two transsulfurylation pathways in klebsiella pneumoniae · this pathway appears to proceed via a...

JOURNAL OF BACTERIOLOGY, Aug. 2006, p. 5762–5774 Vol. 188, No. 160021-9193/06/$08.00�0 doi:10.1128/JB.00347-06Copyright © 2006, American Society for Microbiology. All Rights Reserved.

Two Transsulfurylation Pathways in Klebsiella pneumoniaeThomas A. Seiflein and Jeffrey G. Lawrence*

Department of Biological Sciences, University of Pittsburgh, Pittsburgh, Pennsylvania 15260

Received 10 March 2006/Accepted 7 June 2006

In most bacteria, inorganic sulfur is assimilated into cysteine, which provides sulfur for methionine bio-synthesis via transsulfurylation. Here, cysteine is transferred to the terminal carbon of homoserine via itssulfhydryl group to form cystathionine, which is cleaved to yield homocysteine. In the enteric bacteria Esch-erichia coli and Salmonella enterica, these reactions are catalyzed by irreversible cystathionine-�-synthase andcystathionine-�-lyase enzymes. Alternatively, yeast and some bacteria assimilate sulfur into homocysteine,which serves as a sulfhydryl group donor in the synthesis of cysteine by reverse transsulfurylation with acystathionine-�-synthase and cystathionine-�-lyase. Herein we report that the related enteric bacteriumKlebsiella pneumoniae encodes genes for both transsulfurylation pathways; genetic and biochemical analysesshow that they are coordinately regulated to prevent futile cycling. Klebsiella uses reverse transsulfurylation torecycle methionine to cysteine during periods of sulfate starvation. This methionine-to-cysteine (mtc) trans-sulfurylation pathway is activated by cysteine starvation via the CysB protein, by adenosyl-phosphosulfatestarvation via the Cbl protein, and by methionine excess via the MetJ protein. While mtc mutants cannot usemethionine as a sulfur source on solid medium, they will utilize methionine in liquid medium via a sulfideintermediate, suggesting that an additional nontranssulfurylation methionine-to-cysteine recycling pathway(s)operates under these conditions.

Bacterial physiology is a complex network of interwovenbiochemical pathways involving numerous shared intermedi-ates, disparately regulated enzymes, and keystone metaboliteswith important roles as substrates, products, regulatory effec-tors, or a combination of all three. When genomes gain andlose genes—as they do with startling frequency (20)—the ad-dition of new enzymes and loss of others alter pools of metab-olites, overall energy demands, drains on carbon and nitrogenpools, and the balance between transport and de novo biosyn-thesis. While comparative genomics provides a record ofchanges in the gene inventory, insight into how the cell re-sponds to such changes is achieved by comparative geneticanalyses, where metabolic differences among related organ-isms are correlated with changes both in their genetic andgenomic makeup and in the deployment of their regulatoryapparatuses.

Here we examine differences in sulfur amino acid metabo-lism among three well-studied enteric bacteria, Escherichiacoli, Salmonella enterica, and Klebsiella pneumoniae. E. coli andSalmonella serve as model systems for the physiology of sulfurassimilation. In these organisms, sulfate is reduced to sulfideand assimilated into acyl-activated serine to form cysteine (14);cysteine then serves as the sulfur group donor, either directlyor indirectly, for the synthesis of all other sulfur-bearing mol-ecules in the cell, including methionine. In methionine biosyn-thesis, cysteine is combined with acyl-activated homoserine toform an intermediate compound, LL-cystathionine (7). Cysta-thionine is cleaved to form homocysteine, pyruvate, and am-monia (Fig. 1), and homocysteine is subsequently methylatedto form methionine. These two reactions—together termedtranssulfurylation—are catalyzed by the cystathionine-�-syn-

thase (MetB) and cystathionine-�-lyase (MetC) enzymes.Transsulfurylation is an irreversible process, and neither E. colinor S. enterica can use homocysteine as a sulfur group donor inthe synthesis of cysteine.

Alternatively, yeast and some bacteria assimilate sulfide intoacyl-activated homoserine to form homocysteine directly (31),which is used in the synthesis of both methionine and cysteine.In these cases, an L-allo-cystathionine intermediate is formedwhen homocysteine donates its sulfhydryl group to activatedserine through a cystathionine-�-synthase; cystathionine iscleaved by cystathionine-�-lyase to yield cysteine, �-keto-bu-tyrate, and ammonia (Fig. 1). This process is termed reversetranssulfurylation. In both transsulfurylation pathways, a sulf-hydryl group from the donor molecule (either cysteine or ho-mocysteine) is transferred to an activated alcohol (either ho-moserine or serine) to form a cystathionine intermediate,which is cleaved on the opposite side of the central sulfur atomto complete sulfhydryl group transfer.

K. pneumoniae—formally designated Klebsiella aerogenes—isan enteric bacterium closely related to both E. coli and S.enterica. Studies of physiological processes in Klebsiella providecontext for interpreting the physiology of the these othermodel organisms (2). We reported previously that, unlike E.coli and S. enterica, Klebsiella is able to use methionine as a solesource of sulfur (26). We suggested that Klebsiella may harbortwo transsulfurylation pathways, one that is used in methioninebiosynthesis (MetBC) and one that is used in methionine-to-cysteine recycling (MtcBC). Homocysteine would be createdfrom methionine intracellularly during the use and recycling ofthe methyl group donor S-adenosylmethionine (SAM), therebyallowing methionine to serve as a sole source of sulfur for cellgrowth.

Here we report physical, genetic, and biochemical evidencefor both transsulfurylation pathways in K. pneumoniae. ThemetB and metC genes encode the methionine biosynthetic en-

* Corresponding author. Mailing address: Department of BiologicalSciences, University of Pittsburgh, Pittsburgh, PA 15260. Phone: (412)624-4204. Fax: (412) 624-4759. E-mail: [email protected].

5762

on August 14, 2019 by guest

http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

zymes cystathionine-�-synthase and cystathionine-�-lyase, re-spectively. These genes are highly similar to their well-studiedE. coli and S. enterica homologues and are repressed duringmethionine excess; loss of their function results in methionineauxotrophy. We have also characterized the mtcBC operon,which encodes cystathionine-�-synthase and cystathionine-�-lyase, respectively. These proteins are expressed during peri-ods of sulfur starvation, and the presence of cysteine or sulfatein the medium reduces their expression. While mtcBC mutantsfail to use methionine as a sole sulfur source on solid medium,significant methionine-to-cysteine recycling activity is found incells grown in liquid medium, suggesting the presence of at

least one additional methionine-recycling pathway in Klebsiella;this pathway appears to proceed via a sulfide intermediate,indicating that it is metabolically distinct from the MtcBC-catalyzed pathway for methionine utilization. Partial dissectionof the regulation of the metB, metC, and mtcBC loci providesa framework for understanding both the physiological signifi-cance of the pathways and the selective forces leading to re-tention or loss of the genes over evolutionary time.

