(a1)) of the central 9-guaninyl (G) to the 9-guanylate ion (G-), and its electrostatic effect on each of the internucleotidic phosphate anions, as measured from the resultant delta 31P shifts (pKa2) in the isosequential heptameric ssRNAs vis-à-vis ssDNAs: [d/r(5'-Cp1Ap2Q1p3Gp4Q2p5Ap6C-3'): Q1 = Q2 = A (5a/5b) or C (8a/8b), Q(1) = A, Q(2) = C (6a/6b), Q1 = C, Q2 = A (7a/7b)]. These oligos with single ionizable G in the centre are chosen because of the fact that the pseudoaromatic character of G can be easily modulated in a pH-dependent manner by its transformation to G- (the 2'-OH to 2-O- ionization effect is not detectable below pH 11.6 as evident from the N(1-Me)-G analog), thereby modulating/titrating the nature of the electrostatic interactions of G to G- with the phosphates, which therefore constitute simple models to interrogate how the variable pseudoaromatic characters of nucleobases under different sequence context (J. Am. Chem. Soc., 2004, 126, 8674-8681) can actually influence the reactivity of the internucleotide phosphates as a result of modulation of sequence context-specific electrostatic interactions. In order to better understand the impact of the electrostatic effect of the G to G- on the tunability of the electronic character of internucleotidic phosphates in the heptameric ssRNAs 5b, 6b, 7b and 8b, we have also performed their alkaline hydrolysis at pH 12.5 at 20 degrees C, and have identified the preferences of the cleavage sites at various phosphates, which are p2, p3 and p4 (Fig.3). The results of these alkaline hydrolysis studies have been compared with the hydrolysis of analogous N(1-Me)-G heptameric ssRNA sequences 5c, 7c and 8c under identical conditions in order to establish the role of the electrostatic effect of the 9-guanylate ion (and the 2'-OH to 2-O- ionization) on the internucleotidic phosphate. It turned out that the relative alkaline hydrolysis rate at those particular phosphates (p2, p3 and p4) in the N(1-Me)-G heptamers was reduced from 16-78% compared to those in the native counterparts [Fig. 4, and ESI 2 (Fig. S11)]. Thus, these physico-chemical studies have shown that those p2, p3 and p4 phosphates in the native heptameric RNAs, which show pKa2 as well as more deshielding (owing to weaker 31P screening) in the alkaline pH compared to those at the neutral pH, are more prone to the alkaline hydrolysis because of their relatively enhanced electrophilic character resulting from weaker 31P screening. This screening effect originates as a result of the systematic charge repulsion effect between the electron cloud in the outermost orbitals of phosphorus and the central guanylate ion, leading to delocalization of the phosphorus p(pi) charge into its dpi orbitals. It is thus likely that, just as in the non-enzymatic hydrolysis, the enzymatic hydrolysis of a specific phosphate in RNA by general base-catalysis in RNA-cleaving proteins (RNase A, RNA phosphodiesterase or nuclease) can potentially be electrostatically influenced by tuning the transient charge on the nucleobase in the steric proximity or as a result of specific sequence context owing to nearest-neighbor interactions.

Significant pKa perturbation of nucleobases is an intrinsic property of the sequence context in DNA and RNA.

The pH titration and NMR studies (pH 6.6-12.5) in the heptameric isosequential ssDNA and ssRNA molecules, [d/r(5'-CAQ1GQ2AC-3', with variable Q1/Q2)], show that the pKa of the central G residue within the heptameric ssDNAs (DeltapKa = 0.67 +/- 0.03) and ssRNAs (DeltapKa = 0.49 +/- 0.02) is sequence-dependent. This variable pKa of the G clearly shows that its pseudoaromatic character, hence, its chemical reactivity, is strongly modulated and tuned by its sequence context. In contradistinction to the ssDNAs, the electrostatic transmission of the pKa of the G moiety to the neighboring A or C residues in the heptameric ssRNAs (as observed by the response of the aromatic marker protons of As or Cs) is found to be uniquely dependent upon the sequence composition. This demonstrates that the neighboring As or Cs in ssRNAs have variable electrostatic efficiency to interact with the central G/G-, which is owing to the variable pseudoaromatic characters (giving variable chemical reactivities) of the flanking As or Cs compared to those of the isosequential ssDNAs. The sequence-dependent variation of pKa of the central G and the modulation of its pKa transmission through the nearest-neighbors by variable electrostatic interaction is owing to the electronically coupled nature of the constituent nucleobases across the single strand, which demonstrates the unique chemical basis of the sequence context specificity of DNA or RNA in dictating the biological interaction, recognition, and function with any specific ligand.