G-quadruplexes within prion mRNA: the missing link in prion disease?

Abstract

Cellular ribonucleic acid (RNA) plays a crucial role in the initial conversion of cellular prion protein PrP(C) to infectious PrP(Sc) or scrapie. The nature of this RNA remains elusive. Previously, RNA aptamers against PrP(C) have been isolated and found to form G-quadruplexes (G4s). PrP(C) binding to G4 RNAs destabilizes its structure and is thought to trigger its conversion to PrP(Sc). Here it is shown that PrP messenger RNA (mRNA) itself contains several G4 motifs, located in the octarepeat region. Investigation of the RNA structure in one of these repeats by circular dichroism, nuclear magnetic resonance and ultraviolet melting studies shows evidence of G4 formation. In vitro translation of full-length PrP mRNA, naturally harboring five consecutive G4 motifs, was specifically affected by G4-binding ligands, lending support to G4 formation in PrP mRNA. A possible role of PrP binding to its own mRNA and the role of anti-prion drugs, many of which are G4-binding ligands, in prion disease are discussed.

Figure 1.

Putative G-quadruplex structures in human PrP mRNA. (A) Schematic representation of the mature PrP 23-230. Black boxes represent RNA-binding motifs. The peptide sequence of the five ORs is shown with the nucleotide sequence of the first three repeats depicted below. Guanines capable of forming G-quadruplexes are shown underlined/in red. (B) Cartoon showing the formation of G-quadruplexes structures in the presence of potassium ions. Hydrogen bonds between guanines are shown as dashed red lines. Connecting loops within G-quadruplexes are indicated by solid blue lines. (C) RNA constructs corresponding to different regions of the first OR which were used in this study.

Figure 2.

CD spectra of the four synthetic RNAs. RNAs were dissolved in 10-mM sodium phosphate buffer (pH 6.8) containing MgCl₂ or KCl, and their spectra recorded at 20°C. (A) Pri-One RNA. (B) Pri-Two RNA. (C) Pri-Hp RNA. (D) Pri-Qd RNA..

Figure 3.

1D proton NMR spectroscopy analysis of two synthetic RNAs. Spectra were recorded at 5°C. (A) Pri-Two RNA (60 μM) in the presence of 0.1 mM MgCl2. (B). Pri-Two RNA (60 μM) in the presence of 100 mM KCl. (C) Pri-Hp RNA (250 μM). (D) Pri-Hp RNA (250 μM) in the presence of 100 mM KCl.

Figure 4.

Thermal difference spectra and absorption profiles of Pri-One RNA in the absence and presence of KCl. (A) Normalized thermal difference spectra of Pri-One RNA in 10-mM sodium phosphate buffer (pH 6.7) with (blue dashed curve) or without (red curve) 50-mM KCl. (B) Normalized UV absorption profiles at 295 nm of Pri-One RNA in 10-mM sodium phosphate buffer (pH 6.7) with (blue dashed curves) or without (red curves) 50-mM KCl.

Figure 5.

Effect of potassium and G4 ligand on translation of PrP mRNA in vitro (A)–(C). SDS-PAGE of ³⁵S-methionine-labeled products from in vitro translation reactions using RRL. (A) PrP mRNA and FLuc mRNA were preincubated with or without KCl. As undiluted RRL contains ∼100-mM KCl, the final concentrations are 50, 75 and 100 mM. The relative yields of the bands corresponding to full-length PrP protein or FLuc protein are indicated. Standard deviation at 100-mM KCl was determined from four (PrP) and two (Fluc) experiments. (B) FLuc, PrP and PrP89-253 mRNA were preincubated for 1 h with or without PhenDC3 (final concentration 1 μM). Positions of the expected products are indicated at the right. Standard deviations were determined from two (FLuc and PrP89-253) and four (PrP) experiments. (C) Preincubation as under (B). Standard deviations were determined from five (Pri-RLuc) and four (RLuc and Mut-RLuc) experiments. (D) Final concentrations of KCl and PhenDC3 were 100 mM and 1 μM, respectively. The Y-axis shows the relative light units normalized to the reference samples without KCl or PhenDC3. Error bars represent the standard deviations of nine (Pri-RLuc), six (Mut-Rluc) and five (RLuc2) experiments.