Tra G Lld E Mar il Jua . L *Dep Ba 0802 ol Harv hus Recei To bran and muta bloc cons eithe acqu and Subs anid ton Com tive sizin by th show thre mM. its o on L Ke cola ton m D-L 2-hy Esch (for dized zym gene code 0– ne rm rte rtr Al id be un e l e A e t A rce vo e a tie ea ate AT Bac dy introducing multiple null mutations into the chromosome of strain MC4100. First, a Ddld allele was created by the following procedure. 1 P Biote 2 To dress Biochemical and Biophysical Research Communications 290, 824–829 (2002) doi:10.1006/bbrc.2001.6255, available online at http://www.idealibrary.com on 0006- © 2002 All righ actate, L-lactate and glycolate, three different droxymonocarboxilic acids, can all be utilized by erichia coli as a sole source of carbon and energy reviews see 1–3). Each of these compounds is oxi- to its corresponding keto acid by a different en- e: D-lactate dehydrogenase (encoded by the dld at 47.8 min) (4–9), L-lactate dehydrogenase (en- d by the D gene of the lld operon at 80.1 min) A 59-flanking sequence (0.8-kb) of dld was PCR-amplified from the chromosomal DNA with primers Dld-5N (59-CCAGCGGCCGCA- GATACGTATCTTCGCC-39) and Dld-5C (59-CCTGGATCCGTGAG- CAGGTGTGAAGAACC-39). The PCR product was digested with NotlI and BamHI and cloned between the corresponding sites of pKO3 (22), giving pDld1. A 39-flanking sequence (0.9-kb) of dld was then PCR-amplified with primers Dld-3N (59-CCAGGATCCGA- CCAACAGCATGAATCC-39) and Dld-3C (59-CACGTCGACGAAC- CACCCGGATCTG-39). The PCR product was digested with BamHI and SalI and cloned between the corresponding sites of pDld1, re- sulting in pDld2. Plasmid pDld2 was transformed into strain MC4100 (23) to delete the dld gene by homologous recombination as described previously (22), yielding strain ECL5103. Deletion of dld gene was confirmed by PCR. Next, a Dlld::Kanr allele was created. A Kanr cassette was isolated resent address: Korea Research Institute of Bioscience and chnology, Yusong, Taejon 305-600, Korea. nsport of L-Lactate, D-Lactate, and P and GlcA Membrane Carriers of ı´a Felisa. Nu´n˜ez,* Ohsuk Kwon,†,1 T. Hastings W n Aguilar,*,2 Laura Baldoma,* and Edmund C. C artment of Biochemistry, Faculty of Pharmacy, University of 8 Barcelona, Spain; and †Department of Microbiology and M ard Medical School, 200 Longwood Avenue, Boston, Massac ved December 17, 2001 examine the substrate specificity of the mem- e transport carriers LldP (L-lactate permease) GlcA (glycolate permease) of Escherichia coli, a nt strain lacking their structural genes and ked in the metabolism of the tested substrates was tructed and transformed with a plasmid bearing r the lldP or the glcA gene. Each transformant ired the ability to accumulate L-lactate, D-lactate, glycolate against a high concentration gradient. trate accumulation was inhibited by carbonyl cy- e m-chlorophenylhydrazone, a hydrophobic pro- conductor that dissipates proton motive force. petition of 14C-L-lactate transport by nonradioac- L-lactate, D-lactate, and glycolate in LldP synthe- g cells and competition of 14C-glycolate transport e same three substrates in GlcA synthesizing cells ed that both carriers effectively transported all e substrates with a Ki value ranging from 10 to 20 D-Lactate does not appear to have a permease of wn. Utilization of the compound depends mainly ldP. © 2002 Elsevier Science y Words: Escherichia coli; L-lactate; D-lactate; gly- te; membrane transport; substrate specificity; pro- otive force. (1 ge fo ve ta ac to po th th at fo in w er m ic M stu as a 1 tween whom correspondence and reprint request should be ad- ed. Fax: 34 93 402 4520. E-mail:
