β-Galactosidase activity, determined by monitoring the cps::lacZ reporter gene. (A) ST001 (wild-type) cells were grown overnight in EB medium at 37°C. Aliquots were inoculated into fresh EB medium with and without 0.4% glucose and incubated at 37°C for 1.5 h, and then the cultures were transferred to a lower temperature (20°C). During growth at the lower temperature, β-galactosidase activities were monitored. As references, mutant derivatives of ST001 (ΔrcsC and ΔyojN) were also examined. (Insets) ST001 cells were grown on EB medium at 20°C (on agar plates) in the presence and absence of glucose (0.4%) for 48 h. The resulting colonies were photographed. (B) Growth curves for the cells which were examined in theexperiments whose results are shown in panel A. The symbols are the same as those used in panel A. OD₆₆₀, optical density at 660 nm. (C) Culture incubated at 37°C as described above, as a control. (D) Growth curves for the cells which were examined in the experiment whose results are shown in panel C. The experiments were carried out at least three times, and the results of representative experiments are shown. (E) ST001 cells were grown at 20°C for both 8 and 12 h as described above in EB medium supplemented with either glucose (Glu), fructose (Fru), galactose (Gal), pyruvate (Pyr), or maltose (Mal) at a concentration of 0.4%. Then β-galactosidase activities were measured. Under both experimental conditions consistent results were obtained, and representative data (12 h) are shown.

β-Galactosidase activity, determined by monitoring the cps::lacZ reporter gene. (A) ST001 (wild type) and a derivative of this strain (ΔrcsC) were grown as described in the legend to Fig. 1. In these experiments, glucose and/or zinc (1 mM ZnCl₂) was added, as indicated. β-Galactosidase activities were measured during growth at 20°C. The experiments were carried out at least three times, and the results of representative experiments are shown. (B) Growth curves for the cells which were examined in the experiment whose results are shown in panel A. The symbols are the same as those used in panel A. OD₆₆₀, optical density at 660 nm. (C) ST001 cells were grown at 20°C for both 8 and 12 h (see panel D) in glucose-EB medium containing a divalent cation (1 mM ZnCl₂, 1 mM CuSO₄, 1 mM NiCl₂, 10 mM CaCl₂, or 30 mM MgCl₂). Then β-galactosidase activities were measured. Under both experimental conditions consistent results were obtained, and representative data (12 h) are shown. (D) Growth curves for the cells which were examined in the experiment whose results are shown in panel A. For an explanation of the symbols, see panel C.

β-Galactosidase activity, determined by monitoring the cps::lacZ reporter gene. (A) ST001 (wild type [WT]) and mutant derivatives of this strain (ΔrcsC, ΔyojN, and ΔrcsB) were transformed with a multicopy plasmid harboring the rcsF gene. The resulting cells (+plasmid), together with the control cells (WT), were grown in glucose-EB medium for 8 and 12 h at 37°C. Then β-galactosidase activities were measured. Under both experimental conditions, consistent results were obtained, and representative data (8 h) are shown. (B) ST001 (wild type) and mutant derivatives of this strain (ΔrcsC and ΔrcsF) were grown for 12 h at 20°C in glucose-EB medium with and without zinc, as described in the legend to Fig. 1. Then β-galactosidase activities were measured. Under both experimental conditions consistent results were obtained, and representative data (12 h) are shown.

Schematic representation of microarray data obtained with mutants with mutations in the Rcs signaling system. Three types of microarray analyses were carried out. First, the transcriptome of ST001 (wild-type) cells grown for 8 h at 20°C in EB medium containing 0.4% glucose and 1 mM ZnCl₂ (labeled with Cy5) was compared with that of cells grown in the absence of zinc (labeled with Cy3). The data analyses revealed that 213 genes showed mean values for Cy5/Cy3 that were greater than 2. These genes were assumed to be the genes whose expression was up-regulated by zinc (designated zinc-up genes). Second, the transcriptome of ΔrcsC cells grown at 20°C in EB medium containing 0.4% glucose and 1 mM ZnCl₂ (labeled with Cy5) was compared with that of ST001 (wild-type) cells grown under the same conditions (labeled with Cy3). The data analyses revealed that 58 genes showed mean values for Cy5/Cy3 that were less than 0.5. These genes were assumed to be the genes whose expression was down-regulated in the ΔrcsC background (designated ΔrcsC-down genes). Similarly, 120 ΔrcsF-down genes were identified. Based on the microarray data, each overlapping set of genes was analyzed, as shown in panels A and B. These independent analyses revealed the putative Rcs regulon members, shown in panel C. (D) Putative Rcs regulon genes. The 17 cps genes are located together in cps operons on the E. coli chromosome. The 15 non-cps genes are located at different positions on the chromosome, although the bdm and osmC genes and the spy and ydjR genes are located next to each other, as indicated by arrows. The microarray data used in this study are available as supplemental data at the following web site: http://www.agr.nagoya-u.ac.jp/%7emicrobio/. Also, to gain additional ideas for each gene, see the web site of Colibri (http://www.genolist.pasteur.fr/Colibri/). CPS, capsular polysaccharide.

