Characterization of the myo-inositol utilization island of Salmonella enterica serovar Typhimurium

Abstract

Knockout mutation of STM4432 resulted in a growth-deficient phenotype of Salmonella enterica serovar Typhimurium in the presence of myo-inositol (MI) as the sole carbon source. STM4432 is part of a 22.6-kb genomic island which spans STM4417 to STM4436 (genomic island 4417/4436) and is responsible for MI degradation. Genome comparison revealed the presence of this island in only six Salmonella strains and a high variability of the iol gene organization in gram-negative bacteria. Upon nonpolar deletion of 11 island loci, the genes involved in six enzymatic steps of the MI pathway were identified. The generation time of S. enterica serovar Typhimurium in minimal medium with MI decreases with higher concentrations of this polyol. Reverse transcriptase PCR showed five separate transcriptional units encompassing the genes iolA-iolB, iolE-iolG1, iolC1-iolC2, iolD1-iolD2-iolG2, and iolI2-iolH. Luciferase reporter assays revealed a strong induction of their promoters in the presence of MI but not glucose. The main regulator, IolR, was identified due to a reduced lag phase of a strain mutated in STM4417 (iolR). Deletion of iolR resulted in stimulation of the iol operons, indicating its negative effect on the iol genes of S. enterica serovar Typhimurium in rich medium at a transcriptional level. Bandshift assays demonstrated the binding of this putative repressor to promoter sequences of iolA, iolC1, and iolD1. Binding of IolR to its own promoter and induced iolR expression in an IolR-negative background demonstrate that its transcription is autoregulated. This is the first characterization of MI degradation in a gram-negative bacterium, revealing a complex transcriptional organization and regulation of the S. enterica serovar Typhimurium iol genes.

FIG. 1.

(A) Examples of iol divergons. GEI 4417/4436 (22.6 kb) of S. enterica serovar Typhimurium is presented in comparison to the structural organization of iol genes from B. subtilis and several gram-negative bacteria. Salmonella genes experimentally demonstrated in this study to belong to the inositol divergon are depicted in black; their homologues in other pathogens are shown in gray. Genes encoding putative permeases are hatched. 1, in this organism, iolG is transcribed in the same orientation as the other genes; 2, the iol cluster of P. luminescens is similar to that of E. carotovora but lacks two of three putative permease genes. (B) Reconstruction of the pathway for MI degradation in S. enterica serovar Typhimurium. Seven stepwise reactions are involved in MI degradation to glyceraldehyde-3-phosphate and acetyl coenzyme A (acteyl-CoA). None of the genes from GEI 4417/4436 encodes a homologue of a biphosphate aldolase. Chemical structures were taken from the Kyoto encyclopedia of genes and genomes (13).

FIG. 2.

Growth curves of the S. enterica serovar Typhimurium wild-type strain 14028 in MM without or with increasing concentrations of MI. The iolD2 and iolE deletion mutants were cultivated in the presence of 55.5 mM MI. Zero growth of the wild-type strain and the two mutants in the absence of this carbon source was monitored for at least 100 h after inoculation. Standard deviations from at least three independent experiments are shown. The molarity of MM with respect to MI is indicated. WT, wild type.

FIG. 3.

Transcriptional organization of S. enterica serovar Typhimurium MI utilization genes. Strain 14028s was grown in MM with 55.5 mM MI at 37°C, and mRNA was extracted at an OD₆₀₀ of 0.4. cDNA was amplified with reverse primers listed in Table S1 in the supplemental material. RT-PCR was performed with primer pairs specific for the indicated regions 1 to 17. All PCR products were separated by 2% agarose gel electrophoresis. As controls, PCR amplification products with genomic DNA and DNase-treated RNA samples as template are shown. Line numbers correspond to PCR product numbers depicted above the gel lanes. Arrows indicate promoters identified in this study.

FIG. 4.

Growth curve of strain 14028s ΔiolR. Exponential growth of the strain lacking the repressor IolR starts approximately 10 h earlier than that of the wild-type strain. The phenotype of 14028s ΔiolR could partially be complemented upon in trans expression of iolR via pBR322. Average values from three independent experiments are shown. Standard deviations are not given due to a variable lag phase, and all growth curves were normalized to a lag phase ending 60 h after inoculation.

FIG. 5.

Promoter binding activity of IolR. The interaction of IolR with the regulatory region of nine genes of GEI 4417/4436 is shown. One hundred nanograms of DNA was used in each experiment. The promoter fragments were incubated without or with increasing amounts (7 ng, 14 ng, 28 ng, and 49 ng [221 fM, 442 fM, 884 fM, and 1547 fM, respectively]) of the purified IolR protein. No bandshift was observed when a maximum of 210 to 280 ng IolR, corresponding to a maximal 17-fold molar excess, was incubated with promoter DNAs of iolE, iolG2, iolI, and iolH (data not shown). Protein-DNA complexes were separated on a 9.5% native polyacrylamide gel. A 200-bp sequence of the argS promoter served as a negative control.

FIG. 6.

Regulation of MI utilization in S. enterica serovar Typhimurium. The wild-type strain and an iolR deletion mutant carrying recombinant pDEW201-constructs were grown in LB or in MM with MI or glucose. Promoter induction is depicted by arrows of different sizes (from smallest to largest): <10⁴ RLU/OD₆₀₀ unit, <10⁵ RLU/OD₆₀₀ unit, <10⁶ RLU/OD₆₀₀ unit, <10⁷ RLU/OD₆₀₀ unit, and <10⁸ RLU/OD₆₀₀ unit. Induction of iol genes was similar in both strains in the presence of MI. Binding sites of IolR are indicated by open circles. WT, wild type.