The Escherichia coli metD Locus Encodes an ABC Transporter Which Includes Abc (MetN), YaeE (MetI), and YaeC (MetQ)

JOURNAL OF BACTERIOLOGY, Oct. 2002, p. 5513–5517 0021-9193/02/$04.00⫹0 DOI: 10.1128/JB.184.19.5513–5517.2002 Copyright © 2002, American Society for Microbiology. All Rights Reserved.

Vol. 184, No. 19

The Escherichia coli metD Locus Encodes an ABC Transporter Which Includes Abc (MetN), YaeE (MetI), and YaeC (MetQ) Christophe Merlin, Gregory Gardiner, Sylvain Durand, and Millicent Masters* University of Edinburgh, Institute for Cell and Molecular Biology, Edinburgh EH9 3JR, Scotland Received 4 March 2002/Accepted 2 July 2002

and abc (Fig. 2A). According to the SwissProt web site (http: //www.expasy.ch/sprot/), YaeB displays no defining features. YaeD has been recently characterized as GmhB, a D,D-heptose 1,7-bisphosphate phosphatase (15). YaeC and YaeF are annotated as hypothetical lipoproteins of the PS00013 Prosite family (http://ca.expasy.org/prosite/) with a recognizable lipobox in positions 19 to 23 and 17 to 21, respectively. YaeC is included in the TIGRFAM lipoprotein family “TIGR00363.” At http: //www.tigr.org/TIGRFAMs/index.shtml, sequence alignments with other lipoproteins are available. Abc and YaeE display features of an ABC transporter ATPase and permease, respectively. The Abc protein is a member of the PS00211 Prosite family of ABC ATPases, and its sequence possesses the canonical ABC ATPase motifs: a Walker A motif, a linker peptide (or ABC signature), and a Walker B motif at positions 38 to 46, 141 to 149, and 161 to 167, respectively. YaeE, a member of the PS00402 Prosite family of ABC permeases, contains five potential transmembrane domains at positions 21 to 41, 58 to 78, 81 to 101, 152 to 172, and 186 to 206. It has previously been

Although most open reading frames (ORFs) on the Escherichia coli chromosome have now been identified, the functions of the proteins encoded by nearly half of them are either uncertain or completely unknown. In addition, several hundred approximately mapped loci have not been matched with ORFs (1). As part of a program to discover the functions of E. coli genes, we have been trying to match these loci with their corresponding ORFs. We present our analysis of the metD locus here. MetD is a methionine transport system. Early biochemical and kinetic studies demonstrated that methionine uptake in E. coli involves at least two specific transporters: the high-affinity MetD and low-affinity MetP transport systems (10, 11). Both are regulated by the internal methionine pool size and, for MetD, MetJ-mediated repression has been inferred (12, 14). D-Methionine transport is both ATP dependent and osmotic shock sensitive, the latter suggesting the involvement of periplasmic protein(s) (13). These attributes are characteristic of ABC transporters. Although both MetD and MetP import L-methionine, MetD, but not MetP, can also import D-methionine, which can be converted to L-methionine in the cell. Competition experiments suggest that MetD possesses a distinct substrate-binding site for each stereoisomer (14). A Dmethionine transporter with the same uptake properties has also been described in Salmonella enterica serovar Typhimurium and has also been suggested to be an ABC transporter (6). Both metD loci map to corresponding chromosomal locations; genetic studies of Salmonella metD mutants suggest four complementation groups (6, 8). Figure 1 shows a model for methionine transport based on these observations. Kadner and Watson (11) mapped the metD locus to between 3.6 and 5.6 min. The extensive later use of metD to map other loci has allowed us to refine the position of metD as between proS (4.7 min) and rrnH (4.8 min) (3, 7). We analyzed this region by using the GenBank sequence database (accession no. U00096) and found six genes of unknown function: yaeBCDEF

FIG. 1. Model of methionine transport in E. coli. MetD imports Dand L-methionine, while the genetically uncharacterized transporter MetP imports only L-methionine. MetD is represented as a typical ABC transporter with its three components: A, E, and C represent Abc (ATPase), YaeE (permease), and YaeC (D-methionine-binding protein), respectively. IM, inner membrane; P, periplasm; OM, outer membrane.