MATERIALS AND METHODS

Bacterial strains and plasmids. The strains used in these studies (Table 1)were constructed from K. pneumoniae (formally designated K. aerogenes) W70

FIG. 1. Transsulfurylation pathways in K. pneumoniae W70. The MetBC enzymes catalyze the transsulfurylation of homoserine to formhomocysteine, a precursor of methionine. The MtcBC enzymes catalyze reverse transsulfurylation of serine to form cysteine; both pathways employcystathionine intermediates. SAH, S-adenosylhomocysteine; HS-CoA, coenzyme A.

VOL. 188, 2006 TRANSSULFURYLATION IN KLEBSIELLA 5763


http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

derivative KC2668 [hsdR suc� hutC515(Con) dadA lac �bla-2]. Plasmid pTAS1bears the lamB region from E. coli and renders Klebsiella sensitive to bacterio-phage � (26); plasmid pMMK1 (13) is pTAS1 with the gene encoding the alteredtarget specificity Tn10 transposase (12).

Media and antibiotics. The rich medium used was LB; the minimal definedmedium was E (34). Sulfur-free E medium (NSE) was created by substitutingMgCl2 for MgSO4; glucose was used as a carbon source at 0.2%. Solid mediawere made by the addition of Bacto Agar (Difco) or agarose (Gibco BRL) to1.2%. P1 buffer contained 5 mM CaCl2 and 10 mM MgSO4. Ampicillin was usedat 200 �g/ml; kanamycin was used at 20 �g/ml; chloramphenicol was used at 50�g/ml for selection of Tn10dCm transposition mutants and at 20 �g/ml other-wise; tetracycline was used at 10 �g/ml for selection of transductants and at 20�g/ml otherwise; and zeomycin was used at 100 �g/ml. For growth curve deter-

mination, cells from a fresh overnight culture were rinsed and diluted 200-foldinto NSE-glucose medium; growth curves were determined with 200-�l samplesin a PowerwaveX 96-well spectrophotometer, where absorbance at 600 nm wasmeasured every 5 min.

Genetic methods. Transposition-defective derivatives of Tn10-Tn10LK (kana-mycin resistant), Tn10dTc (tetracycline resistant), and Tn10dCm (chloramphen-icol resistant) were delivered into pTAS1- or pMMK1-bearing strains of K.pneumoniae with bacteriophage � delivery vectors �NK1205, �NK1323, and�NK1324 as previously described (13). All insertion mutations were transducedinto their maternal parent strain following mutagenesis to ensure 100% linkageof the antibiotic resistance marker to the observed mutant phenotype; transduc-tion was mediated by bacteriophage P1-vir as previously described (13).

Enzyme assays. Cells were grown to mid-log phase, concentrated by centrif-ugation, rinsed, and resuspended in 1/30 volume of 100 mM KxPO4 (pH 8.0).Cells were disrupted in a Parr cell bomb by two successive rounds of compressionfor 5 min under 1,000 lb/in2 N2. Cell debris was removed by centrifugation, andthe protein-bearing supernatant was desalted on a PD-10 gel filtration column(Pharmacia), eluting in 10 mM KxPO4 (pH 8.0). Enzyme extracts were usedimmediately in assays for cystathionine-�-lyase activity (18), with final concen-trations of 8.4 mM cystathionine (all four stereoisomers) as the substrate anddetecting �-ketobutyrate as a product; assays were performed in triplicate. Spe-cific activities were calculated as micromoles of �-keto-butyrate produced perminute per milligram of protein. Activities were calculated for the substrate only,the extract only, and the complete reaction mixture (for calculation of substrate-only activities, the protein concentration of the corresponding extract was used);values were normalized to no-substrate controls, leading to some insignificantlynegative activities. Assays for �-galactosidase activity were performed as previ-ously described (17). Protein concentrations were determined by the Bradfordassay (4).

Molecular methods. For the creation of chromosomal libraries, genomic DNAfrom Tn10dCm-bearing cells was partially digested with Sau3AI and ligated intopNEB193, which was digested with BamHI and treated with alkaline phos-phatase. Plasmids were introduced into E. coli XL-2 Gold (Stratagene) selectingfor ampicillin resistance. Chloramphenicol-resistant colonies were detected byreplica printing to LB-chloramphenicol medium, and plasmids were preparedwith QIAprep Spin Miniprep kits (QIAGEN). DNA sequences were determinedon an ABI 3700 automated sequencer according to the manufacturer’s instruc-tions. The sequences across transposon insertion sites were determined fromDNA fragments amplified by the PCR from DNA prepared from wild-type cells.Inverse PCR was performed as previously described (19).

For directed gene knockouts, genes conferring resistance to chloramphenicol(cat), kanamycin (aph), and zeomycin (ble) were amplified by PCR with primerswhose 3� ends were complementary to the antibiotic resistance genes but whose5� ends comprised 45 bases identical to regions flanking the target gene to bereplaced. Following amplification of the appropriate antibiotic resistance gene byPCR with these tailed primers, allelic replacement was performed by electropo-ration of the fragment into a host cell bearing plasmid pTP223 (24), whichexpresses the � red recombinase system. Replacement of the chromosomal genewith the antibiotic resistance gene was confirmed by (i) amplification of thetarget locus via PCR and determination of the nucleotide sequences of the joinpoints between native chromosomal DNA and the introduced antibiotic resis-tance genes and (ii) linkage analysis of the antibiotic resistance gene to knownneighboring loci via P1 cotransduction assays.

Nucleotide sequence accession numbers. The sequences determined in thisstudy have been deposited in GenBank and assigned accession numbersDQ643993 to DQ64998.

RESULTS

Characterization of the metB and metC loci. Transposoninsertion mutations in the metB and metC loci were isolatedas methionine auxotrophs as described previously (26). ThemetB locus was identified by its mutant phenotype (methi-onine auxotrophy corrected by cystathionine) and by linkageto the arg and rha loci (26). The metC locus was identified byits methionine auxotrophy being corrected by homocysteinebut not by cystathionine (26). Chromosomal libraries wereconstructed from a strain bearing a metB4031::Tn10dCm ora metC4029::Tn10dCm insertion, chloramphenicol-resistanttransformants were isolated, and the DNA sequences of

TABLE 1. Bacterial strains and plasmids used in this study

Straina Relevant genotype Source orreference

LD561 hsdR suc� hutC515(Con) dadA lac-�bla-2 KC2668;R. Bender

LD806 pTAS1 26LD821 pMMK1 13LD826 cysDN4070::Tn10dTc 26LD828 cysC4022::Tn10dCm 26LD834 metB4031::Tn10dCm 26LD850 metC4029::Tn10dCm 26LD859 mtcB4028::Tn10dCm This studyLD860 mtcC4016::Tn10dCm This studyLD862 metB4046::Tn10LK metC4010::Tn10dTc This studyLD863 metB4046::Tn10LK metC4010::Tn10dTc

mtcB4028::Tn10dCmThis study

LD864 metB4046::Tn10LK metC4010::Tn10dTcmtcC4016::Tn10dCm

This study

LD865 metA4045::Tn10LK This studyLD866 metB4046::Tn10LK This studyLD867 metJ4048::ble metA4045::Tn10LK (TAS909) This studyLD868 metJ4048::ble metB4046::Tn10LK This studyLD870 cbl-4001::aph This studyLD871 cysB4075::cat This studyLD872 mtcB4001::Tn10dTc This studyLD873 cbl-4001::aph mtc-4028::Tn10dCm (TAS958) This studyLD874 cysIJ4009::Tn10dTc mtc-4028::Tn10dCm

