Acid-Inducible Transcription of the Operon Encoding the Citrate Lyase Complex of Lactococcus lactis...

JOURNAL OF BACTERIOLOGY, Sept. 2004, p. 5649–5660 Vol. 186, No. 170021-9193/04/$08.00�0 DOI: 10.1128/JB.186.17.5649–5660.2004Copyright © 2004, American Society for Microbiology. All Rights Reserved.

Acid-Inducible Transcription of the Operon Encoding the CitrateLyase Complex of Lactococcus lactis Biovar

diacetylactis CRL264Mauricio G. Martín, Pablo D. Sender, Salvador Peiru, Diego de Mendoza, and Christian Magni*

Instituto de Biología Molecular y Celular de Rosario (IBR-CONICET) and Departamento de Microbiología,Facultad de Ciencias Bioquímicas y Farmaceuticas, Universidad Nacional de Rosario, Rosario, Argentina

Received 28 January 2004/Accepted 24 May 2004

Although Lactococcus is one of the most extensively studied lactic acid bacteria and is the paradigm forbiochemical studies of citrate metabolism, little information is available on the regulation of the citrate lyasecomplex. In order to fill this gap, we characterized the genes encoding the subunits of the citrate lyase ofLactococcus lactis CRL264, which are located on an 11.4-kb chromosomal DNA region. Nucleotide sequenceanalysis revealed a cluster of eight genes in a new type of genetic organization. The citM-citCDEFXG operon (citoperon) is transcribed as a single polycistronic mRNA of 8.6 kb. This operon carries a gene encoding a malicenzyme (CitM, a putative oxaloacetate decarboxylase), the structural genes coding for the citrate lyase subunits(citD, citE, and citF), and the accessory genes required for the synthesis of an active citrate lyase complex (citC,citX, and citG). We have found that the cit operon is induced by natural acidification of the medium during cellgrowth or by a shift to media buffered at acidic pHs. Between the citM and citC genes is a divergent open readingframe whose expression was also increased at acidic pH, which was designated citI. This inducible response toacid stress takes place at the transcriptional level and correlates with increased activity of citrate lyase. It issuggested that coordinated induction of the citrate transporter, CitP, and citrate lyase by acid stress providesa mechanism to make the cells (more) resistant to the inhibitory effects of the fermentation product (lactate)that accumulates under these conditions.

Many bacteria can utilize citrate under fermentative condi-tions. The citrate pathway has been extensively studied in en-terobacteria (4, 26). In all known citrate fermentation path-ways, after its uptake into the cell, citrate is split into acetateand oxaloacetate by the enzyme citrate lyase. In Klebsiellapneumoniae the citrate-specific fermentation genes form acluster of two divergent operons (4, 26). This cluster includesthe genes citDEF, encoding the citrate lyase subunits �, �, and�, and the citS gene, encoding a citrate H2�/Na1� proton mo-tive force-dissipating transporter. Associated with this cluster,K. pneumoniae contains the oadGAB genes, encoding the bi-otin oxaloacetate decarboxylase, which allows growth with ci-trate as the sole carbon and energy source (Fig. 1). In Esche-richia coli, the cit cluster includes the genes encoding citratelyase and the CitT citrate/succinate antiporter (22) (Fig. 1). Inthis bacterium the citrate fermentation is dependent on thepresence of an oxidable cosubstrate, due to the lack of genesencoding an oxaloacetate decarboxylase activity. Thus, citrateis converted via malate and fumarate to succinate, and thereducing equivalents required for this conversion are providedby the oxidation of glucose or glycerol (14). In both organisms,the citrate fermentation (cit) clusters are regulated by a sensorkinase, CitA, and a response regulator, CitB (4).