[email protected]. 824291X/02 $35.00 Elsevier Science ts reserved. lycolate by the scherichia coli son,† in† rcelona, Avenida Diagonal 643, ecular Genetics, etts 02115 12), and glycolate oxidase (encoded by the DEF s of the glc operon at 67.3 min) (2). The product ed by glycolate oxidase, glyoxylate, is in turn con- d to malate by malate synthase G (3, 13–18) or to onic semialdehyde by glyoxylate carboligase (19). though each of the three 2-hydroxymonocarboxilic s is metabolized by a different enzyme, there seems only two potential routes of entry for these com- ds. L-Lactate permease (encoded by the P gene of ld operon) (10) and glycolate permease (encoded by gene of the glc operon) were both found to medi- he uptake of D-lactate, L-lactate, and glycolate (20). D-lactate transport system driven by proton motive has been reported, although the gene and protein lved have not been characterized (21). In this study nalyzed the substrate specificity and kinetic prop- s of L-lactate permease (LldP) and glycolate per- se (GlcA) and found no permease specifically ded- d to D-lactate transport. ERIALS AND METHODS terial strains. E. coli strains, phage, plasmids used in this are listed in Table 1. Strain ECL5106 was constructed by .6-kb BamHI fragment from pUC4-KIXX (24) and cloned be- the BamHI and BglII sites of pLct2 (10), resulting in pDlld repla previ strain ECL5 Thi obtain lambd by gly were corres AAAT GTGG son w of the Fin was P Con 1.8-kb from (59-CA CGGA geste ing si Rec mid D DNA purification Kit (Promega, Madison, WI) and Quiaprep spin Miniprep kit (Quiagen, Valencia, CA). Oligonucleotides used in this study were synthesized either by Biopolymer Laboratory at the De- rtm rd era eci rifi ). icro ne Gro re d c , an g L- M sein col rm ioga cub Per 0.1 g dr luti e a act col we re-s the ese n o CC S an ag On on co ork ow e a te To vid CL et pla eth ow rm r q er re FN rt ox TABLE 1 E. coli K-12 Strains, Bacteriophage, and Plasmids St pha pla Strain MC JM JC7 JA2 JA2 JA2 EC EC EC EC Phag P1v Plasm pBR pEX pKO pLc pU pFN pGl PD pDl Vol. 290, No. 2, 2002 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS ::Kanr. This plasmid was transformed into strain JC7623 to ce the chromosomal lldPRD operon by the Dlld::Kanr allele as ously described (25). The Dlld::Kanr allele of the recombinant ECL5104 was subsequently P1 transduced into the strain 103 to give strain ECL5105. rd, the glycolate oxidase deficient mutant strain JA203 was ed by Tn10 mutagenesis of strain JA200 (glcA::Cmr) with a NK1098. Tetracycline resistant transductants were selected colate negative phenotype. The Tn10 chromosomal insertions mapped by inverse PCR and nucleotide sequencing of the ponding junctions using primers miniTn10-A (59-CCA- CATTAGGGGATTCATCAG-39) and miniTn10-B (59-TAAG- ATACACATCTTGTC-39). In strain JA203 the Tn10 transpo- as found to be located in gene glcE, 10 nucleotides downstream ATG codon. ally, the glcA::Cmr glcE::miniTn10(Tetr) locus of strain JA203 1 transduced into the strain ECL5105 to yield strain ECL5106. struction of a plasmid that expresses glycolate permease. A DNA fragment containing glcA sequence was PCR-amplified the chromosomal DNA of strain MC4100 with primers GlcA-N CGAATTCATGGTTACCTGGACC-39) and GlcA-C (59-CT- TCCTAGATGTTCTCACGCTC-39). The PCR product was di- d with EcoRI and BamHI and cloned between the correspond- tes of pEXT20 (26) resulting pGlcA. ombinant DNA techniques and PCR. Chromosomal and plas- NA were isolated, by using respectively the Wizard Genomic pa va m Pr pu CA M Ge we an 40 in 0.1 ca gly pe th in in m di giv L-l gly ml po of pr tio (C RE Tr K a w sh ar la di E (T a M sh pe fo th tu (p po pr Used in This Study rains, ge, and smids Relevant genotype Source/ Reference s 4100 F2 araD139 D(argF-lac)U169 rpsL150 relA1 flbB5301deoC ptsF25 rbsR 23 109 recA1 endA1 gyrA96 thi hsdR17 supE44 relA1 D(lac-proAB)/ F9 