(TAS 937)This study

LD875 mtcB4001::Tn10dTc cysM4077::Tn10dCmcysK4080::aph

This study

LD876 mtcB4001::Tn10dTc cysM4077::Tn10dCm(TAS1006)

This study

LD877 mtcB4001::Tn10dTc cysK4080::aph(TAS1007)

This study

LD878 cysW4082::Tn10dTc This studyLD879 cbl-4001::aph cysW4082::Tn10dTc This studyLD880 cysD4070::Tn10dTc cbl-4001::aph This studyLD881 metR4021::Tn10dKn (ABW411) This studyLD882 mtcC4014::Tn10dCm This studyLD883 metB4045::aph This studyLD884 metJ4048::ble (TAS899) This studyLD885 metJ4048::ble metB4050::aph (TAS1045) This studyLD886 metJ4048::ble mtcC4016::Tn10dCm This studyLD887 metC4010::Tn10dTc metB4046::Tn10LK

cysD4072::Tn10dCmThis study

LD888 metC4010::Tn10dTc metB4046::Tn10LKcysC4022::Tn10dCm

This study

LD889 cysC4022::Tn10dCm cbl-4001::aph This studyLD890 metJ4048::ble cysD4070::Tn10dTc This studyLD891 cysE4074::Tn10dCm This studyLD892 cysK4078::Tn10dKn This studyLD893 cysJ4034::Tn10dCm This studyLD894 metC4010::Tn10dTc This studyLD895 cysB4075::cat cysDN4070::Tn10dTc This study

a All strains were derived from LD561.

5764 SEIFLEIN AND LAWRENCE J. BACTERIOL.


http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

appropriate plasmid inserts were determined as describedabove. A contiguous sequence was constructed for both themetB and metC regions from overlapping clones where ap-propriate (Fig. 2A and B).

The Klebsiella metB locus encodes homologues of the E. colimetJ, metB, and metL genes (Fig. 2A); gene functions can beinferred by the high degree of similarity (proteins are 92 to96% identical) to their E. coli homologues. The metB geneencodes a cystathionine-�-synthase (Fig. 1), and disruption ofthe metB gene by a Tn10dCm insertion at codon 222 results inmethionine auxotrophy. The metL gene encodes an aspartatekinase which produces aspartyl-phosphate; this compoundserves as a precursor in lysine biosynthesis, as well as a pre-cursor in the synthesis of homoserine, an intermediate in me-thionine and threonine biosyntheses. No metB insertion resultsin lysine or threonine auxotrophy by polarity on the metL gene,suggesting that Klebsiella, like E. coli and Salmonella, mayharbor multiple genes encoding aspartate kinases, each poten-tially differentially regulated by an amino acid synthesizedfrom aspartate, i.e., methionine, threonine, or lysine.

The divergently transcribed metJ gene encodes a repressorof methionine biosynthetic genes; in E. coli and Salmonella,MetJ senses high levels of methionine by binding to SAM andeffecting transcriptional repression. Good consensus MetJbinding sites are found upstream of the Klebsiella metJ geneand metBL operon (Fig. 2C) in appropriate positions to affecttranscriptional repression of PmetB, PmetJ1, and PmetJ2 (Fig. 2C).In addition, good conservation of constitutive PmetJ3 is alsoseen upstream of the Klebsiella metJ gene. The conservation ofthe MetJ binding sites among E. coli, S. enterica, and Klebsiellasuggests that a similar mode of regulation may be present inKlebsiella.

The Klebsiella metC genes encodes a protein that is 92%identical to its E. coli metC homolog (Fig. 2B). The metC geneencodes a cystathionine-�-lyase (Fig. 1), and disruption by aTn10 insertion at codon 77 or codon 280 results in methionineauxotrophy (Fig. 2B). Two MetJ binding sites (each matchingseven of eight bases to the consensus sequence) are foundupstream of the metC gene (Fig. 2D); the conservation of thesesites in E. coli, Salmonella, and Klebsiella suggests that MetJmay repress transcription initiation at the Klebsiella metC geneduring periods of methionine excess. Homologs of the E. coliyghB and exbB genes are found adjacent to the metC gene (Fig.2B); in addition, a gene unannotated in E. coli, B3007 (graygene in Fig. 2B), appears to be encoded upstream of the exbBgene since this region shows the excess of synonymous substi-tutions expected for a protein-coding region. These data sup-port the hypothesis that this locus in Klebsiella is the bona fidemetC gene and not merely a related sequence.

Regulation of methionine biosynthetic genes. Mutations inthe metB and metC loci both result in methionine auxotrophyin Klebsiella (26), as do mutations in their E. coli and Salmo-nella homologues. If the primary role for the Klebsiella metBCgene products were in methionine biosynthesis, one wouldpredict that these genes would be induced under periods ofmethionine starvation and be repressed during periods of me-thionine excess. In E. coli and Salmonella, this regulation isprovided by the MetJ protein, which senses excess methioninevia excess SAM and represses methionine biosynthetic genes(11). The presence of a metJ homologue in Klebsiella—and

good consensus MetJ binding sites upstream of the metJ,metBL, and metC transcription units—suggests that the sameregulatory paradigm may apply.

To test this hypothesis, we isolated 12 mutants with Tn10LKinsertions in methionine biosynthesis genes; two fusions ap-peared to be in frame and showed regulated behavior (Table 2).Linkage analysis showed that one insertion affected the metAgene and one insertion affected the metB gene (creating afusion at codon 381 of the MetB protein; Fig. 2A). TheTn10LK insertions result in MetA::LacZ and MetB::LacZ fu-sion proteins that are transcribed and translated from endog-enous promoters and ribosome-binding signals. In both cases,expression of the gene fusions was decreased when the cellswere provided with excess methionine (Table 2). When thecells were provided with the methionine biosynthetic interme-diate homocysteine (Fig. 1) as a sole sulfur source, the metAand metB genes were expressed to 100-fold greater levels(Table 2). To determine if the repression of the metA and metBgenes was mediated by the MetJ protein, a metJ::ble muta-tion—whereby the entirety of the metJ coding sequence isreplaced with a cassette expressing a protein conferring zeo-mycin resistance—was constructed by directed gene knockout(24). As expected, the MetA::LacZ and MetB::LacZ fusionproteins were expressed to high levels in the presence of me-thionine in a metJ::ble mutant background (Table 2). Theseresults support the hypothesis that the MetBC transsulfuryla-tion pathway is expressed only during periods of methioninestarvation and serves to synthesize methionine from cysteine.

Isolation of mtc mutations. Unlike E. coli and Salmonella,Klebsiella can utilize methionine as a sole source of sulfur. Onsolid medium, auxotrophs with lesions in the cysPTWA orcysDN locus are completely corrected by cysteine or by methi-onine (26). We proposed (26) that methionine utilizationcould proceed by a reverse transsulfurylation pathway, onedistinct from the metBC-encoded transsulfurylation pathwayinferred from genetic analyses and characterized above. WithcysDN::Tn10 insertion mutants as parents, we isolated 14Tn10dCm, 6 Tn10dTc, and 4 Tn10LK insertion mutants thatfailed to grow on methionine on solid medium; the cysDNmutation served to eliminate growth on the residual sulfatethat is seen in wild-type strains propagated on “sulfur-free”medium. All 24 mtc mutants failed to show significant growthon methionine as a sole sulfur source after 5 days of growth onsolid medium but grew well on cysteine, sulfite, and sulfide (thecysDN mutation prevents growth on sulfate); most of the mtcmutations fell into a single linkage group (the exception, a cysEmutation, is described below).

