S-Methylmethionine Metabolism in Escherichia coli

Physiological role of the homocysteine methyltransferase.To elucidate the role of the YagD protein in methionine metabolism, a mutant was constructed with an in-frame deletion in both themetE and metH genes. For this purpose, chromosomal copies of the two genes were amplified by PCR. After deletion of residues 493 to 1665 of metE and residues 1198 to 1554 of metH, the mutant genes were re-introduced into wild-type strain KL19 (20) by homologous recombination according to the method of Hamilton et al. (13). Stationary cells (see Fig. 1) of the resulting strain, MTD23, were then used to inoculate minimal medium containing 15 μM (each) different supplements to an optical density at 600 nm (OD₆₀₀) of 0.05. Throughout this work M9 minimal medium (30) containing 0.8% glucose and the supplement indicated was used.L-S-methylmethionine was from Acros Organics (Geel, Belgium). All other compounds were purchased from Sigma (Deisenhofen, Germany). After 18 h of aerobic incubation at 37°C, the cell densities of the cultures were measured. The OD₆₀₀ values (± standard deviations) obtained were 0.056 (± 0.003) without any supplement, 0.280 (± 0.002) with dl-methionine sulfoxide, 0.380 (± 0.006) with l-methionine, 0.074 (± 0.002) withS-adenosyl-l-methionine, 0.771 (± 0.001) withdl-S-methylmethionine, and 0.679 (± 0.005) withl-S-methylmethionine. These results demonstrate that l-methionine, dl-methionine sulfoxide, andS-methylmethionine complement auxotrophy of the strain. YagD does not contribute to the utilization of methionine sulfoxide, because no transfer of the methyl group of this compound to homocysteine takes place (data not shown). Presumably, methionine sulfoxide is converted to methionine via reduction (7). The lower OD value obtained for this substrate may be due to its slower utilization in comparison to methionine (7, 12). WithS-adenosyl-l-methionine (AdoMet), only marginal growth could be observed, which probably resulted from utilization of impurities contained in the AdoMet preparation. This finding is in accordance with previous observations, namely that E. coliK-12 and its derivatives are largely impermeable to AdoMet (14). Growth yield with the l form ofS-methylmethionine and the racemate was about twice that obtained with l-methionine.

The YagD homocysteine methyltransferase does not utilize thed stereoisomer of S-methylmethionine as a substrate (25). It was, therefore, surprising that thedl racemate yielded the same cell mass as the lform. This result was supported in an experiment in which a series of different concentrations ofdl-S-methylmethionine andl-methionine were supplemented (Fig.1). The data indicate thatd-methylmethionine is converted to the l form prior to its utilization, possibly via the same pathway as the conversion of d-methionine to l-methionine (5).

• Open in new tab
• Download powerpoint

Fig. 1.

Effects of ΔyagD and ΔykfDmutations on the utilization of S-methylmethionine. Cells of strains MTD23 (ΔmetE ΔmetH), MTD123 (ΔyagD ΔmetE ΔmetH), and MTD234 (ΔmetE ΔmetH ΔykfD) were grown in minimal medium containing 100 μMl-methionine. After 18 h of aerobic incubation at 37°C, they were washed, resuspended in 0.9% NaCl to an OD₆₀₀ of 10, and transferred into minimal medium containing different concentrations of l-methionine (met) ordl-S-methylmethionine (methyl-met), respectively. The initial OD₆₀₀ was 0.05. After 18 h of aerobic incubation at 37°C, the resulting cell densities were determined. The data shown represent the mean values of at least two independent growth experiments.

To assess the involvement of the yagD gene product inS-methylmethionine metabolism, an in-frame deletion (residues 270 to 548) was introduced into the corresponding gene of strain MTD23 according to the method of Hamilton et al. (13), yielding the triple mutant MTD123 (ΔyagD ΔmetE ΔmetH). An immunoblotting analysis using antiserum directed against homocysteine methyltransferase confirmed that there was no cross-reacting material (data not shown). The results of the growth experiment shown in Fig. 1 prove that metabolism ofS-methylmethionine requires a functional YagD protein, since inactivation of the yagD gene abolishes the capacity to grow on S-methylmethionine.

ykfD, a gene coding for a putativeS-methylmethionine permease.The results described above point to a role for the YagD homocysteine methyltransferase in the utilization of external S-methylmethionine as a source for methionine. Since there is no evidence in the literature thatE. coli is able to synthesize S-methylmethionine, there might be a transport system for its uptake from the medium. The results presented previously do not exclude thatS-methylmethionine uses the methionine uptake system (16, 17).

An inspection of the genomic sequence of E. coli in the vicinity of yagD revealed that upstream of yagDthere is an open reading frame of 1,533 bp (ykfD) coding for a putative protein and overlapping with the yagD reading frame by 14 bp (4) (Fig. 2). Its amino acid sequence shares significant sequence similarity with that of amino acid permeases (data not shown). This similarity and the overlapping reading frames might be an indication that ykfDcodes for a S-methylmethionine transporter which is cotranscribed with the yagD gene.

• Open in new tab
• Download powerpoint

Fig. 2.