Like in K. pneumoniae, in lactic acid bacteria the pathwayinvolved in the conversion of citrate to pyruvate requires threespecific enzymatic activities: a citrate permease, a citrate lyase,and an oxaloacetate decarboxylase (9). In several strains of

Lactococcus lactis, the ability to transport citrate has beenassociated with an 8.0-kb family of plasmids, which carry thecitQRP operon, whereas in Leuconostoc spp., the gene involvedin citrate transport is linked to 23-kb plasmids (9, 19, 27). Wehave recently reported that in Weissella paramesenteroides J1(previously named Leuconostoc paramesenteroides), the citPgene, encoding the citrate permease, is carried on plasmidpCitJ1 (19, 20). This gene is part of the citMCDEFGRP oper-on, which also encodes the �, �, and � subunits of the citratelyase complex (citF, citE, and citD genes, respectively) and thecomplementary activities required for the biosynthesis and ac-tivation of the prosthetic group (citG and citC genes) (19, 20)(Fig. 1). The expression of the plasmid-carried citMCDEFGRPoperon in W. paramesenteroides J1 is induced when the cells aregrown in the presence of citrate, and this induction depends onthe transcriptional activator CitI (19, 20). Analogous organiza-tion and regulation have also been shown for the chromosomalcit cluster from Leuconostoc mesenteroides 195 (2) (Fig. 1).

In contrast to the case for the citrate-regulated genes fromWeissella and Leuconostoc, the expression of the L. lactis citPgene is not influenced by the presence of citrate in the growthmedium (15). Instead, the expression of this gene is induced ata transcriptional level by acidification of the medium (8). Wehave previously reported that in L. lactis the citrate fermenta-tion pathway has an important physiological role, allowing cellsto improve the cometabolism of glucose and citrate at low pHand to detoxify the lactate accumulated at the end of theexponential growth phase (17).

Although citrate lyase catalyzes the first committed step incitrate metabolism, little is known about the regulation of thegenes encoding this important enzyme in L. lactis. In this

* Corresponding author. Mailing address: Departamento de Micro-biología, Suipacha 531, S2002LRK Rosario, Argentina. Phone: 54-341-435 0661. Fax: 54-341-439 0465. E-mail: [email protected].

5649

on Novem

ber 22, 2015 by guesthttp://jb.asm

.org/D

ownloaded from

http://jb.asm.org/

paper we describe the identification of a chromosomal citM-citCDEFXG operon of L. lactis CRL264 that contains thegenes encoding the three subunits of the citrate lyase and thecitM gene, encoding a protein that is highly homologous tomalic enzymes. We also show that the expression of the citM-citCDEFXG operon as well as the citrate lyase activity is in-creased when cells are grown under acidic pH conditions.These results suggest that in L. lactis the pH-controlled tran-scriptional regulation of the citrate fermentation pathway hasevolved as a mechanism of resistance to acidic pH conditions.In addition, we provide information suggesting that CitI couldbe involved in the regulation of the citM-citCDEFXG andcitQRP operons in L. lactis CRL264.

MATERIALS AND METHODS

Bacterial strains, plasmids, and growth conditions. The bacterial strains andplasmids used in this work are listed in Table 1. L. lactis strains were grown at30°C in a pH-controlled fermentor in M17 broth containing 0.5% (wt/vol) glu-cose (M17G) at pH 7.0 or 5.0. The fermentor (Bioflo110 fermentor bioreactor;New Brunswick Scientific). The pH was continuously monitored and kept con-stant with 1 M NaOH solution. Alternatively, cells of L. lactis were grown inbatch culture at 30°C without shaking in M17G adjusted to pH 7.0 or pH 5.0 withHCl (8). To analyze the induction of the cit operon expression by citrate, M17Gmedium was supplemented with 1% sodium citrate (M17GC). Cells of L. lactiswere grown under stress conditions (300 mM NaCl, 1 mM H2O2, or 37°C) inM17G at an initial pH of 7.0. E. coli was grown aerobically at 37°C in Luria-Bertani medium (24). Erythromycin (1 �g/ml, for L. lactis) or ampicillin (100�g/ml, for E. coli) was added to the medium when necessary.