Characterization of the mtcBC locus. A chromosomal librarywas constructed from strain LD859, bearing the mtc-4028::Tn10dCm mutation, and chloramphenicol-resistant transfor-mants were isolated as described above. The nucleotide se-quences were determined for the inserts of several clones togenerate a 3.4-kb contiguous sequence (Fig. 3A); the sequenceof the promoter region of the mtcBC operon was determinedfrom a fragment generated by inverse PCR (19) with genomicDNA as the template. The mtc-4028::Tn10dCm insertion af-fects the mtcB gene, which encodes a likely cystathionine-�-synthase. The predicted MtcB protein sequence is homologousto predicted cystathionine-�-synthases from numerous bacte-ria and eukaryotes.



http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

FIG. 2. Comparison of the metB and metC loci of K. pneumoniae and E. coli. Open reading frames in Klebsiella are named after their E. colihomologues; percent nucleotide and protein identities to homologous Klebsiella open reading frames are noted above each E. coli gene. The graybars denote the extents of plasmid inserts. (A) The metJBL locus. The triangles denote the sites of Tn10 insertions in strains LD859, LD866, andLD834. (B) The metC locus. The triangles denote the sites of Tn10 insertions in strains LD850 and LD894. (C) Comparison of the metJB promoterregion among K. pneumoniae (Kpn), E. coli (Eco), and S. enterica (Sen). MetJ binding sites lie between vertical lines; consensus 35 and 10binding sites for four promoters are noted. (D) Comparison of the metC promoter region. Consensus MetJ binding sites are shown between verticallines; sites in E. coli were inferred on the basis of footprinting (1, 11, 28) and the crystal structure of the MetJ::SAM::DNA ternary complex (29).

5766


http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

The gene downstream of the mtcB gene, termed mtcC, en-codes a likely cystathionine-�-lyase; the predicted MtcC pro-tein sequence is homologous to predicted cystathionine-�-lyases from bacterial species and eukaryotes. These twoproteins are sufficient to catalyze the reverse transsulfurylationpathway, allowing methionine to serve as a sulfur donor in thesynthesis on methionine via SAM, S-adenosylhomocysteine,and homocysteine intermediates. Unlike the metB and metCgenes, there are no good consensus MetJ binding sites betweenmtcBC operon and the upstream pheV gene, suggesting thatthe mtcBC promoter is not regulated directly by the MetJrepressor. However, there is a good consensus CysB bindingsite upstream of a potential mtcBC promoter (Fig. 3B).

The gene downstream of the mtcC gene encodes a likelyLysR family transcriptional regulator; although these genes areconvergently transcribed and their stop codons abut one an-other, there is no obvious rho-independent terminator for ei-ther transcript. The pheV gene, which encodes a tRNAPhe, islocated upstream of the mtcBC operon. A pheV homologue islocated at min 67 of the E. coli genetic map, 42 kb from themetC locus. Linkage analysis by P1 cotransduction shows thatthe mtcBC operon is 55% linked to the metC gene in K.pneumoniae, indicating that it is situated at a correspondinglocation in the chromosome. However, no evidence of anmtcBC operon is found in the genome sequences of E. colistrain K-12 (3), O157:H7 (23), or CFT073 (36) or S. entericaserovar Typhimurium (16), Typhi (22), or Choleraesuis (6) atthis or any other chromosomal location.

A nontranssulfurylation pathway allows methionine utiliza-tion in liquid medium. Although all mtcBC mutants fail to usemethionine as a sole sulfur source on solid medium, they doutilize methionine as a sole sulfur source when grown in liquidmedium (Fig. 4). The apparent diauxie of these growth curvesis consistent with the utilization of methionine only after in-ternal pools of cysteine and other sulfur sources have beenexhausted; external methionine spares these resources, allow-ing cells to achieve a higher density. Three possible routes forthe Klebsiella secondary pathway are possible. (i) A gaseousmethanethiol intermediate (26) may be reduced directly toformaldehyde and sulfide; (ii) methanethiol may be oxidizedto methanesulfonate, as has been proposed in Pseudomonas(10, 33), and then reduced to sulfide via a sulfite interme-diate; or (iii) a third transsulfurylation pathway operates inliquid medium.

To discriminate among these alternatives, mutants with

changes in the cysJ (sulfite reductase), cysK (cysteine synthaseI), and cysM (cysteine synthase II) genes were isolated. Asshown in Fig. 4, the alternative pathway appears to proceed viaa sulfide intermediate, as a cysM cysK double mutant (elimi-nating all cysteine synthase activity) eliminates methionine uti-lization in an mtcB background, while mtc mutants also lackingthe CysIJ sulfite reductase are still able to grow on methioninein liquid medium. These data indicate that mtcBC-indepen-dent methionine utilization in liquid medium proceeds througha sulfide intermediate (Fig. 5) but not a sulfite intermediate, asproposed in Pseudomonas putida (33), also excluding the ac-tion of a third transsulfurylation pathway in Klebsiella. Impor-tantly, this behavior facilitates the characterization of MtcBCfunctions, since mtcBC mutants do grow with methionine as asole sulfur source in liquid medium.

FIG. 3. (A) mtcBC locus of K. pneumoniae. The triangles denotesthe sites of Tn10 insertions in strains LD859, LD860, and LD872. Thegray bars denote the extents of plasmid inserts or inverse PCR (iPCR)fragments. (B) Promoter region of the mtcBC operon. Possible ribo-some binding and �70 binding sites are underlined. (C) Locations ofmutations in the Klebsiella cysJIH (LD893), cysE (LD891), and cysK(LD892) loci.

TABLE 2. Expression of met::lacZ translational fusions

Strain Genotype

�-Galactosidase activitya

Homocysteine Methionine Foldrepression

LD865 metA4045::Tn10LK 157.5 � 1.1 1.6 � 1.5 96LD866 metB4046::Tn10LK 101.4 � 14.7 0.92 � 0.7 111LD867 metJ4048::ble

metA4045::Tn10LK234.7 � 64.9 280.1 � 43.0 0.8

LD868 metJ4048::blemetB4046::Tn10LK

273.9 � 39.0 246.9 � 40.2 1.1

a �-Galactosidase activity was measured in strains grown in NSE–0.2% glucosewith the indicated compound serving as the sole sulfur source. Fold repressionwas calculated as the ratio of activities for cells growing on homocysteine relativeto those growing on methionine.



http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

Cysteine, but not sulfate, represses MtcC activity. To testthe hypothesis that the mtcC gene encodes a cystathionine-�-lyase, we assayed for this activity in wild-type and mtcB andmtcC mutant strains. To avoid any residual activity potentiallyconferred by the MetB or MetC enzyme, assays were per-formed in both the wild-type and metB metC mutant back-grounds. While no cystathionine-�-lyase activity was observedin mtcC::Tn10dCm mutants (Fig. 6A), wild-type cells and non-polar mtcB::Tn10dCm mutants (where the mtcC gene is ex-pressed from the Tn10dCm promoter) showed cystathionine-�-lyase activity, indicating that the mtcC gene encodes thecystathionine-�-lyase; we infer that the mtcB gene encodes thecystathionine-�-synthase. Cystathionine-�-lyase activity is notevident in wild-type cells grown in the presence of cysteine(Fig. 6B) or sulfate (Fig. 6B and C). These data suggest thatcysteine is a preferred sulfur source and the mtcBC operon isinduced only during periods of sulfur starvation.