* Corresponding author. Mailing address: University of Edinburgh, Institute of Cell and Molecular Biology, Darwin Building, King’s Buildings, Mayfield Rd., Edinburgh EH9 3JR, Scotland. Phone: 44 (0)131 650 5355. Fax: 44 (0)131 650 8650. E-mail: M.Masters @ed.ac.uk. 5513

Downloaded from http://jb.asm.org/ on February 5, 2017 by guest

We report that the genes abc, yaeC, and yaeE comprise metD, an Escherichia coli locus encoding a DLmethionine uptake system. MetD is an ABC transporter with Abc the ATPase, YaeE the permease, and YaeC the likely substrate binding protein. Expression of these genes is regulated by L-methionine and MetJ, a common repressor of the methionine regulon. We propose to rename abc, yaeE, and yaeC as metN, metI, and metQ, respectively.

5514

NOTES

J. BACTERIOL.

suggested, based on bioinformatic analysis, that Abc, YaeE, and YaeC form an ABC transporter of unknown specificity (21; www.biology.ucsd.edu/⬃ipaulsen/transport/ecoli.html). ABC transporters require three activities combined in one or more proteins: a permease, an ATPase, and a substratebinding domain (4, 9). The last of these vary greatly, since they differ in substrate specificity, and, not surprisingly, none of the unknown genes in the proS-rrnH region shares similarity with a known substrate-binding protein. Because both yaeC and yaeF are close to abc and yaeE, we surmised that one of them might be the D-methionine-binding protein and examined their cooccurrence with yaeE and abc. When the YaeC protein sequence is compared to the entire bacterial database by BlastP (http://www.ncbi.nlm.nih.gov:80/BLAST/), all of its orthologs are encoded by genes associated with an abc-like gene and a

yaeE-like gene. Figure 2B shows the genetic organization of the yaeC neighboring sequences among the best BLAST hits. Although yaeC, yaeE, and abc genes are always clustered, the colocalization of yaeB and yaeD with abc and yaeE is limited to the ␥ subdivision of the Enterobacteriaceae, while yaeF does not colocalize with these genes other than in E. coli. The clustering of yaeC, yaeE, and abc suggests a common function; the existence of a MetJ-binding sequence upstream of abc (16) suggests that the common function could be the uptake of Dmethionine. The six candidate genes for the metD locus, abc, yaeC, yaeB, yaeD, yaeE, and yaeF, were independently replaced by a selectable and removable reporter cassette (FLKP2: FRT-lacZ-aphPlac-FRT) (18) in strain EDCM367 (Table 1) to create six replacement strains. The gene deletion-replacement proce-

Downloaded from http://jb.asm.org/ on February 5, 2017 by guest

FIG. 2. Genetic organization of the metD region in E. coli (A) and other sequenced bacteria (B). Functionally uncharacterized genes are white, known genes for which the relative position is conserved are black, and genes without orthologs at the same relative position are hatched. E. coli annotated putative promoters are indicated by small black triangles; the putative MetJ binding site is also shown. In order to facilitate comparison between organisms, the E. coli gene names abc, yaeB, yaeC, yaeD, and yaeE have been used for orthologs in the other species. The annotated gene names YPO1069, YPO1071, YPO1072, YPO1074 in Yersinia pestis are thus yaeB, yaeC, yaeE, yaeD; VC905, VC906, VC907, and VC908 in Vibrio cholerae; plpB, PM1729, PM1728, and PM1727 in Pasteurella multocida; and HI0620, HI0620a, HI0621, and HI0261.1 in Haemophilus influenzae are labeled yaeC, yaeE, abc, and yaeD. AGR_L_761, AGR_L_763, and AGR_L_765 in Agrobacterium tumefaciens; Smc03157, Smc03158, and Smc03159 in Sinorhizobium meliloti; BMEII0338, BMEII0336, and BMEII0337 in Brucella meliloti; and mll4794, mll4791, and mll4792 in Mesorhizobium loti are orthologs of yaeC, yaeE, and abc.

VOL. 184, 2002

NOTES

5515

TABLE 1. Strains used in this study Name

Genotype/characteristics

CAG18491 EDCM367 JAHK9 MG1655

metE3079::Tn10 (Tcr) MG1655 ⌬(Plac-lacZY) ZIP514 metJ97; CGSCa strain 5463 Sequenced ␭⫺ and F⫺ derivative of K12 ⌬(codB-lacI)3 tsx-93 ␭⫺ trpA49(Am) relA1 rpsL150 spoT1; CGSC strain 5948

ZIP514

a

Source or reference

20 This study CGSC 2 CGSC

CGSC, E. coli Genetic Stock Center.