Construction of plasmids. In order to clone the citrate lyase genes, degenerateoligonucleotide primers (CL1 and CL2 [Table 1]) designed from the amino acidsequences of the � and � subunits of the citrate lyase of L. mesenteroides (2) wereused for PCR amplification from genomic DNA from L. lactis CRL264. The1.2-kb amplification product containing the 5� ends of the citE and citF genes waspurified, digested with BglII, and cloned into the BamHI site of pUC19 vector togive the plasmid pL1 (Fig. 2; Table 1). This fragment was used as a probe in aSouthern blot experiment, and a 5.3-kb HindIII fragment of chromosomal DNAwas identified and cloned into pBScSK(�) to give the pL2 plasmid (Fig. 2; Table1). With the 298-bp HindIII-EcoRI fragment from pL2 as a probe, a 2.5-kbEcoRI fragment of genomic DNA was identified in Southern blot experimentand cloned into pUC18 to give pL5 (Fig. 2; Table 1). E. coli harboring the pL2

and pL5 plasmids was screened by colony hybridization with the 1.2-kb PCRproduct from pL1 and the 298-bp HindIII-EcoRI fragment from pL2, respec-tively, as probes. When the L. lactis IL-1403 genome sequence was available (3),we used the annotated sequence to design oligonucleotide primers for cloningthe remaining cit region of L. lactis CRL264. The pGEMT264 plasmid wasconstructed by cloning a 1.4-kb fragment including the citM gene and 3� adjacentregion into pGEM-T Easy vector (Fig. 2; Table 1). This fragment was obtainedby PCR amplification with the MAE264U (nucleotides [nt] �24 to �56 from thetranscriptional start site) and MAE264L2 (complementary to nt �1443 to �1420from the transcriptional start site) (Table 1) primers. To construct the pHPB21plasmid, a 2.7-kb DNA fragment containing the 5� end of the citM gene and the2.5-kb upstream region of that gene was amplified by using the primersPCIT264O (nt �2478 to �2451 from the transcriptional start site) and P264L(complementary to nt �220 to �194 from the transcriptional start site) (Table1). This product was purified, digested with HindIII and BamHI, and cloned inthe pAK80 vector (10). The 570-bp PCR product obtained by using the oligo-nucleotides P264U2 (nt �348 to �316 from the transcriptional start site) andP264L was digested with HindIII and BamHI and cloned in the pAK80 vector toconstruct the pHPB22 plasmid. The 864-bp amplification product containing theintergenic region between citC and citI was obtained employing the RegIU andRegCL primers (Fig. 2; Table 1). The DNA product was digested with PstI andcloned into the same site of the plasmid pAK80. Both orientations of the putativepromoter regions, PcitI and PcitC, were isolated by restriction pattern, givingplasmids pHPB41 and pHPB42, respectively. In all cases, standard protocolswere used for PCR. The reaction mixture contained 10 mM Tris-HCl (pH 8.3),50 mM KCl, 2 mM MgCl2, a 25 mM concentration of each of the four de-oxynucleoside triphosphates, 2 U of Deep Vent DNA polymerase (New EnglandBiolabs), 20 pmol of each primer, and 50 ng of DNA in a final volume of 50 �l.Typically, samples were subjected to 30 cycles of denaturation (94°C, 1 min),annealing (54°C, 1 min), and extension (72°C, 1 min). An additional round ofamplification (72°C, 20 min) in the presence of 1 U of Taq DNA polymerase(Promega) and 25 mM dATP was performed in the case of the 1.4-kb fragmentcontaining the citM gene in order to enhance pGEM-T Easy cloning efficiency(plasmid pGEMT264).

To construct the pHPB11 plasmid, the 3.6-kb EcoRI-BglII fragment frompFS21 (27) containing the citQRP promoters was purified and cloned into thepAK80 vector (Table 1).

DNA analysis and manipulation. Plasmid DNA from E. coli cells was preparedwith a Wizard Plus Minipreps DNA purification system (Promega). PlasmidDNA extractions from lactic acid bacteria were done as described by O’Sullivanand Klaenhammer (21). Treatment of DNA with restriction-modification en-zymes was performed as recommended by the suppliers. L. lactis transformation

FIG. 1. Genetic organization of cit genes involved in citrate utilization in K. pneumoniae (4), E. coli (22), L. mesenteroides (2), W. parames-enteroides (20), and L. lactis IL-1403 (The Institute for Genome Research). The shaded box indicates that the organization of the citDEF genesis highly conserved in the different cit clusters.