traD36 proAB1 laclq lacZ DM15 Promega 623 recB21 recC22 sbcB15 sbcC201 31 00 MC4100 glcA::Cmr 20 02 MC4100 glcA::Cmr lldP::Tn5 20 03 glcA::Cmr glcE::miniTn10(Tetr) This study L5103 MC4100 but Ddld This study L5104 JC7623 but Dlld::Kanr This study L5105 MC4100 but Ddld Dlld::Kanr This study L5106 MC4100 but Ddld Dlld::Kanr glcA::Cmr glcE::miniTn10(Tetr) This study es ir Laboratory stock ids 322 Biolabs T20 26 3 22 t2 lldPRD1 10 C4-KIXX 24 20 lldP1 20 cA glcA1 This study ld2 Ddld This study ld::Kanr This study 825 ent of Biological Chemistry and Molecular Pharmacology, Har- Medical School or by Integrated DNA Technologies Inc. Poly- se chain reactions (PCR) were carried out, by using the TaqPlus sion PCR system (Stratagene, La Jolla, CA). PCR products were ed, using QIAquick PCR purification kit (Quiagen, Valencia, Sequence verification of PCR-amplified DNA was performed at Core Facility of the Department of Microbiology and Molecular tics, Harvard Medical School. wth of cells. Luria-Bertani (LB) broth and LB-agar (17 g/liter) used for routine growth. Ampicillin, tetracycline, kanamycin, hloramphenicol were provided at a final concentration of 50, 12, d 20 mg/ml, respectively. For permease assay, strain express- lactate permease was cultured in minimal medium containing 3-[N-morpholino] propanesulfonic acid (MOPS) (pH 7.6), 1% hydrolysate, and 20 mM D-xylose (10). The strain expressing ate permease was grown in LB. The expression of glycolate ease was induced by addition of 100 mM isopropyl b-D- lactopyranoside into the exponentially growing cultures and ated for additional 2 h before harvest. mease assays. Cells were collected, washed twice, suspended M Mops and 0.5 mM Mg21 (pH 7.0) at a final cell density of 0.5 y wt/ml, and placed at 15°C. The rate of uptake was assayed by ng the 14C-labeled substrate 10-fold with the cell suspension to final concentration of 4 mM. The radioactive substrates: 14C- ate (116 mCi/mmol), 14C-D-lactate (56 mCi/mmol) and 14C- ate (50 mCi/mmol) were purchased from ICN. Samples of 100 re taken at different intervals and filtered through 0.65-mm- ize cellulose nitrate filters. The filters were washed with 4 ml same MOPS buffer, placed in plastic vials, and counted in the nce of Emulsifier-safe (Packard, Meriden, CT). The concentra- f the uncoupler carbonyl cyanide m-chlorophenylhydrazone P) used was 100 mM. ULTS sport of D-Lactate, L-Lactate, and Glycolate ainst Concentration Gradient by LldP and GlcA the basis of substrate competition, Matin and ings suggested that D-lactate and L-lactate shared mmon transport system (27). Nun˜ez and co- ers extended the study by mutant analysis which ed that two different permeases, LldP and GlcA, ctive on D-lactate, as well as L-lactate and glyco- (20). characterize the transport properties of each in- ual transport carrier, we transformed strain 5106 [Ddld Dlld::Kanr glcA::Cmr glcE:: miniTn10- r)] with a plasmid pFN20 bearing the lldP gene or smid pGlcA bearing glcA gene (see Materials and ods and Table 1). A preliminary experiment ed that transport of the three substrates by each ease was excessively rapid at room temperature uantitative studies. All subsequent assays were efore conducted at 15°C. At this lower tempera- , the transport of L-lactate by strain ECL5106 20) was close to linear for 1 min, and the trans- of glycolate by strain ECL5106 (pGlcA) was ap- imately linear for about 20 s. Control strains bear- ing a rate Af eithe mid L-lac trace of gr intra Malo conc cells effec max plas is no actu bran dete bs by Pr ect lac ov e p ). ed ot ob rm th ta ol tr o on i of In bs FIG (A) L- (pBR (B) G ECL5 ECL5 TABLE 2 Substrate Accumulation against Concentration Gradient by ran bst Lac Lac yco a A b pB Eff T co a C L5 b E c En Vol. 