Putative RcsA/RcsB recognition sequences in the Rcs regulon genes and β-galactosidase activity as determined by monitoring the katE::lacZ reporter gene. (A) Using the previously proposed RcsA/RcsB recognition consensus sequence (TAAGAATATTCCTA) (47), we searched for nucleotide sequences similar to the consensus sequence in each promoter region of the non-cps genes (see Fig. 4D). The numbers in parentheses indicate the position (last nucleotide) of the 14-nucleotide sequence; the first nucleotide of the inferred ATG initiation codon of the gene in question was defined as position 1. (B) Strain GY37 carries the katE::lacZ reporter gene (49). This wild-type strain and mutant derivatives (ΔrcsC, ΔyojN, and ΔrcsB) were grown in EB medium containing 0.4% glucose and 1 mM ZnCl₂ at 20°C, as described in the legend to Fig. 1. β-Galactosidase activities were measured during growth.

β-Galactosidase activity, determined by monitoring the cps::lacZ reporter gene in a set of 37 two-component system mutants. ST001 (wild type) and 37 two-component system mutants (each designated by a gene designation[s]) were grown in EB medium containing 0.4% glucose and 1 mM ZnCl₂ at 20°C for 12 h, as described in the legend to Fig. 1. Then β-galactosidase activities were measured. The experiments were carried out three times, and the data shown are means. The basS/R mutant cells did not grow well in the medium used. ND, not determined.

β-Galactosidase activity, determined by monitoring the cps::lacZ reporter gene in two-component mutant backgrounds. ST001 (wild type) and mutant derivatives of this strain were grown in EB medium containing 0.4% glucose and 1 mM ZnCl₂ at 20°C, as described in the legend to Fig. 1. Then the β-galactosidase activities were measured at intervals. (A) ST001 (wild type); (B) ST001(ΔrcsC); (C) ST001(ΔarcA); (D) ST001[Δ(envZ/ompR)]; (E) ST001ΔphoQ/P; (F) ΔphoQ; (G) ΔphoP.

β-Galactosidase activity, determined by monitoring the cps::lacZ reporter gene with cells grown in the presence of both magnesium and zinc. (A) ST001 cells were grown at 20°C in EB containing 0.4% glucose and 1 mM ZnCl₂ with and without 30 mM MgCl₂ (also see the legend to Fig. 1). During growth, β-galactosidase activities were measured. (B) ST001 cells were grown at 20°C in EB containing 0.4% glucose and 1 mM ZnCl₂, as well as different concentrations of MgCl₂. After 8 and 12 h of incubation, β-galactosidase activities were measured. Under both experimental conditions consistent results were obtained, and representative data (12 h) are shown.

β-Galactosidase activity, determined by monitoring the mgrB::lacZ reporter gene with cells grown in the presence of both magnesium and zinc. (A) MG1301 cells carrying the mgrB::lacZ reporter gene were grown at 20°C in EB medium containing 0.4% glucose with and without 1 mM ZnCl₂ and/or 30 mM MgCl₂. During growth, β-galactosidase activities were measured. (B) MG1301cells carrying the mgrB::lacZ reporter gene were grown at 20°C in EB medium containing 0.4% glucose and 1 mM ZnCl₂ with different concentrations of MgCl₂. Then β-galactosidase activities were measured after 8 and 12 h of incubation. Under both experimental conditions consistent results were obtained, and representative data (12 h) are shown.

Schematic representation of microarray data obtained with a mutant with a mutation in the PhoQ/P signaling system. Microarray analyses were carried out as indicted. The transcriptome of ΔphoQ/P cells grown at 20°C in EB medium containing 0.4% glucose and 1 mM ZnCl₂ (labeled with Cy5) was compared with that of ST001 (wild-type) cells grown under the same conditions (labeled with Cy3). The data analyses revealed that 100 genes showed mean values for Cy5/Cy3 that were less than 0.5. These genes were assumed to be the genes whose expression was down-regulated in the ΔphoQ/P background (designated ΔphoQ/P-down genes). (A) Uncovered genes were compared with zinc-up genes (see Fig. 4). This analysis allowed us to select 61 ΔphoQ/P-down genes, whose zinc-responsive expression was assumed to be mediated by the PhoQ/P signaling system. (B) Genes compared with genes belonging to the Rcs regulon (see Fig. 4). Note that most of the Rcs regulon members were also found to be ΔphoQ/P-down genes. The microarray data used in this study are available as supplemental data at the following web site: http://www.agr.nagoya-u.ac.jp/%7emicrobio/.

Proposed organization of the PhoQ/P-Rcs signaling network. The magnesium-regulated PhoQ/P signaling system appears to function upstream of the Rcs signaling system. The results of this study suggested that the PhoQ/P signaling system is activated by an external zinc signal, resulting in activation of the Rcs signaling system. The function of RcsF is not clear in this model. However, RcsF is essential for the regulatory system (see Fig. 3), and the rcsF mutation is epstatic to the phoQ mutation (data not shown). We propose that such a sophisticated PhoQ/P-Rcs signaling network ensures that E. coli cells properly regulate the cell surface structures by modifying lipopolysaccharide (LPS) and/or capsular polysaccharide (CPS) in response to the external states of crucial divalent cations (magnesium and zinc). Details are discussed in the text.