Heme rescues a two-component system Leptospira biflexa mutant

Abstract

Background: Heme is typically a major iron source for bacteria, but little is known about how bacteria of the Leptospira genus, composed of both saprophytic and pathogenic species, access heme.

Results: In this study, we analysed a two-component system of the saprophyte Leptospira biflexa. In vitro phosphorylation and site-directed mutagenesis assays showed that Hklep is a histidine kinase which, after autophosphorylation of a conserved histidine, transfers the phosphate to an essential aspartate of the response regulator Rrlep. Hklep/Rrlep two-component system mutants were generated in L. biflexa. The mutants could only grow in medium supplemented with hemin or delta-aminolevulinic acid (ALA). In the pathogen L. interrogans, the hklep and rrlep orthologous genes are located between hemE and hemL genes, which encode proteins involved in heme biosynthesis. The L. biflexa hklep mutant could be complemented with a replicative plasmid harbouring the L. interrogans orthologous gene, suggesting that these two-component systems are functionally similar. By real-time quantitative reverse transcription-PCR, we also observed that this two-component system might influence the expression of heme biosynthetic genes.

Conclusion: These findings demonstrate that the Hklep/Rrlep regulatory system is critical for the in vitro growth of L. biflexa, and suggest that this two-component system is involved in a complex mechanism that regulates the heme biosynthetic pathway.

Figure 1

Genetic organization of the Hklep/Rrlep two-component system in Leptospira spp. (A) Organization of hklep/rrlep genes in L. biflexa. The two-component system genes are shaded in black. Gene labels are given for each surrounding genes (white). The arrow indicates the insertion site of the Himar1 transposon. (B) Schematic representation of the heme biosynthesis genes in L. interrogans. The hemACDBLENYH genes are shaded in grey. The homologous two-component system genes, hklep and rrlep, are shaded in black. The L. biflexa Hklep and Rrlep proteins share 66 % and 69 % similarities with L. interrogans LB014 and LB015, respectively.

Figure 2

Growth of L. biflexa hklep and rrlep mutant strains in EMJH medium. A. The rrlep and hklep mutants grew as well as the wild-type strain (wt) in EMJH liquid medium supplemented with hemin or δ-aminolevulinic acid (ALA). However, the mutants did not grow in EMJH liquid medium. B. Complementation of the L. biflexa rrlep mutant. Transformation of L. biflexa rrlep mutant with the empty shuttle vector and the shuttle vector carrying the L. biflexa rrlep gene plated onto EMJH and EMJH supplemented with hemin. The growth of the rrlep mutant complemented by the rrlep wild type gene was similar to the wild type growth onto EMJH plates. Strains were incubated at 30°C for one week.

Figure 3

Schematic representation of Hklep and Rrlep proteins tested for phosphorylation assays and complementation of L. biflexa mutants. (A) The L. biflexa Hklep protein is composed of three domains. Hklep is putatively anchored into the inner membrane through two transmembrane segments (TM1-TM2) separated by a periplasmic loop (PL) that could correspond to the sensing domain. The residue histidine 98 is predicted to be the site of phosphorylation. The asterisk marks its mutation into an alanine residue in Hklep(H98A). In HklepΔ, the periplasmic loop is replaced by another sequence (see Materials and Methods). In Hklep (80–253), the putative sensing domain is deleted and the original promoter of hklep is replaced by a leptospiral promoter. (B) The L. biflexa Rrlep protein is composed of two domains. The residue aspartate 53 is predicted to be the site of phosphorylation. The asterisk marks its mutation into an alanine residue in Rrlep(D53A). Lbi and Lint refer to the L. biflexa and L. interrogans alleles, respectively. "-" indicates no in vitro phosphorylation or absence of complementation of the L. biflexa mutant strain. "+" indicates in vitro phosphorylation or complementation of the L. biflexa mutant strain. ND: not determined. The expression of HklepH98A, Hklep(80–253), Rrlep(D53A), or Rrlep-Lint (LB015) failed to complement their respective mutant and was characterized by an absence of growth (-) in EMJH medium like observed for hklep and rrlep mutants. The expression of HklepΔ or Hklep-Lint (LB014) complemented hklep mutant restoring a wild type growth (+) in EMJH medium.

Figure 4

Phosphorylation assays with L. biflexa Hklep and Rrlep proteins. (A) For autophosphorylation assays, Hklep was incubated with [γ-³²P]ATP for 1 h 30. The arrow indicates the putative dimer form of Hklep. (B) For phosphotranfer assays, an equal amount of Rrlep protein was added to the phosphorylated Hklep and the reaction was further incubated for 5, 15, 30, and 60 min. Proteins were separated on 12 % SDS-polyacrylamide gel and visualized using autoradiography. The molecular mass is indicated in kDa.

Figure 5

Phosphorylation assays of L. biflexa Hklep and Rrlep mutant proteins. (A) Site-directed mutagenesis of the L. biflexa histidine kinase protein. The Hklep or Hklep(H98A) proteins were incubated with [γ-³²P]ATP for 1 h 30. (B) Site-directed mutagenesis of the L. biflexa response regulator protein. Hklep was incubated with [γ-³²P]ATP for 1 h 30 and an equal amount of Rrlep or Rrlep(D53A) proteins was added. The reaction was further incubated for 1 h. Proteins were separated on 12 % SDS-polyacrylamide gel and visualized using autoradiography. The molecular mass is indicated in kDa.

Figure 6

Relative gene expression of L. biflexa hemA, hemE, hemL, and hemH genes measured by real-time quantitative RT-PCR. The levels of specific mRNA transcript of hemAELH genes were quantified in wild-type (bars in white), rrlep mutant (bars in grey), and hklep mutant (bars in black) grown in EMJH supplemented with 2 (left bar) or 10 μM (right bar) hemin. The amounts of mRNA transcript are shown relative to the quantity of that particular mRNA transcript in EMJH condition (without hemin). As an endogenous control, the 23S rRNA (23S rRNA = 1.0) was used for normalization of transcript levels. Experiments were performed in triplicate from distinct cultures to establish standard deviations.