Caulobacter crescentus RNase E condensation contributes to autoregulation and fitness

Abstract

RNase E is the most common RNA decay nuclease in bacteria, setting the global mRNA decay rate and scaffolding formation of the RNA degradosome complex and BR-bodies. To properly set the global mRNA decay rate, RNase E from Escherichia coli and neighboring γ-proteobacteria were found to autoregulate RNase E levels via the decay of its mRNA's 5' untranslated region (UTR). While the 5' UTR is absent from other groups of bacteria in the Rfam database, we identified that the α-proteobacterium Caulobacter crescentus RNase E contains a similar 5' UTR structure that promotes RNase E autoregulation. In both bacteria, the C-terminal intrinsically disordered region (IDR) of RNase E is required for proper autoregulation to occur, and this IDR is also necessary and sufficient for RNase E to phase-separate, generating BR-bodies. Using in vitro purified RNase E, we find that the IDR's ability to promote phase separation correlates with enhanced 5' UTR cleavage, suggesting that phase separation of RNase E with the 5' UTR enhances autoregulation. Finally, using growth competition experiments, we find that a strain capable of autoregulation rapidly outcompetes a strain with a 5' UTR mutation that cannot autoregulate, suggesting autoregulation promotes optimal cellular fitness.

FIGURE 1:

Depletion of Caulobacter RNase E leads to slow growth and loss in colony formation. (A) Growth curves of depletion strains of RNase E (triangles) and empty vector controls (squares) in the presence of xylose (black) or the absence of xylose (light gray). All cells were grown in PYE with kanamycin and each timepoint is the average OD600 measured from three replicate cultures. Error bars represent SD. (B) Colony forming unit assay to determine cell viability in the depletion strains. Depletion strains were grown in PYE/Kan in the presence of xylose (black) or absence of xylose (light gray) for 8 h, then spotted on PYE/Kan/Xyl plates. Three replicates of the experiment were performed, and one was chosen as a representative image. EV = Empty vector; RNE = RNase E.

FIGURE 2:

Overexpression of Caulobacter RNase E leads to growth arrest. (A) Growth curves of overexpression strains of RNase E (triangles), and empty vector controls (squares) grown in the presence of xylose (black) or the absence of xylose (light gray). All cells were grown in PYE with kanamycin and each time point is the average OD600 measured from three replicate cultures. Error bars represent SD. (B) Colony forming unit assay to determine cell viability in the overexpression strains harboring pBX vectors with the insert on the left. Three replicates of the experiment were performed, and one was chosen as a representative image.

FIGURE 3:

Caulobacter RNase E autoregulation requires activity on its 5′ UTR. (A) Western blot of RNase E in the indicated strains. Vanillate-induced RNase E lacks the 5′ UTR. RNase E-YFP and RNase E bands are indicated. Three replicates of the experiment were performed, and one was chosen as a representative image. (B) Predicted mRNA secondary structures generated by turbofold for C. crescentus (left) and for E. coli (right). RNA cleavage sites are mapped with arrows. RNA decay sites identification in the 5′ UTR. 5′ P-sites from (Zhou et al., 2015) and RNA-seq data indicating the mRNA 5′ UTR from (Schrader et al., 2014).

FIGURE 4:

Caulobacter RNase E autoregulation contributes to cellular fitness. (A) Cartoon of strains JS249 containing the rne 5′ UTR, and JS38 which has a 5′ UTR from a plasmid expression system. YFP intensity distributions of strains JS249 and JS38, Gaussian fits were performed with kaleidagraph from >100 cells each. (Right) Doubling times of JS38 and JS249 as grown in PYE gent vanillate media. The average value is shown, with the error bars representing the SD of doubling times calculated from three replicate growth curves. A one-tailed t test with uneven variance yielded a P value of 0.025. (B) Growth competition of equally mixed cultures containing JS249 and JS38 after multiple days of growth. Error bars are from three biological replicates of the growth competition. The fraction of cells were calculated from the relative area under the curve from a dual Gaussian curve fit as shown in panel C based on a YFP quantitation >100 cells per replicate. (C) YFP distributions of the mixed cultures at the beginning of the growth competition or after 3 d of culture. Dual Gaussian fits were performed using kaleidagraph on >100 cells. The relative area under the JS249 peak compared with the area under the JS38 peak was calculated as the ratio of JS249/JS38.

FIGURE 5:

Caulobacter RNase E 5′ UTR cleavage is stimulated by the IDR. (A) Purified solutions of full-length RNase E containing the IDR (left) and the ΔIDR variant (right) after incubated with RNA for 30 min. Scale bar is 5 µm. Three replicates of the experiment were performed, and one was chosen as a representative image. (B) RNase E 5′ UTR cleavage assay. RNase E’s 5′ UTR was incubated with either full-length RNase E or the ΔIDR variant for the indicated time periods before running on a 7% denaturing PAGE and stained by SYBR gold. Each gel is a representative gel of three independent replicates.

FIGURE 6:

Caulobacter IDR mutants show a correlation between RNase E 5′ UTR cleavage rate and foci formation of YFP-fusions. The RNase E 5′ UTR half-life measured in vivo by rifampicin qRT-PCR time course is correlated with the YFP-foci per cell of various RNase E IDR mutants (data measured previously in Al-Husini et al. Mol Cell 2018 based on triplicate half-life measurements and YFP quantitation from >50 cells). The indicated species of the NTD-IDR fusions is indicated in lower case (CC = C. crescentus, AT = A. tumefaciens, and SM = S. meliloti). The ΔDBS mutant has deletions for the binding sites for RNase D, PNPase, and Aconitase in the IDR. The Y-axis is plotted on a log₂ scale, and the curve fit and R² was generated from an exponential equation in Microsoft excel.