290, No. 2, 2002 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS plasmid vector without insert showed a negligible of transport (Fig. 1). ter 1 min of incubation, cells transformed with r the lldP-bearing or the glcA-bearing plas- were able to accumulate radiolabeled D-lactate, tate and glycolate to a concentration gradient (in- llular concentration/extracellular concentration) eater than 10. The calculation was based on the cellular volume of water determined according to ney et al. (28) (Table 2). Cells synthesizing LldP entrated L-lactate most effectively (21-fold), whereas synthesizing GlcA concentrated glycolate most tively (70-fold). A control experiment showed that imal substrate concentration by cells bearing a mid vector without insert did not exceed 3-fold. It t clear whether this baseline retention of substrate ally represented transport across the cell mem- e, because the assay did not permit the direct rmination of zero time values. Su el L- m th 30 m pr ph no of up ab on tw ti K su . 1. Rates of cognate substrate uptake by LldP and GlcA. Lactate accumulation. Open circles, control strain ECL5106 322); filled circles, LldP producing strain ECL5106 (pFN20). lycolate accumulation. Open squares, control strain 106 (pEXT20); filled squares, GlcA producing strain 106 (pGlcA). T Su L- D- Gl by Re EC 826 trate Transport by LldP and GlcA Driven Proton Motive Force evious observations on the effects of inhibitors of ron transport chains on the transport of D- and tate transport and the association of substrate ement with proton flux led to the suggestion that rocess is driven by proton motive force (21, 27, 29, To test whether the substrate transport processes iated by LldP and GlcA are both dependent on on motive force, we employed CCCP as a hydro- ic proton conductor. This reagent abolishes the al proton gradient across the plasma membrane e cell. If proton cotransport is the mechanism of ke of the substrates, their accumulation should be ished by CCCP. Table 3 shows the effect of CCCP ansport of L-lactate, D-lactate and glycolate by the carriers. Strong inhibition of substrate accumula- by CCCP was observed in all cases. the Three Substrates for Both Permeases order to compare the relative affinity of the three trates for LldP, the inhibitory effects of nonradio- sformants of Strain ECL5106 Bearing Different Vectors rate Concentration gradient (inside/outside)a Transport by LldP Transport by GlcA pBR322 control pFN20b pEXT20 control pGlcA tate 0.40 6 0.1 21 6 2.0 0.42 6 0.06 52 6 11 tate 0.66 6 0.2 11 6 1.6 0.82 6 0.04 55 6 12 late 2.60 6 0.1 15 6 2.6 1.70 6 0.3 70 6 15 ssayed after 1 min of incubation. R322 bearing lldP. TABLE 3 ect of CCCP on the Transport of the Three Substrates ransformants of Strain ECL5106 Bearing Different mbinant Plasmids Substrates % Inhibitiona LldPb GlcAc L-Lactate 91 6 1.8 94 6 1.4 D-Lactate 77 6 1.1 89 6 1.3 Glycolate 93 6 1.3 97 6 0.7 alculated as % inhibition of substrate accumulation in strain 106 (pFN20) or strain ECL5106 (pGlcA) by 100 mM CCCP. ncoded by pFN20. coded by pGlcA. active substrates (5 to 50 mM) on 14C-L-lactate (4 mM) uptake were tested. The data are presented as the recip hibit inhib Ki. A the G The varie Be thre poun tran Cons mM) both ar Be ete at ta gg g b na M gn ne de 20 a e ed ISC Th dP m ate en bs rm ns 13 e p Th d ce e L qu % em an nc co ry FIG uptak produ ECL5 glycol TABLE 4 Apparent Affinity of Substrates for LldP and a Th b E c En Vol. 290, No. 2, 2002 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS rocal of % inhibition against the reciprocal of in- or concentration in Fig. 2A. The concentration of itor that exerted 50% inhibition was taken as the set of similar experiments was carried out with lcA with 