Ribonucleotide reductases of Salmonella typhimurium: transcriptional regulation and differential role in pathogenesis

Abstract

Ribonucleotide reductases (RNRs) are essential enzymes that carry out the de novo synthesis of deoxyribonucleotides by reducing ribonucleotides. There are three different classes of RNRs (I, II and III), all having different oxygen dependency and biochemical characteristics. Salmonella enterica serovar Typhimurium (S. Typhimurium) harbors class Ia, class Ib and class III RNRs in its genome. We have studied the transcriptional regulation of these three RNR classes in S. Typhimurium as well as their differential function during infection of macrophage and epithelial cells. Deletion of both NrdR and Fur, two main transcriptional regulators, indicates that Fur specifically represses the class Ib enzyme and that NrdR acts as a global repressor of all three classes. A Fur recognition sequence within the nrdHIEF promoter has also been described and confirmed by electrophoretic mobility shift assays (EMSA). In order to elucidate the role of each RNR class during infection, S. Typhimurium single and double RNR mutants (as well as Fur and NrdR mutants) were used in infection assays with macrophage and epithelial cell lines. Our results indicate class Ia to be mainly responsible for deoxyribonucleotide production during invasion and proliferation inside macrophages and epithelial cells. Neither class Ib nor class III seem to be essential for growth under these conditions. However, class Ib is able to maintain certain growth in an nrdAB mutant during the first hours of macrophage infection. Our results suggest that, during the early stages of macrophage infection, class Ib may contribute to deoxyribonucleotide synthesis by means of both an NrdR and a Fur-dependent derepression of nrdHIEF due to hydrogen peroxide production and DNA damage associated with the oxidative burst, thus helping to overcome the host defenses.

Figure 1. NrdR-dependent expression of RNR.

NrdR regulation of S. Typhimurium LT2 RNR expression throughout the growth curve (early exponential (OD₅₅₀ = 0.2), final exponential (OD₅₅₀ = 0.8), early stationary phase (OD₅₅₀ = 1.4)). β-galactosidase activity of nrdAB, nrdHIEF, and nrdDG promoter regions fused to lacZ expressed in Miller Units (MU). Wild-type strain shown in black, nrdR mutant strain in dark grey, and nrdR mutant complemented strain in light grey. Results are the mean values of duplicates of at least three independent experiments. Error bars represent the standard deviation of experiments.

Figure 2. Fur-dependent expression of nrdHIEF.

Effect of Δfur mutation on nrdHIEF expression. β-galactosidase activity is expressed in Miller Units (MU) for the wild-type strain (Wt), mutant Δfur strain (Fur^-), and complemented mutant strain (cFur). Results are the mean values of duplicates of at least three independent experiments. Error bars represent the standard deviation of experiments.

Figure 3. Hydroxyurea-dependent expression of RNR.

Effect of hydroxyurea on nrdAB, nrdHIEF and nrdDG expression of wild-type and ΔnrdR mutant strains. β-galactosidase activity is expressed in Miller Units (MU). Overnight cultures of each strain were reseeded and grown for 1 hour previous to inoculation with 10 mM hydroxyurea. nrdDG fusion strains were grown either in aerobiosis (in black) or anaerobiosis (in grey). Results are the mean values of duplicates of at least three independent experiments. Error bars represent the standard deviation of experiments.

Figure 4. NrdR binding sites.

A. S. Typhimurium LT2 nrdAB, nrdHIEF and nrdDG promoter regions. Black boxes indicate NrdR recognition sites and mutated sequences are showed below in bold lower case. −10 and −35 boxes are showed in bold. Transcriptional start site is indicated in bold and underlined as is the 19 bp Fur box in the nrdHIEF promoter region. The DnaA and FNR boxes are underlined for the nrdAB and nrdDG promoter regions, respectively. B. β-galactosidase activities of nrdAB, nrdHIEF, and nrdDG transcriptional fusions with mutagenized NrdR boxes are expressed in Miller Units (MU) for the wild-type strain (Wt), ΔnrdR strain (R⁻), NrdR1 box mutant strain (R1⁻), NrdR2 box mutant strain (R2⁻), and NrdR1 and NrdR2 box double mutant strain (R1⁻/R2⁻). Results are the mean values of duplicates of at least three independent experiments. Error bars represent the standard deviation of experiments.

Figure 5. Fur binding sites.

A. Fur box mutated sequence for the nrdHIEF promoter region. Multiple alignment with ClustalW of the E. coli Fur box consensus sequence, S. Typhimurium LT2 Fur box within the fur promoter region and S. Typhimurium LT2 Fur box in the nrdHIEF promoter region. All sequences were obtained from the NCBI database (http://www.ncbi.nml.nih.gov/). Conserved bases in all three sequences are shown in black, while bases conserved in two out of the three sequences are shown in grey. Base changes in the Fur box sequence are shown in bold lower case. B. β-galactosidase activities of the nrdHIEF-lacZ fusion expressed in Miller Units (MU) for the wild-type strain (Wt), mutant Δfur strain (Fur⁻), fur-complemented mutant strain (cFur), Fur box mutant strain (Furbox⁻), ΔnrdR mutant strain (R⁻) and ΔfurΔnrdR double mutant strain (Fur⁻/R⁻). Results are the mean values of duplicates of at least three independent experiments. Error bars represent the standard deviation of experiments.

Figure 6. Fur binding to nrdHIEF.

EMSA of (A) nrdHIEF and (B) fur promoter region probes of S. Typhimurium LT2, with increasing concentrations of Fur protein: 1–7 lanes: 0, 0.2, 0.4, 0.8, 1, 1.2, and 1.5 µg. MWM: molecular weight marker.

Figure 7. Role of RNR during macrophage infection.

A. Proliferation indexes (PI) at 24 h p. i. of wild-type, ΔnrdAB (IG136), ΔnrdEF (IG138), ΔnrdDG (IG139), and ΔnrdEFΔnrdDG (IG140) mutant strains grown in RAW24.7 (black) or RAW264.7 Nramp1^+/+ (grey) mouse macrophage-like cell lines. B. Proliferation indexes (PI) at 24 h p.i. of wild-type and ΔnrdAB (IG136) strains infecting RAW264.7 (black), RAW264.7 + 65 µM 2,2′-dipyridil (DIP) (light grey) and RAW264.7 Nramp1^+/+ (dark grey). C. Proliferation indexes (PI) at 4, 6 and 24 h p.i. shown in black, light grey, and dark grey, respectively, of wild-type and IG137 strains infecting RAW264.7 and RAW264.7 Nramp1^+/+ mouse macrophage-like cell lines. Results are the mean values of at least three independent experiments. Error bars represent the standard deviation of the mean. P values were determined by Student's t test for proliferation indexes. ** indicates significant differences, P<0.05.

Figure 8. Role of nrdHIEF during macrophage infection.

Proliferation indexes (PI) at 4, 6 and 24 h p.i. for wild-type, ΔnrdABΔnrdR (IG143) and ΔnrdABΔfur (IG144) double mutant strains and ΔnrdABΔnrdDGΔnrdR (IG145) and ΔnrdABΔnrdDGΔfur (IG146) triple mutant strains infecting (A) RAW264.7 and (B) RAW264.7 Nramp1^+/+. Results are the mean values of at least three independent experiments. Error bars represent the standard deviation of the mean. Significant differences were determined by Student's t test for proliferation indexes with P<0.05.