Control of the mtcBC operon by the CysB protein. Sincecysteine has a repressive effect on cystathionine-�-lyase activityin liquid medium (Fig. 6), we postulated that the CysB proteinmay regulate mtcBC expression; a potential CysB binding siteis found in the mtcBC promoter region (Fig. 3B). The CysBprotein detects sulfur starvation as accumulation of o-acetyl-serine (via its isomer N-acetyl-serine [15]), the product of the

CysE protein. In the screen detailed above, a cysE mutationprevented growth on methionine on solid and liquid media andcaused cysteine auxotrophy even when transduced into an oth-erwise wild-type background. The Tn10dCm element in strainLD891 is located 4 bp upstream of the translation initiationsite of the cysE gene and likely results in cysteine auxotrophybecause of polar effects.

We posited that the cysE mutant fails to express the mtcBCoperon because the lack of N-acetyl-serine fails to activateCysB (Fig. 5). To test this hypothesis, we created a cysB nullmutant by directed insertion of a chloramphenicol resistancegene (cat) into the cysB coding region, replacing the cysB genewith the cat gene and its promoter. The cysB mutation preventsutilization of methionine as a sole sulfur source on solid me-dium (Table 3). Methionine utilization is also impaired inliquid medium (Fig. 4); the cysB strain grows more slowly,exhausting internal sulfur stores at 9 h (see no-sulfur andsulfate curves). These results indicate that the mtcBC operon isnot expressed in the cysB::cat mutant background.

The mtcBC operon is regulated by the Cbl protein on solidmedium. The E. coli Cbl (CysB-like) protein regulates manygenes involved in the use of alternative sources of sulfur uponsulfate depletion (9, 32). It has been proposed that Cbl recog-nizes the sulfate reduction intermediate adenosyl-phosphosul-

FIG. 4. Growth of Klebsiella strains on sulfur-free minimal medium. Sole sulfur sources were added as indicated. Strains used were LD561 (wildtype), LD870 (cbl), LD871 (cysB), LD872 (mtcB), LD873 (mtcB cbl), LD874 (mtcB cysIJ), LD876 (mtcB cysM), LD877 (mtcB cysK), and LD875(mtcB cysM cysK). Triangles denote the point during methionine utilization when internal sulfur stores have been exhausted (see text). cysB strainsgrow significantly more slowly and consistently reach this point at a later time. Values indicate the doubling times of strains growing on methionineduring exponential growth after this point. OD600, optical density at 600 nm.



http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

fate (APS) as an effector, which inhibits its ability to activatetranscription (5), and purified Cbl protein appears to bind APS(30). The phenotypes of Klebsiella cysDN and cysC mutantssupport the hypothesis that Cbl mediates gene expression; cysCmutants would accumulate APS and do not express mtcBC onsolid medium, whereas cysTWA and cysDN mutants, affecting

sulfate transport and APS synthesis, do. This behavior is con-sistent with APS acting as an anti-inducer via Cbl.

This hypothesis was tested by creating a cbl::aph mutation bydirected gene knockout; linkage of this mutation to the naclocus (25) confirmed that we had eliminated the cbl ortho-logue. MtcBC function was assessed by testing the abilities of

FIG. 5. Sulfur amino acid metabolism in K. pneumoniae. Steps for cysteine synthesis are shown with black arrows, methionine synthesis is shownwith gray arrows, and methionine-recycling pathways are shown with open arrows. Regulatory interactions are shown with dashed lines.

FIG. 6. Cystathionine-�-lyase activity in K. pneumoniae. Specific activities were calculated as micromoles of �-keto-butyrate produced perminute per milligram of protein. Strains assayed were LD862, LD863, LD864, LD860, LD887, LD888, LD561, LD826, and LD828.



http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

strains bearing cbl::aph and cysB::cat mutations to utilize me-thionine as a sole source of sulfur. While cysD mutants utilizedmethionine as a sole sulfur source on solid medium, neither acysD cysB nor a cysD cbl double mutant could use methionineas a sole sulfur source on solid medium (Table 3). These datasuggest that the Cbl protein may activate the mtcBC operon inresponse to the lack of APS in the cell, rather than repressingthe mtc operon in the absence of APS. Consistent with thishypothesis, cbl mutations did not allow cysC mutants to utilizemethionine as a sulfur source (Table 3).

Although cbl mutants fail to use methionine on solid me-dium, they do utilize methionine in liquid medium. The addi-tion of a cbl mutation results in a short lag before growth onmethionine at rates comparable to that of wild-type cells (Fig.4). In addition, cbl mtcB mutants (whereby all cysteine is madevia the alternative pathway) show no growth defect relative tocbl� mtcB strains aside from the short lag following exhaustionof internal sulfur stores, indicating that expression of the al-ternative pathway is not affected by Cbl. The less dramaticphenotype of cbl mutations relative to cysB mutations (Table 3;Fig. 4) suggests that CysB does not regulate mtcBC expressionsolely through Cbl.

The MetJ protein affects MtcBC activity. To test if MetJplays a role in methionine recycling, we examined the growthof wild-type and metJ mutant strains on various sulfur sources.On solid medium, cysD metJ double mutants failed to utilizemethionine as a sulfur source whereas cysD mutants grew well,suggesting that MetJ activity is required for methionine utili-zation (Table 3). On liquid medium (Fig. 7A to F), the lack ofthe MetJ protein decreased methionine-recycling capabilities.This result may have two underlying causes. First, the lack ofMetJ repression may cause constitutive expression of the metBand metC genes, causing a futile cycle to occur. Second, MetJmay be required for proper expression of the mtcBC operonor other genes involved in methionine utilization as a sulfursource. To separate these phenomena, we tested the effect of

a metJ mutation in the metB and mtcC mutant backgrounds,thereby eliminating any deleterious effect of futile recycling(Fig. 7). In these strains, a defect in methionine-recycling abil-ities was still observed, suggesting a direct or indirect role ofthe MetJ protein in regulating mtcBC operon expression. Webelieve the effect is likely indirect, since there is no evidence ofconsensus MetJ binding sites upstream of the mtcBC pro-moter.

The MetR protein plays no role in methionine utilization.The MetR protein up-regulates genes responsible for the con-version of homocysteine to methionine (metE and metH) uponthe accumulation of excess homocysteine concentrations. Sincethis pathway is insensitive to the absolute concentration ofeither methionine or homocysteine, we did not expect thisprotein to regulate the mtcBC operon. We examined metRmutants for defects in methionine utilization and found noneon either solid medium (data not shown) or liquid medium(Fig. 7G and H).