14C-glycolate as the substrate (Fig. 2B). Ki values of both carriers for all three substrates d from 10 to 20 mM (Table 4). cause of structural similarity of acetate with the e substrates, we also tested the two-carbon com- d as a nonradioactive inhibitor for 14C-L-lactate sport by LldP and 14C-glycolate transport by GlcA. istent with a previous conclusion (20), acetate (0.5 was found to be a very weak inhibitor. The Ki for carriers were .100 mM (data not shown). Se pl th up su in ge m si ge un JA as em m D Ll is tr di su pe co of th an an th se 80 se m co in ve . 2. Double reciprocal plots of % inhibition of 14C-L-lactate e against inhibitor concentration for LldP and GlcA. (A) LldP cing strain ECL5106 (pFN20). (B) GlcA producing strain 106 (pGlcA). Circles, L-lactate; triangles, D-lactate; squares, ate. 827 ch for a Specific D-Lactate Transporter cause a mutant lacking both LldP and GlcA com- ly failed to grow on D-lactate, it was suggested E. coli has not a specific permease dedicated to the ke of this substrate (20). In confirmation of this estion, we found that when strain ECL5106, lack- oth LldP and GlcA as well as the three dehydro- ses, were grown on D-xylose in the presence of 20 D-lactate as inducer, the cells were still unable to ificantly accumulate D-lactate. To test for a cryptic for D-lactate permease, several attempts were rtaken to select for spontaneous mutants of strain 2 (dld1glcA::cat lldP::Km) for growth on D-lactate sole carbon and energy source. No mutants rged after screening 1010 cells on agar D-lactate ium plated at different cell densities. USSION e substrate specificity and kinetic properties of and GlcA are strikingly similar. The resemblance anifested both in the effective ability to concen- L-lactate, D-lactate, and glycolate against a gra- t and in the high apparent affinities for the three trates. The variation of the Ki values of the two eases for the three substrates from 10 to 20 mM is istent with the finding of Kang who reported a Km mM for the transport of D-lactate by cells grown in resence of D-lactate (21). e similarity in the functional properties of LldP GlcA probably reflect a relatively recent common stry of the two proteins, both of which belong to ldP family of membrane carriers. The amino acid ences of the LldP and GlcA show 65% identity and similarity (20). Nonetheless, each gene product s to be well on its way to specialization. This is ifested by the facts that LldP is most effective in entrating L-lactate, whereas GlcA is most effective ncentrating glycolate. The functional divergence probably followed the integration of the dupli- GlcA Transport Proteins Substrates Ki (mM)a LldPb GlcAc L-Lactate 10.5 6 0.2 10.2 6 1.2 D-Lactate 13.3 6 1.6 9.2 6 2.4 Glycolate 16.6 6 2.4 15.8 6 1.2 e values were derived from the data presented in Fig. 2. ncoded by pFN20. coded by pGlcA. cated ancestral genes into separate operons under dif- ferent regulation. On GlcA ogni L-lac indic carb inhe D-lac D-lac pre-e sion degr allow tran show impa (20). nor L by D incre strai migh indu ACK We was s Inves Publi Instit REFE 1. G co F M U cr 2. P lo su B 3. P a E iz 4. D st sp 5. K tr ti g 6. K d er 7. M and generation of the proton electrochemical gradient in mem- brane vesicles from Escherichia coli GR19N and in proteolipo- somes reconstituted with purified D-lactate dehydrogenase and cy . R (1 to 1 . T m a li . D E u . F L p . N h g ti F . H in . K co ro J . L B . M (1 th E . O la 1 . V g ri . C u er d 3 . N L co fu 1 . K n B . L g w ch . C g p . B fo G . O Vol. 290, No. 2, 2002 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS the other hand, during the course of LldP and evolution, the basic mechanism of substrate rec- tion seems to have been retained. Recognition