Bacillus subtilis ResD induces expression of the potential regulatory genes yclJK upon oxygen limitation

Abstract

Transcription of the yclJK operon, which encodes a potential two-component regulatory system, is activated in response to oxygen limitation in Bacillus subtilis. Northern blot analysis and assays of yclJ-lacZ reporter gene fusion activity revealed that the anaerobic induction is dependent on another two-component signal transduction system encoded by resDE. ResDE was previously shown to be required for the induction of anaerobic energy metabolism. Electrophoretic mobility shift assays and DNase I footprinting experiments showed that the response regulator ResD binds specifically to the yclJK regulatory region upstream of the transcriptional start site. In vitro transcription experiments demonstrated that ResD is sufficient to activate yclJ transcription. The phosphorylation of ResD by its sensor kinase, ResE, highly stimulates its activity as a transcriptional activator. Multiple nucleotide substitutions in the ResD binding regions of the yclJ promoter abolished ResD binding in vitro and prevented the anaerobic induction of yclJK in vivo. A weight matrix for the ResD binding site was defined by a bioinformatic approach. The results obtained suggest the existence of a new branch of the complex regulatory system employed for the adaptation of B. subtilis to anaerobic growth conditions.

FIG. 1.

Anaerobic expression of yclJK operon is ResDE dependent. Total RNAs were extracted from wild-type strain JH642 and the resDE mutant strain LAB2135 grown under the following growth conditions: aerobic (1), fermentative (2), anaerobic plus nitrate (3), and anaerobic plus nitrite (4). RNAs (10 μg) were separated in a 1% denaturing agarose gel and analyzed by Northern blotting. A yclJ-specific RNA probe was used for hybridization. A single transcript of 2.1 kb was detected, which corresponds to the size of a yclJK transcript. Ethidium bromide staining of the gel showed that equal amounts of RNA were analyzed. The size standards are RNA molecular weight marker no. 1 (Roche Diagnostics GmbH) and the 16S and 23S rRNA species.

FIG. 2.

Determination of transcription start site of yclJK by primer extension analysis. The total RNA was isolated and analyzed from JH642 cells grown aerobically (1) and under fermentative conditions (2). The same primer used for the primer extension analysis was used for sequencing reactions (lanes G, A, T, and C). Arrows indicate the primer extension products and asterisks mark the 5′ end of the yclJK mRNA in the sequence.

FIG. 3.

Expression of yclJ-lacZ in various regulatory mutant strains. β-Galactosidase activities were measured in the JH642 wild-type strain and resDE, resE, fnr, and yclJ mutant strains after 3 h of culture under the following growth conditions: aerobic (1), fermentative (2), anaerobic plus nitrate (3), and anaerobic plus nitrite (4). Error bars indicate standard deviations (n ≥ 3).

FIG. 4.

EMSAs to detect binding of ResD to yclJ promoter. (A) An end-labeled DNA fragment containing the yclJ promoter was incubated with ResD in the presence (ResD-P) or absence (ResD) of 0.5 μM ResE. The amounts of ResD used were 0.06 μM (lanes 2 and 6), 0.13 μM (lanes 3 and 7), 0.25 μM (lanes 4 and 8), and 0.5 μM (lanes 5 and 9). Lane 1, probe only. (B) ResD does not bind to yclJ promoters carrying mutations in the putative ResD box. The same DNA fragments containing the wild-type and mutant promoters were used to examine interactions with ResD phosphorylated with 0.5 μM ResE. The ResD concentrations used were 0.13 μM (lanes 2, 5, and 8), 0.25 μM (lanes 3, 6, and 9), and 0.5 μM (lanes 4, 7, and 10). Lane 1, probe only. (C) Competition experiment with cold DNA fragment containing the wild-type and mutant promoters. The labeled yclJ promoter fragment was incubated with 0.5 μM ResD and ResE (lane 2). Increasing amounts of cold DNA, i.e., 2.5 nM (lanes 3, 6, and 9), 5 nM (lanes 4, 7, and 10), and 10 nM (lanes 5, 8, and 11), were included as competitor DNA in the reaction mixtures. Lane 1, probe only. The image shows the effect of the mutations in the putative ResD box on binding by ResD. Probes muta and mutb indicate the yclJ promoter carrying the binding site a and binding site b mutations, respectively.

FIG. 5.

DNase I footprinting experiment with ResD and the yclJ promoter. Increasing concentrations of ResD and/or ResE (2, 4, 8, and 16 μM) were incubated with ³²P-end-labeled coding and noncoding strands of the yclJ promoter. G and A sequencing ladders are included to localize the binding sites. The vertical brackets indicate protected regions. Dotted brackets show weakly protected regions. Positions relative to the transcription start site are shown.

FIG. 6.

In vitro transcription analysis of yclJ promoter. Transcription was carried out with 25 nM purified RNA polymerase and 5 nM templates without ResD and ResE (lane 1), with increasing amounts of wild-type ResD (lanes 2 to 4), with wild-type ResD and ResE (lanes 5 to 7), with the D57A mutant ResD (lanes 8 to 10), with the D57A mutant ResD and ResE (lanes 11 to 13), and with ResE only (lane 14). The amounts of ResD used were 0.35, 0.7, and 1.4 μM, and the amount of ResE used was 1 μM. An RNA size marker (in nucleotides) is shown (M).

FIG. 7.

New definition of ResD binding site and location in the yclJ promoter. (A) Sequence logo of the ResD binding site based on the information weight matrix model. The height of each stack of letters is the sequence conservation, measured in bits of information according to the equation given in Materials and Methods. The height of each letter within a stack is proportional to its frequency at that position in the binding site. The letters are sorted, with the most frequent on top. (B) ResD-dependent promoter of yclJ. The transcription start site obtained from primer extension analysis is marked “+1.” Potential ResD binding sites, a and b, are boxed, and their positions with respect to the transcriptional start sites are given. The solid line marks the protected region from the footprinting experiments; the dashed lines mark weakly protected regions. RBS, ribosome-binding site.

FIG. 8.

Mutations in binding sites a and b prevent expression of yclJ-lacZ. β-Galactosidase activities were measured from yclJ-lacZ wild-type and mutant promoter constructs in the JH642 wild-type strain after 3 h of culture under the following growth conditions: aerobic (1), fermentative (2), anaerobic plus nitrate (3), and anaerobic plus nitrite (4). Error bars indicate standard deviations (n ≥ 3).