FIG. 3. Growth of EDCM367 metE::Tn10 and its deletion derivatives. The deletants were pregrown at 37°C in MM supplemented with L-methionine (20 ␮g/ml) to an optical density at 600 nm (OD600) of 0.2. Cells were washed twice with MM and diluted into MM containing L-methionine, D-methionine, or no supplement. (A) Cultured with no supplement or with L-or D-methionine at 20 ␮g/ml. (B) Like panel A, but including cultures supplemented with 2 ␮g of D-methionine per ml. The deleted gene is indicated in the individual panels.

sion remained at the same level under all growth conditions (Table 2). In order to test whether this expression is MetJ regulated, the reporter cassettes abc具 典FLKP2, yaeC具 典FLKP2, and yaeE具 典FLKP2 were P1 transduced to the metJ strain JAHK9 and to its parent, ZIP514 (Table 1), and the progeny were assayed for ␤-galactosidase activity. (The metJ mutation appears to adversely affect growth or survival, since overnight cultures require many hours to achieve the parental growth rate [data not shown].) In the MetJ⫹ strains (ZIP514 derivatives), abc, yaeC, and yaeE exhibited a 1.5- to 2.5-fold increase in activity when deprived of L-methionine as in the EDCM367 derivatives. In the metJ strains (JAHK9 derivatives), levels of

Downloaded from http://jb.asm.org/ on February 5, 2017 by guest

dure has been extensively described previously (18). The sequences of the primers used can be obtained from the authors. Each replacement strain was made auxotrophic for methionine by P1 transduction (17) of the metE::Tn10 marker from CAG18491 (Table 1). The FLKP2 cassettes were removed by the Flp recombinase provided on plasmid pCP20, as described previously (5, 18), where required to minimize possible polar effects caused by interruption of metD. Each MetE⫺ deletant was tested for growth in VB (22) minimal medium with 1% glucose (MM) supplemented with either L- or D-methionine. All deletants were able to grow in the presence of L-methionine, since MetP transport remains active. However, when D-methionine is the sole methionine source, only strains with an active MetD transport system are expected to grow. We found that of the deletants, the ⌬abc, ⌬yaeE, and ⌬yaeC strains were unable to grow normally when dependent on D-methionine (Fig. 3). For ⌬abc and ⌬yaeE deletion strains, growth was greatly impaired at 20 ␮g of Dmethionine per ml, although it did not entirely cease. The ⌬yaeC strain exhibited reduced growth at this methionine concentration, but failed to grow at all at 2 ␮g of D-methionine per ml. The reference strain EDCM367 metE::Tn10 grew at the same reduced rate at both 20 and 2 ␮g/ml. Thus, it appears that these three genes behave as part of the D-methionine transport system, while the neighboring genes yaeB, yaeD, and yaeF do not. We propose to rename the metD genes abc, yaeE, and yaeC as metN, metI, and metQ, respectively, with the deduced functions of ATPase, D-methionine permease, and D-methionine-binding protein of the D-methionine ABC transporter. We suggest that metD be retained as the name of the locus. The proS-rrnH region was originally annotated with a single consensus Met box (5⬘-AGACGTCT-3⬘) between abc and yaeD, suggesting possible transcriptional regulation by the repressor MetJ (GenBank accession no. U00096). MetJ binding requires at least two Met boxes; a second box at this site has been proposed in a recent model for MetJ repressor binding site recognition (16). If MetJ represses expression of the metD genes, transcription of the genes should increase upon deprivation of its corepressor, methionine. To test this, we measured the expression of all six genes by using the EDCM367 derivatives in which the ORFs are replaced by the FLKP2 cassette (containing a lacZ reporter gene). For the replacements abc具 典FLKP2, yaeC具 典FLKP2, and yaeE具 典FLKP2, a 1.5- to 2.5fold increase in expression followed removal of L-methionine (i.e., in the presence of D-methionine or in the absence of supplement). For the other three replacements, gene expres-

5516

NOTES

J. BACTERIOL. TABLE 2. Transcriptional activity of abc, yaeB, yaeC, yaeD, yaeE, and yaeFa

Host

EDCM367 (⌬lacZY)

