Control of methionine synthesis and uptake by MetR and homocysteine in Streptococcus mutans

Abstract

MetR (formerly Smu.1225), a regulator of the LysR family, controls key genes for methionine supply in Streptococcus mutans. An S. mutans metR mutant is unable to transport l-methionine and to grow in the absence of this amino acid. Accordingly, MetR activates transcription by binding to the promoter regions of two gene clusters and smu.1487, whose products are involved in methionine biosynthesis (MetEF and Smu.1487) and uptake (AtmBDE). Transcriptional activation by MetR requires the presence of a 17-bp palindromic sequence, the Met box. Base substitutions in the Met box hinder the formation of a MetR-DNA complex and abolish MetR-dependent activation, showing that Met boxes correspond to MetR recognition sites. Activation by MetR occurs in methionine-depleted medium and is rapidly triggered under nonactivating conditions by the addition of homocysteine. This intermediate of methionine biosynthesis increases the affinity of MetR for DNA in vitro and appears to be the MetR coeffector in vivo. Homocysteine plays a crucial role in methionine metabolic gene regulation by controlling MetR activity. A similar mechanism of homocysteine- and MetR-dependent control of methionine biosynthetic genes operates in S. thermophilus. These data suggest a common mechanism for the regulation of the methionine supply in streptococci. However, some streptococcal species are unable to synthesize the homocysteine coeffector. This intriguing feature is discussed in the light of comparative genomics and streptococcal ecology.

FIG. 1.

Sulfur amino acid biosynthesis and uptake pathways in S. mutans. The pathways were deduced from (i) the analysis of S. mutans genome sequence data (http://www.ncbi.nlm.nih.gov/genomes), (ii) knowledge of sulfur amino acid biosynthetic pathways in other bacteria, such as L. lactis (41), and (iii) the experimental results of this work. Genes of S. mutans proposed to encode proteins involved in the corresponding reaction are shown; genes belonging to the MetR and homocysteine regulon are in bold, and genes for transporters potentially involved in the uptake of thiosulfate and methionine are in brackets. S, sulfide; TS, thiosulfate; SAH, S-adenosylhomocysteine; SRH, S-ribosylhomocysteine; dashed-line arrow, not-yet-identified transporters; AtmBDE, ABC transporter belonging to the MUT family and associated with d-methionine, l-methionine (low concentrations), selenomethionine, and homocysteine in S. mutans; BcaP (formerly Smu.16), transporter potentially involved in methionine uptake; CysE, serine O-acetyltransferase; CysK, O-acetylserine sulfhydrylase; MetB, cystathionine γ-synthase; MetC, cystathionine β-lyase; MetA, homoserine O-acyltransferase; CysD, O-acylhomoserine sulfhydrylase; Smu.1487, putative homocysteine S-methyltransferase; MetE, methionine synthase; MetF, putative methylenetetrahydrofolate reductase; MetK, methionine adenosyltransferase; Pfs, S-adenosylhomocysteine nucleosidase; LuxS, S-ribosylhomocysteinase.

FIG. 2.

Methionine uptake in various S. mutans strains. Time course for l-[¹⁴C]methionine uptake into S. mutans strains UA159 (wild type; ⧫), JIM8862 (ΔatmB; ▴), JIM8863 (ΔatmBCDE; ▪), and JIM8860 (ΔmetR; ×).

FIG. 3.

Genetic organization and sequence alignment of promoter regions with potential Met boxes in S. mutans. In the upper panel, dark gray arrows correspond to genes with Met boxes in their promoter regions and light gray arrows correspond to methionine metabolism genes. Experimentally mapped promoters are indicated by black arrows. No transcription start site located less than 4 kb upstream of smu.1936c was detected, suggesting the coexpression of smu.1936c and the atm cluster. Lollipop symbols indicate predicted transcription terminators. In the lower panel, the ATG start codon of each gene is indicated in italics. The experimentally determined sites of transcription initiation are given in bold letters. The deduced extended −10, −10, and −35 boxes for each promoter are underlined. Met boxes are shaded. The Met boxes of the atmB, metE, cysD, metA, and smu.1487 genes are positioned as expected for an LTTR DNA-binding sequence.

FIG. 4.

Gel mobility shift analyses of MetR binding to different promoter regions. Labeled DNA probes (0.1 nM) corresponding to atmB, cysD, cysK, metA, metE, metE* (a metE fragment with substitutions in the Met boxes), metR, and smu.1487 were incubated with different amounts of MetR-His₈ (0 to 650 nM) and analyzed by nondenaturing polyacrylamide gel electrophoresis (see Materials and Methods for details). 260*, a 0.1 nM labeled smu.1487 fragment was incubated with 260 nM MetR-His₈ in the presence of a 50-fold excess of an unlabeled smu.1487 fragment.

FIG. 5.

Effects of substitutions in the Met boxes of PmetE-lux transcriptional fusions. The left panel is a schematic representation of transcriptional fusions between the wild-type or modified metE promoter region and luciferase reporter genes. Striped boxes and arrows represent the Met boxes and the metE promoters, respectively. Consistent with its minor role in the expression of metE, P1 is represented by dashed-line arrows. The right panel shows luciferase activity levels in cells expressing the corresponding fusion and cultivated in CDM+CM (+ Met) or CDM+C (− Met). The values reported are averages of those obtained for at least three independent cultures at an OD₆₀₀ of 0.4. Ratios were calculated by dividing the values obtained in the absence of methionine by the values obtained in the presence of methionine.

FIG. 6.

Effects of Met box or MetR inactivation on methionine (A)- and homocysteine (B)-dependent regulation of metE transcription. Levels of luciferase activity (amount of Lux per OD₆₀₀ unit) from the UA159 wild-type (JIM8867 and JIM8870) and ΔmetR mutant (JIM8873) strains expressing the wild-type PmetE-lux fusion and the Met box mutant PmetE-lux fusion (ΔMet-boxes) were measured. (A) Cells were grown in M17 medium to early exponential phase (OD₆₀₀ of 0.2). Bacterial cultures were split in two, washed, and resuspended in CDM+C (⋄) or CDM+CM (⧫) (time zero). Cells were held in these media for 90 min, and luciferase activity was measured. (B) Cells were grown in CDM+CM to early exponential phase (OD₆₀₀ of 0.2). Bacterial cultures were split in two, and 4 mM homocysteine was added to one of the aliquots. The level of luciferase activity from cells cultivated in the presence (⧫) or in the absence (⋄) of 4 mM homocysteine was measured. The x axis shows the period of incubation after the addition of homocysteine (time zero). One curve representative of the results from at least three experiments is presented.

FIG. 7.

Effect of homocysteine on MetR-DNA complex formation. This effect was measured by gel mobility shift assays. The promoter region of smu.1487 was generated by PCR and radiolabeled. Labeled DNA probes (0.1 nM) were incubated with the indicated amounts of MetR-His₈ (0 to 100 nM) in the presence (+) or absence (−) of 4 mM homocysteine. The samples were analyzed by nondenaturing polyacrylamide gel electrophoresis (see Materials and Methods for details).