of tate, D-lactate, and glycolate but not acetate would ate that the active site interacts only with the oxyl and hydroxyl groups of the substrate. This rent ability of the two carriers to transport tate may account for either the failure for a special tate permease to emerge or the loss of such a xisting permease. The relative high basal expres- of the lld operon (10), in contrast to the high ee of inducibility of the glc operon (2), apparently ed LldP to assume the major role as a D-lactate sporter. The importance of LldP in this regard is n by the observation that growth on D-lactate is ired by a null mutation in lldP but not in glcA It might be added that neither L-lactate permease -lactate dehydrogenase were significantly induced -lactate in strain ECL5103 (data not shown). The ased expression of the lld operon in wild type n MC4100 grown in the presence of D-lactate t be attributable to the phenomenon of product ction mediated by pyruvate (20). NOWLEDGMENTS thank Dimitris Georgellis for helpful discussions. This work upported by Grant BCM 2001-3003 of the Direccio´n General de tigacio´n, Ministerio de Cienciay Tecnologı´a, Spain, and by U.S. c Health Service Grant GM40993 from NIGMS of the National utes of Health. RENCES ennis, R. B., and Stewart, V. (1996) Respiration. In Escherichia li and Salmonella: Cellular and Molecular Biology (Neidhardt, . C., Curtiss, R., III, Ingraham, J. L., Lin, E. C. C., Low, K. B., agasanik, B., Reznikoff, W. S., Riley, M., Schaechter, M., and mbarger, H. E., Eds.), pp. 217–261, American Society for Mi- obiology, Washington, D. C. ellicer, M. T., Badia, J., Aguilar, J., and Baldoma, L. (1996) glc cus of Escherichia coli: Characterization of genes encoding the bunits of glycolate oxidase and the glc regulator protein. J. acteriol. 178, 2051–2059. ellicer, M. T., Fernandez, C., Badia, J., Aguilar, J., Lin, E. C. C., nd Baldoma, L. (1999) Cross-induction of glc and ace operons of scherichia coli attributable to pathway intersection. Character- ation of the glc promoter. J. Biol. Chem. 274, 1745–1752. ym, O., Pratt, E. A., Ho, C., and Eisenberg, D. (2000) The crystal ructure of D-lactate dehydrogenase, a peripheral membrane re- iratory enzyme. Proc. Natl. Acad. Sci. USA 97, 9413–9418. ohn, L. D., and Kaback, H. R. (1973) Mechanisms of active ansport in isolated bacterial membrane vesicles. XV. Purifica- on and properties of the membrane-bound D-lactate dehydro- enase from Escherichia coli. J. Biol. Chem. 248, 7012–7017. ovatchev, S., Vaz, W. L. C., and Eibl, H. (1981) Lipid depen- ence of the membrane-bound D-lactate dehydrogenase of Esch- ichia coli. J. Biol. Chem. 256, 10369–10374. atsushita, K., and Kaback, H. R. (1986) D-Lactate oxidation 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 828 tochrome o oxidase. Biochemistry 25, 2321–2327. ule, G. S., Pratt, E. A., Chin, C. C. Q., Wold, F., and Ho, C. 985) Overproduction and nucleotide sequence of the respira- ry D-lactate dehydrogenase of Escherichia coli. J. Bacteriol. 61, 1059–1068. anaka, Y., Anraku, Y., and Futai, M. (1976) Escherichia coli embrane D-lactate dehydrogenase. Isolation of the enzyme in ggregated form and its activation by Triton X-100 and phospho- pids. J. Biochem. (Tokyo) 80, 821–830. ong, J. M., Taylor, J. S., Latour, D. J., Iuchi, S., and Lin, . C. C. (1993) Three overlapping lct genes involved in L-lactate tilization by Escherichia coli. J. Bacteriol. 175, 6671–6678. utai, M., and Kimura, H. (1977) Inducible membrane-bound -lactate dehydrogenase from Escherichia coli. Purification and roperties. J. Biol. Chem. 252, 5820–5827. ishimura, Y., Tan, P. K., Ohgami, Y., Kohgami, K., and Kami- ara, T. (1983) Induction of membrane-bound L-lactate dehydro- enase in Escherichia coli under conditions of nitrate respira- on, fumarate reduction and trimethylamine N-oxide reduction. EMS Microbiol. Lett. 17, 283–286. ansen, R. W., and Hayashi, J. A. (1962) Glycolate metabolism Escherichia coli. J. Bacteriol. 