Caulobacter crescentus CdnL is a non-essential RNA polymerase-binding protein whose depletion impairs normal growth and rRNA transcription

Abstract

CdnL is an essential RNA polymerase (RNAP)-binding activator of rRNA transcription in mycobacteria and myxobacteria but reportedly not in Bacillus. Whether its function and mode of action are conserved in other bacteria thus remains unclear. Because virtually all alphaproteobacteria have a CdnL homolog and none of these have been characterized, we studied the homolog (CdnL_Cc) of the model alphaproteobacterium Caulobacter crescentus. We show that CdnL_Cc is not essential for viability but that its absence or depletion causes slow growth and cell filamentation. CdnL_Cc is degraded in vivo in a manner dependent on its C-terminus, yet excess CdnL_Cc resulting from its stabilization did not adversely affect growth. We find that CdnL_Cc interacts with itself and with the RNAP β subunit, and localizes to at least one rRNA promoter in vivo, whose activity diminishes upon depletion of CdnL_Cc. Interestingly, cells expressing CdnL_Cc mutants unable to interact with the RNAP were cold-sensitive, suggesting that CdnL_Cc interaction with RNAP is especially required at lower than standard growth temperatures in C. crescentus. Our study indicates that despite limited sequence similarities and regulatory differences compared to its myco/myxobacterial homologs, CdnL_Cc may share similar biological functions, since it affects rRNA synthesis, probably by stabilizing open promoter-RNAP complexes.

Figure 1. Growth and cell morphology upon CdnL_Cc depletion.

(a) Growth of the indicated C. crescentus strains. Liquid cultures (OD₆₆₀ ∼ 0.5) were serially diluted, spotted (8 μl) on PYE plates with (+van) or without (−van) 0.5 mM vanillate, and incubated for 2 days at 30 °C. (b) Growth curves for the strains indicated (with symbols in parentheses). Freshly plated cells were innoculated into 10 ml of PYE (with 0.5 mM vanillate for ME4 and ME5) and grown at 30 °C to OD₆₆₀ ∼ 0.8. 50 μl were aliquoted (for ME4 and ME5, after washing three times with PYE to remove the vanillate) into 10 ml of fresh PYE (for ME4 and ME5, one with 0.5 mM vanillate and one without). Growth was monitored at the indicated times, and the average and error of three independent measurements is shown. (c) Cellular morphology of cells from (b). Samples (concentrated ten-fold in the case of ME5 cultivated in the absence of vanillate) of each culture in (b) after ∼20 h of growth were DAPI-stained and examined by DIC (differential interference contrast; top panels) and fluorescence microscopy (bottom panels), as described in Methods. Scale bar: 5 μm.

Figure 2. Analysis of CdnL_Cc protein stability in vivo.

(a) Stability of FLAG-CdnL_Cc, CdnL_Cc-FLAG and FLAG-CdnL_Cc(DD) in vivo. C. crescentus strains ME27, ME28 and ME29 were grown in PYE with vanillate, and subjected to a previously described protocol (see Methods) prior to immunoblot analysis (−xyl: no xylose; +xyl: 0.3% xylose to induce clpX* expression). Total protein from 1 ml aliquots withdrawn every 20 min was detected in immunoblots using anti-FLAG antibodies (left). On the right is a semi-log plot of the relative band intensities (mean of three independent experiments) versus time for FLAG-CdnL_Cc (−xyl, circles; +xyl, diamonds), CdnL_Cc-FLAG (−xyl, triangles) and FLAG-CdnL_Cc(DD) (−xyl, squares). Slopes of the linear fits shown yield the decay rate constants used to estimate half-lives. (b) FLAG-CdnL_Cc levels during the cell cycle. Swarmer cells (SW) from strain ME24 grown in M2G with vanillate were isolated and used for synchronized cell cycle progression (∼150 min doubling time). Total protein from 1 ml aliquots taken every 20 min was subjected to immunoblot analysis using anti-FLAG antibodies. The control CtrA was probed on a separate blot (since its gel mobility is close to that of FLAG-CdnL_Cc) using anti-CtrA antibodies and equivalent samples from the same experiment processed in parallel. Samples at 150 min correspond to SW and ST (with a small proportion of PD) cells isolated from the culture remaining at the end of this assay. A plot of band intensities (% of the maximum value) versus time is shown (right). In (a,b), positions of molecular size markers (kDa) are shown in blue to the right of cropped immunoblots. Note that FLAG-tagged CdnL_Cc (∼19.5 kDa calculated Mw) migrates slower than expected. (c) Growth of C. crescentus strains (ME41, ME39, ME8) expressing CdnL_Cc, CdnL_Cc(DD) or CdnL_Cc-FLAG, respectively, at the endogenous site. Liquid cultures (OD₆₆₀ ∼ 0.5) were serially diluted, spotted (8 μl) on PYE plates with or without vanillate, and incubated for 2 days at 30 °C. (d) Growth curves of strains in (c) expressing CdnL_Cc (squares, red), CdnL_Cc-FLAG (triangles, blue) or CdnL_Cc(DD) (circles, green) in PYE without vanillate at 30 °C.

