Role of the Irr protein in the regulation of iron metabolism in Rhodobacter sphaeroides

Abstract

In Rhizobia the Irr protein is an important regulator for iron-dependent gene expression. We studied the role of the Irr homolog RSP_3179 in the photosynthetic alpha-proteobacterium Rhodobacter sphaeroides. While Irr had little effect on growth under iron-limiting or non-limiting conditions its deletion resulted in increased resistance to hydrogen peroxide and singlet oxygen. This correlates with an elevated expression of katE for catalase in the Irr mutant compared to the wild type under non-stress conditions. Transcriptome studies revealed that Irr affects the expression of genes for iron metabolism, but also has some influence on genes involved in stress response, citric acid cycle, oxidative phosphorylation, transport, and photosynthesis. Most genes showed higher expression levels in the wild type than in the mutant under normal growth conditions indicating an activator function of Irr. Irr was however not required to activate genes of the iron metabolism in response to iron limitation, which showed even stronger induction in the absence of Irr. This was also true for genes mbfA and ccpA, which were verified as direct targets for Irr. Our results suggest that in R. sphaeroides Irr diminishes the strong induction of genes for iron metabolism under iron starvation.

Figure 1. Growth curves of the R. sphaeroides wild type (black) and the 2.4.1Δirr mutant (gray) under normal iron (continuous line) and under iron limitation (dashed line) conditions are shown.

The optical density at 660 nm (OD₆₆₀) of microaerobically grown R. sphaeroides cultures was determined over time. The data represent the mean of at least three independent experiments and error bars indicate standard error of the mean.

Figure 2. Sensitivity of R. sphaeroides wild type, 2.4.1Δirr mutant and complemented mutant to (photo-) oxidative stress.

Inhibition of growth of the wild type (white bars), the irr deletion mutant (black bars) and the complemented mutant (gray bars) to hydrogen peroxide (A) and methylene blue (B) as determined by inhibition zone assays. Each bar represents the mean of at least three independent experiments and error bars indicate standard deviation. Levels of significance are indicated as follows: *P≤0.01.

Figure 3. Relative expression of katE (RSP_2779) in R. sphaeroides wild type and 2.4.1Δirr mutant.

(A) Real-time RT-PCR was used to investigate the relative expression of katE in 2.4.1Δirr mutant 0 min (light gray bar) and 20 min (dark gray bar) after exposure to 1 mM H₂O₂ compared to the wild type. (B) Relative katE expression 20 min of 1 mM H₂O₂ in the R. sphaeroides wild type (white bar) and the irr deletion mutant (black bar). Values were normalized to rpoZ and to the control at time point 0. The data represent the mean of three independent experiments and error bars indicate standard deviation. Levels of significance are indicated as follows: *P≤0.01; **P≤0.05.

Figure 4. Validation of microarray data by real-time RT-PCR.

Relative gene expression (A) in 2.4.1Δirr under normal iron conditions compared to the wild type under normal iron conditions and (B) in 2.4.1Δirr under iron limitation compared to normal iron conditions (light gray bars) and in wild type under iron limitation compared to normal iron conditions (dark gray bars). Values were normalized to rpoZ and to the respective control treatment. The data represent the mean of at least three independent experiments and error bars indicate standard deviation. Numbers in parentheses show the fold change of the respective genes as determined by microarray analysis.

Figure 5. Binding of purified Irr to the promoter of mbfA and ccpA as determined by Electrophoretic Mobility Shift Assays.

(A) Binding of Irr to the promoter region of mbfA (180 bp). All reactions contain the same amount of ³²P end-labeled DNA fragment (3.08 fmol/lane) comprising the promoter sequence. Lanes 1–4 contain no Irr; lanes 3 and 4 contain 0.6 µg BSA; lanes 5 and 7 contain 0.1 µg Irr; lane 6 and 11–13 contain 0.6 µg Irr; lane 8 contains 0.2 µg Irr; lanes 9 and 14–16 contain 0.3 µg Irr; lane 10 contains 0.4 µg. Reactions contain 1 mM MnCl₂ as indicated. Lanes 14–16 contain non-labeled DNA fragment mbfA in excess amount as cold competitor. Lanes 12 and 13 contain radioactively labeled sitA DNA fragment (180 bp) as unspecific DNA. (B) Binding of Irr to the promoter region of ccpA (168 bp). All reactions contain the same amount of ³²P end-labeled DNA fragment (3.68 fmol/lane) comprising the promoter sequence. Lanes 1–3 contain no Irr; lane 3 contains 0.6 µg BSA; lane 4 contains 0.1 µg Irr; lane 5 contains 0.2 µg Irr; lane 6 contains 0.4 µg Irr; lane 7 contains 0.6 µg Irr. Reactions contain 1 mM MnCl₂ as indicated. All reactions contain 1 µg of salmon sperm DNA as unspecific competitor. The asterisks and arrows show the location of free and Irr-bound ³²P end-labeled DNA fragments, respectively.

Figure 6. Determination of 5′ ends of mbfA (RSP_0850) (A) and ccpA (RSP_2395) (B) mRNA by 5′ rapid amplification of cDNA ends (RACE).

Separation of 5′-RACE products mbfA and ccpA obtained from RNA extracts of the wild type strain under normal iron conditions. PCR products obtained after second PCR (nested) were separated on a 10% polyacrylamid gel and stained with ethidium bromide. Determined 5′ ends are indicated by an arrow. The putative translational start is indicated by an asterisk. The Irr-box (ICE, iron control element) is marked by a frame.