5650 MARTIN ET AL. J. BACTERIOL.

on Novem


.org/D

ownloaded from

http://jb.asm.org/

was performed by electroporation according to the procedure of Dornan andCollins (7). E. coli cells were transformed by the standard CaCl2 procedure (24).

The complete DNA sequence of the citrate cluster was determined withautomated DNA sequencing instrumentation at the University of Maine DNAsequencing facility from pL1, pL2, pL5, pGEMT264, pHPB41, pHPB42, pHPB21,and pHPB22 (Fig. 2; Table 1).

�-galactosidase assay. The �-galactosidase activity was measured as describedby Israelsen et al. (10).

Citrate lyase activity assay. To determine citrate lyase activity, lactococcalcultures were grown in M17G at either pH 7.0 or 5.0 to an absorbance of 0.3 at660 nm. Cells were harvested and resuspended in ice-cold 100 mM phosphatebuffer (pH 7.2) supplemented with 3 mM MgCl2 and 1 mM phenylmethylsulfonylfluoride protease inhibitor (Sigma). Total protein extracts were prepared bypassing cells three times through a French pressure cell at 10,000 lb/in2, and celldebris was removed by centrifugation at 20,000 � g for 25 min. Citrate lyaseactivity was determined from samples of these extracts containing 100 or 50 �gof total soluble proteins at 25°C in a coupled spectrophotometric assay withmalate and lactate dehydrogenases as described by Bekal-Si Ali et al. (2). Oneunit of enzyme activity is defined as 1 pmol of citrate converted to acetate andoxaloacetate per min per �g of total protein.

RNA isolation and analysis. For Northern blot and primer extension analysis,RNA was isolated by the method described by Raya et al. (23). The RNAs werechecked for their integrity and yield of the rRNAs in all samples. The patterns ofthe rRNAs were similar in all preparations. Total RNA concentration was de-termined by UV spectrophotometry and by gel quantification with Gel Doc 1000(Bio-Rad). Primer extension analysis was performed as previously described(13). The primer used for detection of the start sites of the cit operon and citIwere pMM264 and RI264U, respectively (Table 1). One picomole of the primerwas annealed to 15 �g of RNA. Primer extension reactions were performed by

incubation of the annealing mixture with 20 U of Moloney murine leukemia virusreverse transcriptase (Promega) at 42°C for 60 min. Determination of the sizesof the reaction products was carried out in 6% polyacrylamide gels containing 8M urea. Extension products were detected by autoradiography on Kodak X-Omat S film.

For Northern blot analysis, samples containing 10 �g of total RNA wereseparated in a 1% agarose gel. Transfer of nucleic acids to nitrocellulose mem-branes and hybridization with radioactive probes were performed as previouslydescribed (13). The single-stranded probes used were synthesized as follows. The1,170-bp BamHI fragment from pGEMT264 containing the citM gene and the1,200-bp SalI fragment from pL2 containing the 3� end of citE and the 5� regionfrom citF were gel purified and �-32P labeled in one strand by using a Sequenasekit (Promega). Primer MAE264L2 and T3 reverse primer were then used to giveprobes I and II, respectively (Fig. 3; Table 1). The 1,086-bp fragment containingthe citI gene was obtained by PCR amplification with CitI264U and RI264L(Table 1). The fragment was purified and �-32P labeled in one strand by using theCitI264U oligonucleotide to give probe III (see Fig. 7). The reaction mixtureincluded 0.02 pmol of DNA; 0.3 pmol of oligonucleotide; 0.3 mM (each) dCTP,dGTP, and dTTP; 0.7 �M unlabeled dATP; and 1 �l of [�-32P]dATP. mRNA mo-lecular sizes were estimated by using 0.28- to 6.58-kb RNA markers (Promega).

Nucleotide sequence accession number. The nucleotide sequence of L. lactisCRL264 that contains the cit region has been deposited at the GenBank databaseunder accession no. AY268077.