Figure 3. CdnL_Cc interacts with itself, with RNAP, and with DNA.

(a) BACTH analysis of CdnL_Cc self-interaction in E. coli BTH101 transformed with plasmids pUT18-cdnL_Cc and pKT25-cdnL_Cc. (b) BACTH analysis of the interaction between CdnL_Cc (in pKT25) and C. crescentus RNAP β subunit fragments β_16-214 and β_16-523 (in pUT18C). In (a,b), the negative control (-) was pKT25 without insert. (c) Western blot of immunoprecipitated CdnL_Cc-FLAG probed for the presence of coprecipitating RNAPβ. Cells expressing CdnL_Cc-FLAG (strain ME17: ΔcdnL_Cc, P_van::cdnL_Cc-flag, vanR) or, as the negative control, untagged CdnL_Cc (strain ME5: ΔcdnL_Cc, P_van::cdnL_Cc, vanR) were immunoprecipitated with anti-FLAG agarose and processed in parallel, as described in Methods. Equal amounts of sample were then resolved by SDS-PAGE for immunoblot analysis. Monoclonal anti-RNAP β antibodies were used to detect RNAPβ (top) that coimmunoprecipitated with CdnL_Cc-FLAG, which was detected using anti-FLAG antibodies (bottom). Molecular size markers are shown to the right of the cropped immunoblots by the lines and corresponding values in kDa. (d) EMSA for the DNA binding of CdnL_Cc, CdnL_Tt, and CdnL_Mx. Reactions were performed as described in Methods with a 350-bp P_rrnA DNA probe, and with 1 μg of poly[dG-dC] or poly[dI-dC] added as nonspecific competitor.

Figure 4. CdnL_Cc localizes at an rRNA promoter and affects its transcription in vivo.

(a) CdnL_Cc binds to P_rrnA in vivo. ChIP-qPCR analysis using an anti-FLAG antibody on cells expressing CdnL_Cc-FLAG (strain ME17: ΔcdnL_Cc, P_van::cdnL_Cc-flag, vanR; unfilled bars) or CdnL_Cc as negative control (strain ME5: ΔcdnL_Cc, P_van::cdnL_Cc, vanR; black bars) showing CdnL_Cc-FLAG enrichment at P_rrnA in vivo, relative to an intergenic region. As positive controls, ChIP-qPCR analysis was carried out using anti-RNAP β or anti-σ^A monoclonal antibodies for enrichment of RNAP or σ⁷³, respectively, at P_rrnA. (b) Effects of CdnL_Cc depletion on P_rrnA promoter activity in vivo. CdnL_Cc was expressed from the P_van promoter in strain ME42. Cells grown in M2G with vanillate to OD₆₆₀~0.4 were washed and then grown in vanillate-free medium to block cdnL_Cc expression, and activity was measured by qRT-PCR at the times indicated. At each time point following vanillate withdrawal, one-half of the sample was washed and resuspended in medium without glucose and the other half remained untreated. Following 15 min incubation at 30 °C, the samples were collected for qRT-PCR analysis. (c) qRT-PCR analysis carried out with strain ME40, which expresses CdnL_Cc-FLAG from the P_van promoter, using a procedure identical to that in (b). Data shown in (a–c) correspond to the mean and standard error from three biological replicates. (d) Immunoblot analysis of CdnL_Cc-FLAG corresponding to samples in (c) with and without glucose deprivation and at the times indicated following vanillate withdrawal (top). As loading control, the same blot was probed using polyclonal anti-RNAP antibodies; the band corresponding to the RNAP β, β′ subunits is shown (bottom). Molecular size markers are shown to the right of the cropped immunoblots by lines and corresponding values in kDa.

Figure 5. Mutational analysis of CdnL_Cc-RNAP interaction in vivo.