Supplement

yaeC具 典FLKP2

yaeD具 典FLKP2

yaeE具 典FLKP2

yaeF具 典FLKP2

337 796 733

370 408 428

55 90 88

372 375 336

229 566 599

575 542 530

L-Met D-Met

462 1,051 1,088

293 301 223

42 77 110

NDb ND ND

416 895 699

ND ND ND

L-Met D-Met

2,516 2,521 2,590

487 474 486

318 321 317

ND ND ND

2,743 2,644 2,625

ND ND ND

None JAHK9 (ZIP514 metJ)

yaeB具 典FLKP2

L-Met D-Met

None ZIP514 (Parental)

␤-Galactosidase sp act (Miller units) with deletion/reporter: abc具 典FLKP2

None a

expression were five- to sevenfold higher than were the Lmethionine repressed levels in the MetJ⫹ controls (Table 2). Repression by L-methionine is, as expected, no longer seen in the metJ strains. In addition, expression levels in the MetJ⫺ strains were increased two- to fourfold compared to the derepressed levels in the MetJ⫹ controls. Because the metJ mutation is deleterious, we measured the expression of yaeB具 典FLKP2, which is not affected by L-methionine level. Its expression was increased 1.5-fold in the MetJ⫺ strains, accounting for a part, but not all, of the increase in the expression of the metD genes in these strains. MetN is the putative ATPase and MetI is the membranespanning region of the MetD ABC transporter. We would thus expect the third component, MetQ, to contain the substratebinding domain. The D-methionine concentration dependence of the ⌬yaeC mutant is consistent with this. As described above, YaeC is a putative lipoprotein. In gram-positive bacteria, where substrate-binding proteins are anchored outside the single membrane by a lipid tail, lipoprotein substrate-binding proteins are common. Since they have not been previously reported in gram-negative bacteria, further study to determine whether YaeC is a membrane-anchored lipoprotein would be worthwhile. The three metD genes are regulated by the MetJ repressor. Although the presence of a good putative MetJ binding site upstream of abc suggests that its action is direct, further molecular study would be needed to confirm the binding of the repressor in this particular location. Kadner and Winkler have shown (12, 14) that the level of MetD transporter is regulated by the intracellular methionine pool, with the possible involvement of MetJ. Consistent with these observations, we show here that the expression of abc (metN), yaeE (metI), and yaeC (metQ) is increased in the absence of methionine in a MetJdependent fashion. Kadner also suggested that MetD contains two distinct substrate-binding domains—one for each methionine stereoisomer. Although YaeC seems to bind D-methionine, what binds L-methionine in the MetD transport system remains an open question. YaeC itself might contain two distinct binding sites—one for each stereoisomer. Alternatively,

the binding protein may be separately encoded, either in this region (i.e., yaeB) or elsewhere. Further analysis of L-methionine transport will first require the identification of the ORFs that constitute metP. We are attempting to do this now. This work was supported by a Project grant from the British Biological Sciences Research Council (BBSRC) to M.M. and a Vacation scholarship from the Society of General Microbiology to G.G. S.D. was supported by a short-term ERASMUS fellowship. We thank Mary Berlyn for bringing MetD to our attention. REFERENCES 1. Berlyn, M. K. 1998. Linkage map of Escherichia coli K-12, edition 10: the traditional map. Microbiol. Mol. Biol. Rev. 62:814–984. 2. Blattner, F. R., G. Plunkett III, C. A. Bloch, N. T. Perna, V. Burland, M. Riley, J. Collado-Vides, J. D. Glasner, C. K. Rode, G. F. Mayhew, J. Gregor, N. W. Davis, H. A. Kirkpatrick, M. A. Goeden, D. J. Rose, B. Mau, and Y. Shao. 1997. The complete genome sequence of Escherichia coli K-12. Science 277:1453–1474. 3. Bohman, K., and L. A. Isaksson. 1980. A temperature-sensitive mutant in prolinyl-tRNA ligase of Escherichia coli K-12. Mol. Gen. Genet. 177:603– 605. 4. Braibant, M., P. Gilot, and J. Content. 2000. The ATP binding cassette (ABC) transport systems of Mycobacterium tuberculosis. FEMS Microbiol. Rev. 24:449–467. 5. Cherepanov, P. P., and W. Wackernagel. 1995. Gene disruption in Escherichia coli: TcR and KmR cassettes with the option of Flp-catalyzed excision of the antibiotic-resistance determinant. Gene 158:9–14. 6. Cottam, A. N., and P. D. Ayling. 1989. Genetic studies of mutants in a high-affinity methionine transport system in Salmonella typhimurium. Mol. Gen. Genet. 215:358–363. 7. Ellwood, M., and M. Nomura. 1982. Chromosomal locations of the genes for rRNA in Escherichia coli K-12. J. Bacteriol. 149:458–468. 8. Grundy, C. E., and P. D. Ayling. 1992. Fine structure mapping and complementation studies of the metD methionine transport system in Salmonella typhimurium. Genet. Res. 60:1–6. 9. Jones, P. M., and A. M. George. 1999. Subunit interactions in ABC transporters: towards a functional architecture. FEMS Microbiol. Lett. 179:187– 202. 10. Kadner, R. J. 1974. Transport systems for L-methionine in Escherichia coli. J. Bacteriol. 117:232–241. 11. Kadner, R. J., and W. J. Watson. 1974. Methionine transport in Escherichia coli: physiological and genetic evidence for two uptake systems. J. Bacteriol. 119:401–409. 12. Kadner, R. J. 1975. Regulation of methionine transport activity in Escherichia coli. J. Bacteriol. 122:110–119. 13. Kadner, R. J., and H. H. Winkler. 1975. Energy coupling for methionine transport in Escherichia coli. J. Bacteriol. 123:985–991. 14. Kadner, R. J. 1977. Transport and utilization of D-methionine and other methionine sources in Escherichia coli. J. Bacteriol. 129:207–216.