RESULTS

Analysis of the chromosomal region containing the citM-citCDEFXG operon. We determined the complete nucleotide

TABLE 1. Bacterial strains, plasmids, and oligonucleotides used in this work

Strain, plasmid, oroligonucleotide Characteristics, genotype, or sequencea Reference

L. lactis subsp. lactisstrains

CRL264 Lac� Pro� Cit�; harboring pCit264 27IL-1403 Trp�; plasmid free 6

PlasmidspL1 1.2-kb PCR amplification fragment containing citE and 5� end of citF from L. lactis CRL264 cloned in pUC19

vectorThis work

pL2 5.3-kb HindIII fragment containing citI, citCDE, and 5� end of citF cloned in pBScSK(�) vector This workpL5 2.5-kb EcoRI fragment containing citX, citG, and 3� end of citF cloned in the pUC18 vector This workpGEMT264 1.4-kb PCR amplification product corresponding to citM gene cloned in pGEM-T Easy vector This workpHPB21 pAK80-derived plasmid containing the lacLM genes under control of the 2.5-kb cit promoter region from

L. lactis CRL264This work

pHPB22 pAK80-derived plasmid containing the lacLM genes under control of the 0.4-kb cit promoter region fromL. lactis CRL264

This work

pHPB11 pAK80-derived plasmid containing the lacLM genes under control of the P1 and P2 promoters of the citQRPoperon from L. lactis CRL264

This work

pHPB41 pAK80-derived plasmid containing the lacLM genes under control of the PcitI promoter of the citI gene fromL. lactis CRL264

This work

pHPB42 pAK80-derived plasmid containing the lacLM genes under control of the putative PcitC from L. lactis CRL264 This workpFS21 8-kb pCit264 plasmid cloned in the EcoRI site of pUC18 vector 27pAK80 Promoter selection vector containing the promotorless L. mesenteroides �-galactosidase (lacL and lacM) genes 10

OligonucleotidesCL1 5�-TTTAGATCTAACNATGATGTTYGTNCCNGG-3� This workCL2 5�-ACCAGATCTAGTDATRTTNGTNACNACNCC-3� This workMAE264U 5�-AGGAAAGAGGATCCATGGTTGATTTTAATAAAG-3� This workMAE264L2 5�-TCCAAGCTTTGAAAAACTTGCTGG-3� This workPCIT264O 5�-TAATTAAGCTTCCTTTGGACTATTCTG-3� This workP264U2 5�-ATGTAGAGAAGCTTCAAAAAAATAATGCACTCC-3� This workP264L 5�-TCGGATCCTTGTGTATTTGCGTTTAGC-3� This workRegIU 5�-CCACTGCAGATCAAATCCATTC-3� This workRegCL 5�-GCAAGTCTGCAGGTATTTTAT This workCitI264U 5�-AACTGCAGCTTTCTATGTTACAACTCTGC-3� This workRI264L 5�-AAACATTCAAATGCAAATCC-3� This workRI264U 5�-AAATCTTGTGCCAAATTGG-3� This workpMM264 5�-AATATCCAATACACCTGTGTG-3� This work

a The restriction sites included in the oligonucleotides are underlined.