83, 679–687. ornberg, H. L., and Sadler, J. R. (1961) The metabolism of C2 mpounds in microorganisms. 8. A dicarboxylic acid cycle as a ute for the oxidation of glycollate by Escherichia coli. Biochem. . 81, 503–513. ord, J. M. (1972) Glycolate oxidoreductase in Escherichia coli. iochem. Biophys. Acta 267, 227–237. olina, I., Pellicer, M. T., Badia, J., Aguilar, J., and Baldoma, L. 994) Molecular characterization of Escherichia coli malate syn- ase G. Differentiation with the malate synthase A isoenzyme. ur. J. Biochem. 224, 541–548. rnston, L. N., and Ornston, M. K. (1969) Regulation of glyoxy- te metabolism in Escherichia coli K12. J. Bacteriol. 98, 1098– 108. anderwinkel, E., and De Vlieghere, M. (1968) Physiologie et e´ne´tique de l’isocitritase et des malate synthase chez Esche- chia coli. Eur. J. Biochem. 5, 81–90. hang, Y. Y., Wang, A. Y., and Cronan, J. E., Jr. (1993) Molec- lar cloning, DNA sequencing, and biochemical analyses of Esch- ichia coli glyoxylate carboligase. An enzyme of the acetohy- roxy acid synthase-pyruvate oxidase family. J. Biol. Chem. 268, 911–3919. un˜ez, M. F., Pellicer, M. T., Badia, J., Aguilar, J., and Baldoma, . (2001) The gene yghK linked to the glc operon of Escherichia li encodes a permease for glycolate that is structurally and nctionally similar to L-lactate permease. Microbiology 147, 069–1077. ang, S. Y. (1978) Mechanism of autoenergized transport and ature of energy coupling for D-lactate in Escherichia coli. J. acteriol. 136, 867–873. ink, A. J., Phillips, D., and Church, G. M. (1997) Methods for enerating precise deletions and insertions in the genome of ild-type Escherichia coli: Application to open reading frame aracterization. J. Bacteriol. 179, 6228–6237. asadaban, M. J. (1976) Transposition and fusion of the lac enes to selected promoters in Escherichia coli using bacterio- hage lambda and Mu. J. Mol. Biol. 104, 541–555. arany, F. (1985) Single-stranded hexameric linkers: A system r in-phase insertion mutagenesis and protein engineering. ene 37, 111–123. den, K. L., DeVeaux, L. C., Vidat, C. R., Cronan, J. E., Jr., and Gennis, R. B. (1990) Genomic replacement in Escherichia coli K-12 using covalently closed circular plasmid DNA. Gene 96, 29–36. 26. Dykxhoorn, D. M., St. Pierre, R., and Linn, T. (1996) A set of compatible lac promoter expression vectors. Gene 177, 133–136. 27. Matin, A., and Konings, W. N. (1973) Transport of lactate and succinate by membrane vesicles of Escherichia coli, Bacillus subtilis and a Pseudomonas species. Eur. J. Biochem. 34, 58–67. 28. Maloney, P. C., Kashket, E. R., and Wilson, T. H. (1996) Methods for studying transport in bacteria. In Methods in Membrane Biology (Korn, E. D., Ed.), Vol. 5, pp. 1–49, Plenum, New York. 29. Collins, S. H., Jarvis, A. W., Lindsay, R. J., and Hamilton, W. A. (1976) Proton movements coupled to lactate and alanine trans- port in Escherichia coli: Isolation of mutants with altered stoi- chiometry in alanine transport. J. Bacteriol. 126, 1232–1244. 30. Ten Brink, B., and Konings, W. N. (1980) Generation of an electrochemical proton gradient by lactate efflux in membrane vesicles of Escherichia coli. Eur. J. Biochem. 111(1), 59–66. 31. Wackernagel, W. (1973) Genetic transformation in Escherichia coli: The inhibitory role of the recBC DNase. Biochem. Biophys. Res. Commun. 51, 306–311. Vol. 290, No. 2, 2002 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS 829 MATERIALS AND METHODS TABLE 1 RESULTS FIG. 1 TABLE 2 TABLE 3 FIG. 2 TABLE 4 DISCUSSION ACKNOWLEDGMENTS REFERENCES