Downloaded from http://jb.asm.org/ on February 5, 2017 by guest

Overnight cultures grown at 37°C on MM supplemented with L-methionine (20 ␮g/ml) were diluted 100-fold into the same medium and grown up to an optical density at 600 nm (OD600) of 0.2. Forty milliliters of each culture was centrifuged, washed twice with 40 ml of MM, resuspended in 20 ml of unsupplemented MM, and then diluted twofold (OD600 of 0.1 to 0.2) into MM plus L-methionine or plus D-methionine or without amino acid supplement. Growth was monitored, and samples were taken for ␤-galactosidase assay after 2 h of growth. Assays were performed as described by Miller (19). ␤-Galactosidase specific activities are expressed in Miller units as described previously. Parental strains EDCM367, ZIP514, and JAHK9 were tested under the same conditions, and no residual ␤-galactosidase activity was found. b ND, not done.

VOL. 184, 2002 15. Kneidinger, B., C. Marolda, M. Graninger, A. Zamyatina, F. McArthur, P. Kosma, M. A. Valvano, and P. Messner. 2002. Biosynthesis pathway of ADPL-glycero-␤-D-manno-heptose in Escherichia coli. J. Bacteriol. 184:363–369. 16. Liu, R., T. W. Blackwell, and D. J. States. 2001. Conformational model for binding site recognition by the E. coli MetJ transcription factor. Bioinformatics 17:622–633. 17. Masters, M. 1977. The frequency of P1 transduction of the genes of Escherichia coli as a function of chromosomal position: preferential transduction of the origin of replication. Mol. Gen. Genet. 155:197–202. 18. Merlin, C., S. McAteer, and M. Masters. 2002. Tools for characterization of Escherichia coli genes of unknown function. J. Bacteriol. 184:4573–4581. 19. Miller, J. M. 1992. A short course in bacterial genetics: A laboratory manual

NOTES

5517

and handbook for Escherichia coli and related bacteria. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y. 20. Nichols, B. P., O. Shafiq, and V. Meiners. 1998. Sequence analysis of Tn10 insertion sites in a collection of Escherichia coli strains used for genetic mapping and strain construction. J. Bacteriol. 180:6408–6411. 21. Paulsen, I. T., M. K. Sliwinski, and M. H. Saier, Jr. 1998. Microbial genome analyses: global comparisons of transport capabilities based on phylogenies, bioenergetics and substrate specificities. J. Mol. Biol. 277: 573–592. 22. Vogel, H. J., and D. M. Bonner. 1956. A convenient growth medium for Escherichia coli and some other microorganisms (medium E). Microb. Genet. Bull. 13:43–44.

Downloaded from http://jb.asm.org/ on February 5, 2017 by guest