Model Uracil-Rich RNAs and Membrane Protein mRNAs Interact Specifically with Cold Shock Proteins in Escherichia coli

Abstract

Are integral membrane protein-encoding mRNAs (MPRs) different from other mRNAs such as those encoding cytosolic mRNAs (CPRs)? This is implied from the emerging concept that MPRs are specifically recognized and delivered to membrane-bound ribosomes in a translation-independent manner. MPRs might be recognized through uracil-rich segments that encode hydrophobic transmembrane helices. To investigate this hypothesis, we designed DNA sequences encoding model untranslatable transcripts that mimic MPRs or CPRs. By utilizing in vitro-synthesized biotinylated RNAs mixed with Escherichia coli extracts, we identified a highly specific interaction that takes place between transcripts that mimic MPRs and the cold shock proteins CspE and CspC, which are normally expressed under physiological conditions. Co-purification studies with E. coli expressing 6His-tagged CspE or CspC confirmed that the specific interaction occurs in vivo not only with the model uracil-rich untranslatable transcripts but also with endogenous MPRs. Our results suggest that the evolutionarily conserved cold shock proteins may have a role, possibly as promiscuous chaperons, in the biogenesis of MPRs.

Fig 1. Characterization of 4 model untranslatable RNAs.

(A) Schematic representation of the Ra-Rd encoding genes [see text, S1 Fig]. (B) Uracil content of the model transcripts, utilizing a sliding window of 55 nucleotides as calculated by the software DNA Strider. (C) Wild type E. coli expressing Ra or Rb were disrupted by sonication and cell extracts were fractionated by sucrose density gradient (a representative gradient is shown). The gradient fractions were analyzed for RNA content (A₂₆₀). The indicated free and ribosomal fractions were pooled. (D) The contents of the indicated R transcripts and endogenous, translatable transcripts encoding PrfA and RpoD as controls, were measured by qPCR in the pooled free and ribosomal fractions. Error bars indicate SEM (n = 3).

Fig 2. Identification of protein-U-rich RNA interactions.

Ra and Rb were synthesized in vitro with or without biotinylated CTP. The RNAs were incubated with E. coli extract (S300) in the presence of 15 mM MgCl₂ and then trapped by streptavidin beads. After extensive wash, the bound material was released in urea and SDS-containing buffer, and analyzed by (A) RNA-PAGE, stained with ethidium bromide and (B) protein Tris-tricine SDS-PAGE, stained with Instant blue. (*) Indicates specific U-rich RNA-binding proteins. These proteins were identified by mass spectrometry as CspE and CspC. (C) Biotinylated Ra-Rd were incubated with purified 6His-CspE or 6His-CspC in the presence of 15 mM Mg²⁺. RNA-protein complexes were trapped by streptavidin beads, eluted with SDS, separated by tris-tricine SDS-PAGE, and the gels were stained with Instant blue. Quantitation was performed by densitometry and expressed as percentage of the proteins that were eluted with Ra. (D) In vitro synthesized biotinylated Ra-Rd were incubated with an E. coli S300 fraction in the presence of 2 or 20 mM Mg⁺². Uper panel, Eluted material was separated by tris-tricine SDS-PAGE and stained with instant blue. Lower panel, Western blot analysis of the eluted material with anti-CspE antibodies. (E) Biotinylated Ra and Rb (0.17 μM) were incubated with increasing concentrations of purified 6His-CspC (0.17–2.2 μM). RNA-protein complexes were trapped by streptavidin beads, and after wash treated with RNaseA and eluted with SDS buffer (see methods). Protein samples containing the same amount of RNA were separated by tris-tricine SDS-PAGE, and analyzed by Western blotting with anti CspE antibodies. (F) Quantitation of protein bands shown in E.

Fig 3. 6His-CspE/C pull-down experiments.

E. coli cells co-expressing Ra, or Rb, or Rc, or Rd together with 6His-CspE or 6His-CspC were lysed, and extracts were incubated with Talon resin, for immobilizing 6His-CspE or 6His-CspC and their bound RNAs. (A) Samples from the various purification steps were analyzed by Westerm blotting with anti CspE antibodies. (B) Left panel, the eluates were treated with DNase or RNase and analyzed by Agarose gel. Right panel, the quality and size of the eluted RNA were analyzed by tapestation. (C) The total extract and the CspE/CspC bound RNAs were analyzed by qPCR with the corresponding primers to Ra, Rb, Rc, or Rd. Primers to rrl, rrs, and rnpB were utilized as controls. Values were calculated as 2^{extract Ct} / 2^{pull-down Ct} (see methods). Error bars indicate SEM (n = 3). (D) RNA was extracted from disrupted cells and the total steady state amount of R transcripts was measured by qPCR. Endogenous RNAs ssrA and rnpB were used as controls to assure unbiased results. Since both of the controls were similarly expressed in the various samples, we chose rnpB expression as a standard for calculating the relative quantity (RQ) of each R transcripts. The RQ of Ra is defined as = 1. (E) 6His-CspE (upper panel) or 6His-CspC (lower panel) were purified from wild type E. coli extracts with Talon resin. The total extracts and the eluates were analyzed by qPCR with primers to various MPRs and CPRs and analyzed as in (C). Error bars indicate SEM (n = 3). Right panels, an average ratio of [bound]/[total] is shown for each experiment.

Fig 4. High throughput sequencing of endogenous RNAs that co-purify with 6His-CspE.

(A) Wild type E. coli expressing CspE-6His were disrupted in the presence of either 2 mM or 15 mM [Mg²⁺] (top and bottom panels, respectively) and the total cell extracts were subjected to metal affinity chromatography using Talon resin. RNA was prepared from the total cell extracts and the imidazole-eluted material (see Fig 3) and subjected to high throughput sequencing. The amount of CspE-6His bound MPRs (left panels) and CPRs (right panels) is plotted as a function of the amount of the same mRNAs in the total extract. (B) CspE-binding of all detected mRNAs was calculated as [RPKM_CspE-bound / RPKM_extract]. The quota of MPRs and CPRs in each 10^th percentile along the CspE-association landscape is presented as a moving average plot. (C) This panel shows the CspE-binding values in the presence of 2 mM [Mg²⁺] for selected MPRs and CPRs that were similarly analyzed by qPCR (see Fig 3D for comparison). (D) An independent experiment that shows the CspE-binding values for selected MPRs and CPRs in the presence of 15 mM [Mg²⁺].

Fig 5. Characterization of CspE-, CspC-, and CspC/E deleted cells.

(A) PCR analysis of the cspE and cspC in E. coli ΔcspE, ΔcspC, or ΔcspCΔcspE::kan ^R. (B) Western blot analysis with anti-CspE antibodies of total extracts from wild type and the deleted E. coli strains. (C) (D) In vitro synthesized biotinylated Ra or Rb were incubated with S300 fractions of wild type E. coli or E. coli ΔcspE, or E. coli ΔcspE ΔcspC in the presence of 2 mM Mg²⁺. The RNA-protein complexes were purified by streptavidin beads, eluted with SDS buffer and separated by tris-tricine SDS-PAGE. The gels were stained with Instant blue.