(a) BACTH analysis of the interaction of CdnL_Cc mutants V39A, R52A and P54A (in pKT25) with C. crescentus RNAPβ fragment β_16-523 (in pUT18C). The negative control (–) bears empty pKT25 and pUT18C-β_16-523. (b) Schematic for the strategy employed to check for cdnL_Cc complementation in C. crescentus. A pMR3552 derivative with the required cdnL_Cc-flag allele (* indicates mutant) flanked by DNA segments upstream (grey) and downstream (black) of cdnL_Cc in the genome was introduced into strain ME5, which bears the ΔcdnL_Cc allele at the endogenous site and P_van-cdnL_Cc at a heterologous site. Merodiploids resulting from plasmid integration by recombination express both CdnL_Cc*-FLAG and CdnL_Cc in the presence of vanillate (+van) and only the former in the absence of vanillate (−van). (c) Complementation analysis in C. crescentus of cells bearing the ΔcdnL_Cc allele or ones expressing at the endogenous site C-terminal FLAG-tagged wild-type CdnL_Cc (WT) or the indicated N-terminal CdnL_Cc variants. PYE plates with (+van) or without (−van) vanillate were spotted with 8 μl of liquid cultures (OD₆₆₀ ∼ 0.5) at the dilutions indicated and incubated at 30 °C for two days or at 25 °C for three days. (d) Growth curves at 30 °C or 25 °C of C. crescentus expressing C-terminal FLAG-tagged CdnL_Cc (WT) or its indicated variants cultivated in liquid PYE without vanillate using the procedures described in Fig. 1b. (e) Cellular morphology examined by DIC microscopy of the wild-type (WT; scale bar: 5 μm) and the indicated mutant cells from (d) grown at 30 °C or 25 °C. (f) Immunoblot analysis to probe the stability of N-terminal CdnL_Cc variants. Cell extracts of strains expressing C-terminal FLAG-tagged CdnL_Cc (WT) or its indicated variants grown at 30 °C or 25 °C in PYE with vanillate were probed using anti-FLAG antibodies (top). As loading control, the same blot was probed using polyclonal anti-RNAP antibodies; the band corresponding to the RNAP β, β′ subunits is shown (bottom). Molecular size markers are shown to the right of the cropped immunoblots by lines and corresponding values in kDa (in blue).

Figure 6. Analysis of C-terminal CdnL_Cc mutations in vivo.

(a) Complementation analysis in C. crescentus of cells bearing the ΔcdnL_Cc allele or ones expressing at the endogenous site C-terminal FLAG-tagged wild-type CdnL_Cc (WT) or the indicated C-terminal CdnL_Cc variants. The analysis was carried out using the same procedures and conditions described in Fig. 5c. (b) Growth curves at 30 °C or 25 °C of C. crescentus strains expressing the C-terminal FLAG-tagged CdnL_Cc (WT) or its indicated variants cultivated in liquid PYE without vanillate using the procedures described in Fig. 1b. (c) Cellular morphology examined by DIC microscopy of the wild-type (WT; scale bar: 5 μm) and the indicated mutant cells from (b) grown at 30 °C or 25 °C. (d) Immunoblot analysis to probe the stability of C-terminal CdnL_Cc variants. Cell extracts of strains expressing C-terminal FLAG-tagged wild-type CdnL_Cc (WT) or the indicated CdnL_Cc mutants grown at 30 °C or 25 °C in PYE with vanillate were probed using anti-FLAG M2 antibodies (top). The negative control “Δ” corresponds to the strain (ME5) with the ΔcdnL_Cc allele at the endogenous site and expressing untagged CdnL_Cc under P_van. As loading control, the same blot was probed using polyclonal anti-RNAP antibodies; the band corresponding to the RNAP β, β′ subunits is shown (bottom). Molecular size markers are shown to the right of the cropped immunoblots by lines and corresponding values in kDa (in blue). (e) P_rrnA promoter activity in vivo at 30 °C or at 25 °C in cells expressing CdnL_Cc-FLAG (WT) or its variant with the R130A mutation under P_van control (strains ME40 and ME38, respectively). Cells grown overnight at 30 °C in M2G with vanillate were diluted into the same medium to OD₆₆₀ ∼ 0.1. One-half was grown at 30 °C and the other at 25 °C to OD₆₆₀ of 0.3–0.4, and RNA was quantitated using qRT-PCR. Data shown correspond to the mean and standard error from three biological replicates.