Chromosomal organization of the ykfD andyagD open reading frames. The restriction sites used in this study are shown. Putative MET boxes in the 5′ region of ykfDare depicted as boxed nucleotide sequences. The degrees of identity with the MET box consensus sequence 5′-AGACGTCT-3′ are indicated. Possible translational start codons are marked by boldface letters.

To test this assumption we introduced an in-frame deletion intoykfD (residues 304 to 1089) and combined this mutation with the metE metH lesions of strain MTD23 according to the method of Hamilton et al. (13). The resulting triple mutant, MTD234 (ΔmetE ΔmetH ΔykfD), was tested for growth on different supplements. Figure 1 demonstrates that the ykfDlesion prevents growth on S-methylmethionine but not on methionine. The control strain MTD23, in contrast, is able to utilizeS-methylmethionine. Immunoblotting analysis of cell lysates from strain MTD234 grown in minimal medium supplemented with 20 μMl-methionine confirmed that the yagD gene was expressed (data not shown). Thus, the mutation in ykfD did not abolish YagD formation by any polarity effect. AtS-methylmethionine concentrations of higher than 50 μM slow but significant growth of strain MTD234 could be observed, which may indicate nonspecific transport of the compound via other systems (data not shown).

Regulation.The phenotypes of the ykfD andyagD mutants described above indicate a physiological role for the gene products in the acquisition of externalS-methylmethionine and in its conversion into methionine. This could imply that the two genes are subject to regulation by the MetJ-S-adenosyl-methionine system. MetJ is a homodimer of 12-kDa promoters (35, 36). It binds to the MET box (5′-AGACGTCT-3′) in the upstream region of genes (metA, metBL, metC, metF,metJ, metR, and metE) whose products are involved in methionine biosynthesis and acts as a transcriptional repressor (3, 18, 28, 29). Between two and five of these MET boxes are organized in tandem repeats. Because of cooperative interactions between the repressor molecules bound, their number and match with the consensus sequence determines the level of repressibility. S-adenosylmethionine, which is synthesized from methionine and adenosine triphosphate by the metK gene product (21), acts as a corepressor (27, 33, 34) and thereby mediates methionine regulation of the metregulon.

Upstream of yagD, no sequence similar to that of the MET box motif can be found. In contrast, there are four of these motifs upstream of ykfD (Fig. 2). Depending on which of the two ATG codons of the ykfD gene is functioning for the start of the translation, the four MET boxes overlap or immediately precede the start of the reading frame.

To analyze whether the ykfD yagD putative transcriptional unit is indeed under control of methionine, cultures of E. coli KL19 (wild type) were grown in minimal medium supplemented with different concentrations of l-methionine ordl-S-methylmethionine. After 230 min of growth, cells were harvested and analyzed for the formation of YagD protein by immunoblotting (Fig. 3). It is evident that the amount of YagD protein was strongly reduced when 40 μMl-methionine or higher concentrations were present in the medium. Supplementation of S-methylmethionine only caused a slight reduction in the amount of YagD, which might be due to its intracellular conversion into methionine. Similar results were obtained when synthesis of YagD was followed by measuring enzyme activity (results not shown).

• Open in new tab
• Download powerpoint

Fig. 3.

Influence of methionine andS-methylmethionine supplementation on the expression ofyagD. Minimal medium containing different concentrations ofl-methionine (A) ordl-S-methylmethionine (B) was inoculated to an OD₆₀₀ of 0.05 with stationary cells (see legend to Fig. 1; grown without l-methionine) of wild-type strain E. coli KL19. After 230 min of aerobic incubation at 37°C, the cells were harvested and lysed. The proteins were separated on a 15% polyacrylamide gel in the presence of sodium dodecyl sulfate (19). The detection of YagD was performed by immunoblotting analysis using anti-YagD antiserum (obtained from Eurogentech, Belgium, by custom immunization of a rabbit with purified YagD) in a 1:4,000 dilution, protein A-horseradish peroxidase conjugate (Bio-Rad, Munich, Germany), and chemiluminscence blotting substrate from Boehringer (Mannheim, Germany).

The results described here identify the function of two unassigned open reading frames from E. coli. Their products are involved in the uptake of S-methylmethionine and in the methyl transfer to homocysteine, rendering two molecules of methionine. Such an activity has been described for crude extracts from E. colicells by Balish and Shapiro (2).S-methylmethionine is a compound synthesized and stored by many plant species (15, 23), and expression ofykfD and yagD enables E. coli to utilize this compound. The abilities of other methionine auxotrophic bacterial species to use S-methylmethionine as a source of methionine (22) might be based on similar systems. In agreement with this physiological role, ykfD andyagD are subject to control of expression by methionine, thus constituting two more genes belonging to the methionine regulon. More detailed studies, however, are required to confirm that MetJ is involved as a regulatory protein. On the basis of the results obtained in this study we propose the new gene designation mmu(S-methylmethionine utilization). Accordingly, the gene coding for the putative S-methylmethionine permease (ykfD) is renamed mmuP, and theS-methylmethionine:homocysteine methyltransferase gene (yagD) is renamed mmuM.