RNA chaperone activity and RNA-binding properties of the E. coli protein StpA

Abstract

The E. coli protein StpA has RNA annealing and strand displacement activities and it promotes folding of RNAs by loosening their structures. To understand the mode of action of StpA, we analysed the relationship of its RNA chaperone activity to its RNA-binding properties. For acceleration of annealing of two short RNAs, StpA binds both molecules simultaneously, showing that annealing is promoted by crowding. StpA binds weakly to RNA with a preference for unstructured molecules. Binding of StpA to RNA is strongly dependent on the ionic strength, suggesting that the interactions are mainly electrostatic. A mutant variant of the protein, with a glycine to valine change in the nucleic-acid-binding domain, displays weaker RNA binding but higher RNA chaperone activity. This suggests that the RNA chaperone activity of StpA results from weak and transient interactions rather than from tight binding to RNA. We further discuss the role that structural disorder in proteins may play in chaperoning RNA folding, using bioinformatic sequence analysis tools, and provide evidence for the importance of conformational disorder and local structural preformation of chaperone nucleic-acid-binding sites.

Figure 1.

Splicing behaviour of td pre-mRNA constructs with varying exon lengths. (A) Schematic representation of the different constructs; exon 1 constructs are 216, 113 and 27 nt long; exon 2 constructs are either 2 or 48 nt long. (B) Splicing assays performed with different exon constructs show the influence of exon length on splicing and folding of the td intron. The splicing reaction is induced by the addition of GTP, in the presence of 5 mM Mg²⁺ at 37°C. The decrease in pre-mRNA versus time is indicated.

Figure 2.

StpA-induced splicing of the td pre-mRNA. (A) Splicing assay of the sho-sho RNA in the presence of increasing amounts of StpA. The reactions were performed at 37°C as described in experimental procedures. Increasing the amount of the RNA chaperone leads to an increase of the fast-reacting RNA population with a peak activity at 1.4 µM StpA. Higher StpA concentrations again reduce the activity. (B) Comparison of the splicing behaviour of the sho-sho RNA in the absence and after addition of StpA to the splicing reaction after 10 min of incubation. StpA causes a burst of activity immediately after addition.

Figure 3.

RNA-binding assays (A) Different RNA constructs were incubated with increasing amounts of protein (1 pM to 4 µM). The td ribozyme construct lacks exon sequences and misses 7 nt of the 5′ end of the intron and 5 nt at the 3′ end; (B) Influence of mono- and divalent metal ions on the binding behaviour of the RNA chaperone to td ribozyme RNA. Increasing amounts of mono- and divalent ions lead to a drastic drop in the binding efficiency of StpA.

Figure 4.

In vitro selection of StpA- and Hfq-binding RNAs from a genomic E. coli library. The percentage of recovered RNA in relation to the input RNA is shown for both proteins for eight consecutive selection cycles. For the Hfq, used here as a positive control, RNAs are enriched and increasing amounts of the input RNA were recovered. No StpA-binding RNAs could be selected and recovery remained at background levels throughout the procedure.

Figure 5.

StpA promotes RNA annealing by simultaneoulsly binding two RNAs. In a microplate reader, 10 nM each of two fluorophore-labelled oligoribonucleotides were injected, mixed and then incubated at 37°C in the absence or presence of 1 µM StpA. With the Cy3 donor dye excited, fluorescence emissions of Cy3 and Cy5 were measured every second, and the FRET index was calculated as a ratio of acceptor to donor fluorescence. (A) To assay RNA annealing, fully complementary RNA 21 mers (21R⁺, 21R⁻) were used that comprise no significant secondary structures and therefore can anneal by themselves. The presence of 1 µM StpA accelerates this reaction 4-fold. (B) Using a non-complementary RNA pair (21R⁺, duplex⁻) the two 21 mers alone cannot hybridize. However, with StpA present, fluorescence resonance energy transfer (FRET) takes place, indicating simultaneous binding of both RNAs to the protein. Graphs were fitted with the second-order reaction equation for equimolar initial reactant concentrations.

Figure 6.

StpA mutant and domains. (A) Schematic representation of the different protein constructs used to specify the regions responsible for the RNA chaperone activity of StpA. The N-terminal (NH₂–StpA) and the C-terminal domains (COOH–StpA) as well as a mutant with a glycine to valine change at position 126 (G126V–StpA) in the DNA-binding domain were prepared. (B) In splicing assays, 1.4 µM of the different protein constructs were tested for RNA chaperone activity with 0.5 pM of the sho-sho pre-mRNA. The graphs show the decrease of pre-mRNA during splicing.

Figure 7.

Equilibrium filter-binding assays with wt and mutant StpA. (A) Increasing amounts of wt StpA or G126V–StpA mutant (50 pM to 15 µM) were bound to 100 pM of sho-sho pre-mRNA construct of wild-type intron sequence or (B) to a sho-sho pre-mRNA construct with a mutation (C865U) in the joining region of J3/4 of the intron RNA. This mutant has a destabilized tertiary structure (14).

Figure 8.

Residue compactness and secondary structure plot of StpA. Predicted compactness and local secondary structural features of StpA are shown as a function of residue position. Large compactness values (black) indicate residue positions typically buried in the interior of the 3D structure, whereas small values are found for residues exposed to the solvent. Corresponding local secondary structure elements (red) are overlayed. Positive values are indicative of α-helical segments. In contrast, continuous negative values are typical for extended or β-strand regions.