Mutagenesis of RpoE-like sigma factor genes in Bdellovibrio reveals differential control of groEL and two groES genes

Abstract

Background: Bdellovibrio bacteriovorus HD100 must regulate genes in response to a variety of environmental conditions as it enters, preys upon and leaves other bacteria, or grows axenically without prey. In addition to "housekeeping" sigma factors, its genome encodes several alternate sigma factors, including 2 Group IV-RpoE-like proteins, which may be involved in the complex regulation of its predatory lifestyle.

Results: We find that one sigma factor gene, bd3314, cannot be deleted from Bdellovibrio in either predatory or prey-independent growth states, and is therefore possibly essential, likely being an alternate sigma 70. Deletion of one of two Group IV-like sigma factor genes, bd0881, affects flagellar gene regulation and results in less efficient predation, although not due to motility changes; deletion of the second, bd0743, showed that it normally represses chaperone gene expression and intriguingly we find an alternative groES gene is expressed at timepoints in the predatory cycle where intensive protein synthesis at Bdellovibrio septation, prior to prey lysis, will be occurring.

Conclusions: We have taken the first step in understanding how alternate sigma factors regulate different processes in the predatory lifecycle of Bdellovibrio and discovered that alternate chaperones regulated by one of them are expressed at different stages of the lifecycle.

Figure 1

Sequence LOGO showing DNA binding region of RpoEs [8]. The first 35 sequences annotated as rpoE in the NCBI database were entered into the Weblogo program (http://weblogo.berkeley.edu/) using default parameters. The red arrows indicate the residues known to bind DNA in E. coli. The residues highlighted in red on the Bdellovibrio sequences show those that align to these using the ClustalW program and indicate that these are different from most RpoEs and each other, suggesting that they may well bind to different DNA motifs. There is also a 4 residue insertion in the Bd0743 sequence relative to the other sequences.

Figure 2

Predation efficiency assay using luminescent prey shows reduced efficiency for the ΔBd0881 mutant. Predatory efficiency plot showing log₁₀ initial ratios of prey to predator against time to reach half of starting luminescence for the strains. Equivalent numbers of the ΔBd0881 mutant Bdellovibrio killed the prey cells more slowly than ΔBd0743 or kanamycin resistant “reconstituted wild-type”, fliC1 merodiploid strain.

Figure 3

RT-PCR showing relative levels of transcription of chaperonin and flagellin genes in total RNA from attack phaseBdellovibrio, under normal and heat-shocked conditions. RT-PCR with transcript specific primers was carried out on matched concentrations of RNA (matched by Nanodrop spectrophotometer readings) from wild-type and mutant attack-phase Bdellovibrio including samples subjected to heat shock (42°C for 10 minutes). Total RNA samples from :-WT- wild-type HD100 attack phase, N- non-heat shocked 29°C, HS- heat shocked at 42°C for 10 minutes, 0881- ΔBd0881 attack phase, 0743- ΔBd0743 attack phase, Lane 7- no template negative control, Lane 8- HD100 genomic DNA positive control. “No reverse transcriptase” controls were performed for each template and were negative for DNA contamination (data not shown). The abundant transcript produced using primers designed to anneal to the fliC1 gene acts as a positive control by showing that there was ample total RNA in all samples.

Figure 4

Lengths of flagella and swimming speeds of the mutants and wild-type. A- Flagellar length of wild type and sigma factor mutants measured from electron micrographs, error bars show 95% confidence intervals. B- Speeds of wild type and mutant predatory strains measured by the Hobson Bactracker, error bars show 95% confidence intervals.

Figure 5

Expression patterns ofrpoE-like genes compared tofliC3in total RNA taken from across the predatory cycle studied by RT-PCR. RT-PCR with transcript-specific primers on total RNA prepared from identical numbers of B. bacteriovorus HD100 predator synchronously invading an E. coli S17-1 prey culture, with samples taken as the predatory infection, and Bdellovibrio development proceeds across a time course. L- NEB 100 bp ladder, AP- attack-phase 15–15 minutes predation, 30–30 minutes predation, 45–45 minutes predation 1-4 h: 1,2,3,4 hours predation respectively. Controls of no template, no reverse transcriptase, E. coli S17-1 only RNA as template and B. bacteriovorus HD100 genomic DNA were carried out. Primers designed to bd0743 give a product in every sample, thus act as a positive control for the RNA, validating the lack of expression in some of the samples. A similar expression pattern was seen for bd0881 and fliC3.

Figure 6

Transcriptional expression patterns of the threeBdellovibriochaperonin genes across the wild type predatory cycle. RT-PCR with transcript-specific primers was performed on total RNA prepared from identical volumes of B. bacteriovorus HD100 predator with E. coli S17-1 prey infection culture as the predatory infection proceeds across a time course. L- NEB 100 bp ladder, AP- attack-phase 15–15 minutes predation, 30–30 minutes predation, 45–45 minutes predation 1-4 h: 1,2,3,4 hours predation respectively. Controls of no template, no reverse transcriptase, E. coli S17-1 only RNA as template and bacteriovorus HD100 genomic DNA were carried out. Primers designed to bd0097 groES1 and bd0099 groEL give a product in every sample, thus act as a positive control for the RNA, validating the lack of expression of bd3349 groES2 in the earlier part of the infectious cycle.

Figure 7

Transcriptional expression patterns of the threeBdellovibriochaperonin genes during axenic Host-Independent growth. RT-PCR with transcript specific primers was carried out on matched concentrations of RNA (matched by Nanodrop spectrophotometer readings) from axenically grown Host-Independent Bdellovibrio. Three independently isolated strains of each sigma factor mutant and each host-independent (HI) wild-type were used to account for HI strain-to strain variation. L- NEB 100 bp ladder –ve - no template negative control + ve- HD100 genomic DNA positive control.