DISCUSSION

Roles of CysB and Cbl in mtcBC expression. MtcBC activity,and likely mtcBC expression, is eliminated in wild-type cells inthe presence of excess cysteine—which would inhibit the CysEprotein (Fig. 5)—or in cysE mutants (Table 3). The inability ofcysE mutants to assimilate sulfide prevents methionine utiliza-tion via the proposed alternative pathway as well, resulting incomplete lack of growth on methionine in liquid medium. TheCysE protein produces O-acetyl-serine, both the substrate forsulfide assimilation and the effector for the CysB activator viaits isomerization product N-acetyl-serine (15). Since MtcBCactivity produces cysteine directly from homocysteine—with-out the use of a sulfide intermediate—we cannot attribute thelack of MtcBC action to a lack of sulfide incorporation. Rather,we suspect that expression of the mtcBC operon requires ac-tivation by CysB, which is unavailable in the cysE mutant be-cause of the lack of N-acetyl-serine.

CysB may act either directly on the mtcBC operon or indi-rectly because CysB activates the cbl operon. There is a plaus-ible CysB binding site upstream of a likely mtcBC promoter,but the sequence requirements of a CysB binding site are fewand have not been readily distinguished from the sequencesdefining a Cbl binding site. To discriminate among these al-ternatives, both cysB and cbl null mutants were created bydirected gene knockout. As expected, the cysB::cat mutant wasunable to grow on methionine as a sole sulfur source on solidor liquid medium. However, cbl mutants showed a more mod-est defect in methionine utilization on solid medium than cysBmutants did (Table 3). Therefore, Cbl protein cannot be thesole activator of the mtcBC operon and cysB mutants likelyhave additional defects beyond their inability to transcribe thecbl gene. The CysB and Cbl proteins may work in tandem atthe mtcBC promoter, or CysB may have additional indirecteffects. The failure of cysDN cbl and cysC cbl mutants to utilizemethionine on solid medium is consistent with the Cbl proteinserving to activate the mtcBC operon, with APS serving as ananti-inducer. The likely accumulation of APS in cysCHIF mu-tants would prevent these strains from using methionine as asulfur source, whereas its absence from cysDN mutants wouldallow Cbl to activate mtcBC.

TABLE 3. Growth of K. pneumoniae strains on solidmedium with various sulfur sources

Strain GenotypeSulfur sourcea

None Sulfate Cysteine Methionine

LD561 Wild type �� LD878 cysW4082::Tn10dTc �� LD826 cysD4070::Tn10dTc �� LD828 cysC4022::Tn10dCm �� LD871 cysB4075::cat � LD895 cysB4075::cat

cysD4070::Tn10dTc �

LD891 cysE4070::Tn10dCm � LD870 cbl-4001::aph �� LD879 cysW4082::Tn10dTc

cbl-4001::aph ��

LD880 cysD4070::Tn10dTccbl-4001::aph

��

LD889 cysC4022::Tn10dCmcbl-4001::aph

��

LD884 metJ4048::ble �� LD890 metJ4048::ble

cysD4070::Tn10dTc ��

a Compound added to NSE medium cast with 1.2% agarose as a solidifyingagent. ��, strong growth; �, moderate growth; �, weak growth; , no growth.



http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

But things are more complex in liquid medium, where thealternative pathway allows for growth on methionine. BecausecysTWA and cysD mutants utilize methionine on solid mediumwhile cysC, cysH, and cysIJ mutants do not, we had anticipatedthat the accumulation of APS in cysC mutants (residual sulfateis present in all solid media) would prevent MtcBC activity inboth solid and liquid media. Although cystathionine-�-lyaseactivity was repressed by the presence of cysteine, the addition

of sulfate had no direct effect for cells grown in liquid medium(Fig. 6B and C). Here, wild-type cells showed no cystathionine-�-lyase activity when grown on methionine plus sulfate, likelybecause cysteine was being synthesized to levels sufficiently highto inhibit CysE, lowering levels of acetyl-serine, thus preventingCysB activity. Yet sulfate had no effect on MtcC activity in liquidmedium (Fig. 6), even though we expect that APS levels should bequite high, preventing Cbl from activating the mtcBC operon. In

FIG. 7. (A to F) Effect of a metJ mutation on methionine utilization. Strains were grown in sulfur-free minimal medium with the sulfur sourcesspecified. Strains used were LD561 (wild type), LD883 (metB), LD882 (mtcC), LD884 (metJ), LD885 (metJ metB), and LD886 (metJ mtcC). (Gto H) Effect of a metR mutation on methionine utilization. Strains were grown in sulfur-free minimal medium with the sulfur sources specified.Strains used were LD561 (wild type) and LD881 (metR). Triangles denote the point during methionine utilization when internal sulfur stores havebeen exhausted (see text). OD600, optical density at 600 nm.



http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

addition, cbl mutations resulted only in a slight lag period follow-ing exhaustion of internal sulfur stores before robust growth onmethionine (Fig. 4). Therefore, either another protein besidesCbl activates the mtcBC operon in liquid medium or cysCHIJmutants fail to grow on solid medium for regulatory reasons thatare specific to solid medium.

Role of MetJ in modulating MtcBC function. We expect thatthe CysB and Cbl proteins would be involved in mtcBC expres-sion if the operon is to be expressed during periods of sulfurstarvation. However, we also expect that mtcBC operon ex-pression may require methionine excess since homocysteine isa substrate. As detailed above, levels of methionine in the cellsare sensed by the MetJ protein via concentrations of SAM.While it normally serves as a repressor of methionine biosyn-thetic genes (8, 27, 35, 37), it is possible that MetJ may act asan activator of the mtcBC operon. Such behavior is not un-precedented, as the CysB protein acts as both a transcriptionalactivator (15) and a transcriptional repressor (21). As ex-pected, elimination of MetJ activity through directed geneknockout showed that this protein is involved in regulatingmtcBC operon expression as well. The MetJ protein operates

as a transcriptional repressor of met genes, occluding the pro-moter when bound to excess SAM. There were no MetJ bind-ing sites detected upstream of the mtcBC operon, while suchsites were readily found upstream of the metBL and metCtranscriptional units (Fig. 2). These data suggest that the MetJprotein acts indirectly to affect mtcBC transcription or that itsbinding site is significantly different when acting as an activatorat this site.