VOL. 186, 2004 L. LACTIS CITRATE LYASE OPERON TRANSCRIPTION 5651

on Novem


.org/D

ownloaded from

http://jb.asm.org/

sequence of a 11.4-kb region comprising the cit loci present inL. lactis strain CRL264 and found that the cluster containingthe genes coding for citrate lyase possessed 99% of identitywith the corresponding genes of L. lactis IL-1403 (3) (seeMaterials and Methods) (Fig. 1 and 2). Examination of thecomplete nucleotide sequence of the citrate cluster from L.lactis CRL264 revealed the presence of eight open readingframes (ORFs). On the basis of their homology with the pre-viously characterized citrate lyase genes from enterobacteria(4, 26), six of these ORFs were identified as citC, citD, citE,citF, citX, and citG (Fig. 2). The citC gene encodes a protein of346 amino acids that shows high homology to the acetate:SH-citrate lyase ligase, which is involved in the activation of theprosthetic group of the citrate lyase complex. The start codonof the citD gene is located 4 bp upstream of the stop codon ofcitC and encodes a protein of 96 amino acids (11.5 kDa). ThecitE gene, which starts 1 bp downstream of the stop codon ofcitD, encodes a protein of 304 amino acids (36.5 kDa). Thestart codon of the citF gene is located 23 bp upstream of thestop codon of citE, and from this DNA sequence we deduceda product of 518 amino acids (62.2 kDa). The citD, citE, andcitF genes encode the three citrate lyase subunits, � (acyl car-rier protein [ACP]), � (citryl-ACP oxaloacetate lyase), and �(acetyl-ACP:citrate ACP-transferase), respectively. High ho-mology was found between the structural subunit �, �, and �proteins and the corresponding proteins present in others cit-rate-fermenting microorganisms. The citX gene starts 56 bpdownstream from the stop codon of citF, and the inferredprotein has 172 amino acids (20.6 kDa). Finally, the citG geneoverlaps the citX coding region by 8 nt. The sequence derivedfrom citG was 274 amino acids long (32.9 kDa). These genesencode the enzymes needed for the synthesis of the prostheticgroup of the citrate lyase (CitX [apo-citrate lyase phosphori-bosyl-dephospho-coenzyme A transferase] and CitG [triphos-phoribosyl-dephospho-coenzyme A synthase]).

Preceding the first ATG of citC, citD, citE, citF, citX, andcitG, putative ribosomal binding sites complementary to the 3�

end of the 16S rRNA from L. lactis were observed. The over-lapping found between the citC-citD, citD-citE, citE-citF, andcitX-citG genes suggests the existence of translational couplingfor these genes.

citM should encode a 40.5-kDa polypeptide. This proteinshowed homology with malic enzyme, which is associated withthe cit cluster in other lactic acid bacteria (Fig. 1). Downstreamfrom citM and on the complementary strand we found an ORF(citI) encoding a protein that shows high homology to severaltranscriptional regulators. Among them, the proteins that hadthe highest homology were the cit operon activator CitI fromW. paramesenteroides (20), the ClyR putative regulator from L.mesenteroides (2), and a putative CitI protein from Clostridiumperfringens. Between the citM and citI genes we found a pseu-dogene (citO) which shows several frameshift mutations (Fig.2). The analysis of its sequence reveals homology to the malatepermease from Oenococcus oeni (GI:5870596) and Clostridi-uim cellulovorans (GI:7363467). However, the frameshift mu-tations that accumulated during evolution indicate that citO isa nonfunctional gene. It is interesting that a full copy of theinsertion element IS983 was found adjacent to the cit cluster(Fig. 2), which is a distinctive feature of L. lactis CRL264compared with the cit cluster of L. lactis IL-1403.

The activity of citrate lyase from L. lactis CRL264 is inducedby acid stress. Previous reports indicate that in several bacteriathe activity of citrate lyase is induced by the presence of citratein the growth medium (2, 4, 20). To test whether this inductionalso takes place in L. lactis CRL264, citrate lyase activity in cellextracts from cultures grown in M17G or M17GC at pH 7.0was analyzed. The citrate lyase activities observed in culturesgrown in the absence or presence of citrate were 0.19 0.04and 0.22 0.06 U/min � �g of protein, respectively. Theseresults indicate that citrate is not an inducer of citrate lyase inL. lactis. Taking into account previous results demonstrating ahigher citrate consumption in L. lactis cells grown under acidicconditions (17), we decided to analyze the citrate lyase activityin extracts of cells grown at neutral or acidic pH. The citrate

FIG. 2. Cloning of the cit locus of L. lactis CRL264. A schematic representation of the lactococcal inserts present in the recombinant plasmidspL1, pL2, pL5, pHPB21, pHPB22, pHPB41, pHPB42, and pGEMT264 is shown. Solid bars indicate chromosomal fragments identified by colonyhybridization (pL2 and pL5); open bars indicate DNA fragments obtained by PCR amplification (see details in Materials and Methods andTable 1).


on Novem


.org/D

ownloaded from

http://jb.asm.org/

lyase activities were 0.19 0.04 and 1.22 0.06 U/min � �g ofprotein in extracts of cells grown at pH 7.0 or 5.0, respectively.Therefore, the citrate lyase activity levels were increased aboutsixfold at acidic pH, as was also demonstrated for the activityof the plasmid-encoded citrate transporter CitP (8, 17).