Evolution of transsulfurylation pathways in members of thefamily Enterobacteriaceae. The absence of homologues of themtcBC genes from the genomes of E. coli and S. enterica couldreflect one of two processes: gain of the mtcBC operon by theKlebsiella lineage after its divergence from the E. coli-Salmo-nella ancestor or their loss from this ancestor’s genome follow-ing relaxation of selection for function. That is, either Klebsiellaacquired the reverse transsulfurylation pathway as a novel traitor the ancestor of E. coli and Salmonella no longer found ituseful. To address this question, we looked for homologues ofthese genes in the genomes of completely sequenced �-pro-teobacteria, including the Enterobacteriaceae (Fig. 8). The dis-tribution of metB and metC genes suggests that the transsul-

FIG. 8. Distribution of transsulfurylation genes among �-proteobacteria. The similarity of homologues to the Klebsiella genes is noted. Thephylogeny of 16S rRNA genes was constructed by the maximum-likelihood method with PhyML. Genes were detected via BLAST. Thick linesdenote lineages bearing mtcBC genes, while gray lines denote lineages with orthologous replacement. C�S, cystathionine-�-synthase; C�L,cystathionine-�-lyase; C�S, cystathionine-�-synthase; C�L, cystathionine-�-lyase; G�, homologues to genes found in gram-positive bacteria (seetext); unk, unknown because a complete genome sequence was unavailable.



http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

furylation pathway is ancestral to the Enterobacteriaceae,although it has been lost in some of the pathogenic lineages.Not only are these genes found throughout these �-proteobac-teria, but—as suggested by their similarity to the Klebsiellagenes (Fig. 8)—their phylogeny is also consistent with verticalinheritance from their common ancestor (phylogeny notshown). The only exception is that the metB genes in thePasteurellaceae have been replaced by an orthologue fromgram-positive bacteria (the closest relatives are found in thegenera Geobacillus, Bacillus, Clostridium, and Listeria, indi-cated by G� in Fig. 8).

In contrast, the mtcBC genes for the reverse transsulfuryla-tion pathway are only found in the genera Klebsiella, Photorhab-dus, and Yersinia (Fig. 8A), where the genes also form anoperon. Homologues are also found in Pantoea agglomerans,although the complete genome sequence of this organism isnot available. The relationships among the mtcBC homologues(Fig. 8B) are consistent with their introduction into the ances-tor of the family Enterobacteriaceae and subsequent loss inseveral lineages, including the ancestor of E. coli and S. en-terica. While the gain of these genes in an ancestor and sub-sequent multiple losses in many lineages are not traditionallythought of as most parsimonious, we favor this scenario for tworeasons. First, the relationships among these genes are consis-tent with vertical inheritance. Second, repeated loss is notunexpected when considering the genomes of pathogenic bac-teria, where gene loss is common. Rather, multiple losses ofthe mtc operon, while the metBC genes are retained, suggestthat the reverse transsulfurylation pathway plays a less impor-tant role in enteric bacterial physiology.

Summary. K. pneumoniae encodes two transsulfurylationpathways, one functioning in methionine biosynthesis and an-other functioning in methionine-to-cysteine recycling duringsulfate starvation. This reverse transsulfurylation pathway isregulated by at least three proteins, including the CysB proteinrequired for activation, the Cbl protein required for activationon solid medium, and the MetJ protein.

ACKNOWLEDGMENTS

We thank L. Kapetanovich for technical assistance in the isolation ofmet::Tn10LK fusions, J. Floyd for assistance with mtc linkage analysis,M. M. Kolko for assistance with nucleotide sequencing, and E. A.Presley for helpful discussions.

This work was supported by grant GM62805 from the NationalInstitutes of Health and by a grant from the David and Lucille PackardFoundation.

REFERENCES

1. Belfazia, J., C. Parsot, A. Martel, C. Bouthier de la Tour, D. Margarita, G. N.Cohen, and I. Saint-Girons. 1986. Evolution in biosynthetic pathways: twoenzymes catalyzing consecutive steps in methionine biosynthesis originatefrom a common ancestor and possess a similar regulatory region. Proc. Natl.Acad. Sci. USA 83:867–871.

2. Bender, R. A. 1996. Variation on a theme by Escherichia, p. 4–9. In F. C.Neidhardt, R. Curtiss III, J. L. Ingraham, E. C. C. Lin, K. B. Low, B.Magasanik, W. S. Reznikoff, M. Riley, M. Schaechter, and H. E. Umbarger(ed.), Escherichia coli and Salmonella: cellular and molecular biology, 2nded. ASM Press, Washington, D.C.

3. Blattner, F. R., G. R. Plunkett, 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. Science277:1453–1474.

4. Bradford, M. M. 1976. A rapid and sensitive method for the quantitation ofmicrogram quantities of protein utilizing the principle of protein-dye bind-ing. Anal. Biochem. 72:248–254.

5. Bykowski, T., J. R. van der Ploeg, R. Iwanicka-Nowicka, and M. M.Hryniewicz. 2002. The switch from inorganic to organic sulphur assimilationin Escherichia coli: adenosine 5�-phosphosulphate (APS) as a signalling mol-ecule for sulphate excess. Mol. Microbiol. 43:1347–1358.

6. Chiu, C. H., P. Tang, C. Chu, S. Hu, Q. Bao, J. Yu, Y. Y. Chou, H. S. Wang,and Y. S. Lee. 2005. The genome sequence of Salmonella enterica serovarCholeraesuis, a highly invasive and resistant zoonotic pathogen. NucleicAcids Res. 33:1690–1698.

7. Greene, R. C. 1996. Biosynthesis of methionine, p. 542–560. In F. C. Neid-hardt, R. Curtiss III, J. L. Ingraham, E. C. C. Lin, K. B. Low, B. Magasanik,W. S. Reznikoff, M. Riley, M. Schaechter, and H. E. Umbarger (ed.), Esch-erichia coli and Salmonella: cellular and molecular biology, 2nd ed. ASMPress, Washington, D.C.

8. He, Y. Y., P. G. Stockley, and L. Gold. 1996. In vitro evolution of the DNAbinding sites of Escherichia coli methionine repressor, MetJ. J. Mol. Biol.255:55–66.

9. Iwanicka-Nowicka, R., and M. M. Hryniewicz. 1995. A new gene, cbl, en-coding a member of the LysR family of transcriptional regulators belongs toEscherichia coli cys regulon. Gene 166:11–17.

10. Kahnert, A., P. Vermeij, C. Wietek, P. James, T. Leisinger, and M. A.Kertesz. 2000. The ssu locus plays a key role in organosulfur metabolism inPseudomonas putida S-313. J. Bacteriol. 182:2869–2878.

11. Kirby, T. W., B. R. Hindenach, and R. C. Greene. 1986. Regulation of in vivotranscription of the Escherichia coli metJBLF gene cluster. J. Bacteriol.165:671–677.

12. Kleckner, N., J. Bender, and S. Gottesman. 1991. Uses of transposons withemphasis on Tn10. Methods Enzymol. 204:139–180.

13. Kolko, M. M., L. A. Kapetanovich, and J. G. Lawrence. 2001. Alternativepathways for siroheme synthesis in Klebsiella aerogenes. J. Bacteriol. 183:328–335.

14. Kredich, N. 1996. Biosynthesis of cysteine, p. 514–527. In F. C. Neidhardt, R.Curtiss III, J. L. Ingraham, E. C. C. Lin, K. B. Low, B. Magasanik, W. S.Reznikoff, M. Riley, M. Schaechter, and H. E. Umbarger (ed.), Escherichiacoli and Salmonella: Cellular and molecular biology, 2nd ed. ASM Press,Washington, D.C.

15. Lynch, A. S., R. Tyrrell, S. J. Smerdon, G. S. Briggs, and A. J. Wilkinson.1994. Characterization of the CysB protein of Klebsiella aerogenes: directevidence that N-acetylserine rather than O-acetylserine serves as the inducerof the cysteine regulon. Biochem. J. 299:129–136.

16. McClelland, M., K. E. Sanderson, J. Spieth, S. W. Clifton, P. Latreille, L.Courtney, S. Porwollik, J. Ali, M. Dante, F. Du, S. Hou, D. Layman, S.Leonard, C. Nguyen, K. Scott, A. Holmes, N. Grewal, E. Mulvaney, E. Ryan,H. Sun, L. Florea, W. Miller, T. Stoneking, M. Nhan, R. Waterston, andR. K. Wilson. 2001. Complete genome sequence of Salmonella enterica ser-ovar Typhimurium LT2. Nature 413:852–856.

17. Miller, J. H. 1972. Experiments in molecular genetics. Cold Spring HarborLaboratory, Cold Spring Harbor, N.Y.

18. Nagasawa, T., H. Kanzaki, and H. Yamada. 1984. Cystathionine gamma-lyase of Streptomyces phaeochromogenes. The occurrence of cystathioninegamma-lyase in filamentous bacteria and its purification and characteriza-tion. J. Biol. Chem. 259:10393–10403.

19. Ochman, H., A. S. Gerber, and D. L. Hartl. 1988. Genetic applications of aninverse polymerase chain reaction. Genetics 120:621–623.

20. Ochman, H., J. G. Lawrence, and E. Groisman. 2000. Lateral gene transferand the nature of bacterial innovation. Nature 405:299–304.

21. Ostrowski, J., and N. M. Kredich. 1991. Negative autoregulation of cysB inSalmonella typhimurium: in vitro interactions of CysB protein with the cysBpromoter. J. Bacteriol. 173:2212–2218.

22. Parkhill, J., G. Dougan, K. D. James, N. R. Thomson, D. Pickard, J. Wain,C. Churcher, K. L. Mungall, S. D. Bentley, M. T. Holden, M. Sebaihia, S.Baker, D. Basham, K. Brooks, T. Chillingworth, P. Connerton, A. Cronin, P.Davis, R. M. Davies, L. Dowd, N. White, J. Farrar, T. Feltwell, N. Hamlin,A. Haque, T. T. Hien, S. Holroyd, K. Jagels, A. Krogh, T. S. Larsen, S.Leather, S. Moule, P. O’Gaora, C. Parry, M. Quail, K. Rutherford, M.Simmonds, J. Skelton, K. Stevens, S. Whitehead, and B. G. Barrell. 2001.Complete genome sequence of a multiple drug resistant Salmonella entericaserovar Typhi CT18. Nature 413:848–852.

23. Perna, N. T., G. Plunkett, V. Burland, B. Mau, J. D. Glasner, D. J. Rose,G. F. Mayhew, P. S. Evans, J. Gregor, H. A. Kirkpatrick, G. Posfai, J.Hackett, S. Klink, A. Boutin, Y. Shao, L. Miller, E. J. Grotbeck, N. W. Davis,A. Lim, E. T. Dimalanta, K. D. Potamousis, J. Apodaca, T. S. Anantharaman,J. Lin, G. Yen, D. C. Schwartz, R. A. Welch, and F. R. Blattner. 2001.Genome sequence of enterohaemorrhagic Escherichia coli O157:H7. Nature409:529–533.

24. Poteete, A. R., and A. C. Fenton. 2000. Genetic requirements of phagelambda red-mediated gene replacement in Escherichia coli K-12. J. Bacteriol.182:2336–2340.

25. Schwacha, A., and R. A. Bender. 1993. The nac (nitrogen assimilation con-trol) gene from Klebsiella aerogenes. J. Bacteriol. 175:2107–2115.

26. Seiflein, T. A., and J. G. Lawrence. 2001. Multiple pathways for methionineto cysteine recycling in Klebsiella aerogenes. J. Bacteriol. 183:336–346.

27. Shoeman, R., B. Redfield, T. Coleman, R. Greene, A. Smith, I. Saint-Girons,



http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

N. Brot, and H. Weissbach. 1985. Regulation of methionine synthesis inEscherichia coli: effect of the metJ gene product and S-adenosylmethionineon the in vitro expression of the metB, metL, and metJ genes. Biochem.Biophys. Res. Commun. 133:731–739.

28. Smith, A. A., R. C. Greene, T. W. Kirby, and B. R. Hindenach. 1985. Isolationand characterization of the product of the methionine regulatory gene, metJ,of Escherichia coli K12. Proc. Natl. Acad. Sci. USA 82:6104–6108.

29. Somers, W. S., J. B. Rafferty, K. Phillips, S. Strathdee, Y. Y. He, T. McNally,I. Manfield, O. Navratil, I. G. Old, I. Saint-Girons, P. G. Stockley, andS. E. V. Phillips. 1994. The Met repressor-operator complex: DNA recog-nition by �-strands. Ann. N. Y. Acad. Sci. 726:105–117.

30. Stec, E., M. Witkowska, M. M. Hryniewicz, A. M. Brzozowski, A. J. Wilkinson,and G. D. Bujacz. 2004. Crystallization and preliminary crystallographicstudies of the cofactor-binding domain of the LysR-type transcriptionalregulator Cbl from Escherichia coli. Acta Crystallogr. D Biol. Crystallogr.60:1654–1657.

31. Thomas, D., and Y. Surdin-Kerjan. 1997. Metabolism of sulphur amino acidsin Saccharomyces cerevisiae. Microbiol. Mol. Biol. Rev. 61:503–532.

32. van der Ploeg, J. R., R. Iwanicka-Nowicka, M. A. Kertesz, T. Leisinger, and

M. M. Hryniewicz. 1997. Involvement of CysB and Cbl regulatory proteins inexpression of the tauABCD operon and other sulfate starvation-induciblegenes in Escherichia coli. J. Bacteriol. 179:7671–7678.

33. Vermeij, P., and M. A. Kertesz. 1999. Pathways of assimilative sulfur metab-olism in Pseudomonas putida. J. Bacteriol. 181:5833–5837.

34. Vogel, H., and F. Jahnig. 1986. Models for the structure of outer-membraneproteins of Escherichia coli derived from Raman spectroscopy and predictionmethods. J. Mol. Biol. 190:191–199.

35. Weissbach, H., and N. Brot. 1991. Regulation of methionine synthesis inEscherichia coli. Mol. Microbiol. 5:1593–1597.

36. Welch, R. A., V. Burland, G. Plunkett III, P. Redford, P. Roesch, D. Rasko,E. L. Buckles, S. R. Liou, A. Boutin, J. Hackett, D. Stroud, G. F. Mayhew,D. J. Rose, S. Zhou, D. C. Schwartz, N. T. Perna, H. L. Mobley, M. S.Donnenberg, and F. R. Blattner. 2002. Extensive mosaic structure revealedby the complete genome sequence of uropathogenic Escherichia coli. Proc.Natl. Acad. Sci. USA 99:17020–17024.

37. Wu, W. F., M. L. Urbanowski, and G. V. Stauffer. 1993. MetJ-mediatedregulation of the Salmonella typhimurium metE and metR genes occursthrough a common operator region. FEMS Microbiol. Lett. 108:145–150.



http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

two transsulfurylation pathways in klebsiella pneumoniae · this pathway appears to proceed via a...

Documents