Transcriptional analysis of the citrate cluster of L. lactisCRL264. To study the transcriptional pattern of the cit clusterand to test whether its transcriptional activity is regulated by

external pH, a Northern blot analysis was performed. Totalcellular RNA was isolated from cultures of L. lactis CRL264grown in M17G at pH 7.0 or pH 5.0 in the presence or absenceof citrate. The RNA was hybridized with two �-32P-labeledsingle-stranded DNA probes (Fig. 3). Probe I includes a frag-ment of 1,170 nt covering the citM gene, and probe II corre-sponds to a 1,200-nt fragment including the 3� end of citE andthe 5� region of citF. Probe I revealed the presence of an 8.6-kb

FIG. 3. Organization of the cit operon in L. lactis CRL264 (A) and Northern blot analysis of the cit operon (B). (A) The 11.4-kb DNA clusterencompassing eight genes involved in citrate utilization is shown at the top. Pcit, promoter of the cit operon. The secondary structure downstreamfrom citG represents a putative Rho-independent transcriptional terminator. Probe I includes a 1.2-kb fragment of citM. Probe II includes a 1.0-kbfragment of citEF. (B) Northern blot analysis was carried out as described in Materials and Methods. Strain CRL264 was grown at pH 7.0 or 5.0in the presence or absence of citrate. mRNA molecular marker are showed on the right of each autoradiograph.


on Novem


.org/D

ownloaded from

http://jb.asm.org/

mRNA and smaller RNA species. With probe II we observedat least six transcripts, the largest one having a size of 8.6 kb(Fig. 3). Considering the size of the cit gene cluster, the largestRNA specie detected (8.6 kb) with the two probes would cor-respond to an operon starting upstream from citM and endingdownstream from citG (Fig. 3). The 3� extreme of this tran-script contains an inverted repeated sequence (cAGUucagACUGGGGAGccacuCUCCUCaAGUacGCUUUUU [underliningshows nucleotides involved in the complementary interaction;lowercase letters represent internal loops]; G°, �10.1 kcal/mol)(28), which could act as a Rho-independent terminator (Fig. 3).The transcriptional pattern of the cit operon shown in Fig. 3suggests that the 8.6-kb transcript could be subject to specificprocessing. Although below we demonstrate the absence of pro-moters upstream of citC, we cannot discard possibility of theexistence of alternative promoters in other regions of the citoperon.

As shown in Fig. 3, the transcription of the cit operon wasdramatically increased when the cells were grown at acidic pH.It is worth noting that the presence of 1% citrate in the growthmedium did not significantly affect transcription of the cit op-eron compared with the induction produced by the acidifica-tion of the medium (Fig. 3). These results indicate that, simi-larly to the plasmid citP gene of L. lactis CRL264, transcriptionof the chromosomal cit cluster is induced by low pH (8).

Analysis of the promoter region and determination of thetranscriptional start site of the citM-citCDEFXG operon. Todetermine if the synthesis of the 8.6-kb transcript is driven bya putative promoter located upstream from citM, the transcrip-tional start site of citM-citCDEFXG was determined by primerextension. Total RNA was extracted from L. lactis CRL264cells grown in M17G at pH 7.0 or 5.0, and a primer extensionassay was performed as described in Materials and Methods.As shown in Fig. 4A, the transcript starts with a guanosine

FIG. 4. Identification of the transcriptional start site of the cit operon from L. lactis CRL264. (A) The autoradiograph shows primer extensionexperiments performed with total RNA extracted from cells grown in M17G medium buffered at pH 7.0 or 5.0. Lanes A, C, G, and T, sequencingladders. (B) Nucleotide sequences of the chromosomal citM-citCDEFXG operon promoter region (Pcit) and the plasmid citQRP promoter region(P1) from L. lactis CRL264. �10 and �35 promoter elements are shaded in gray. Consensus promoter sequences recognized by the L. lactis �39

transcription factor are indicated over these boxes. The bent arrow indicates the transcriptional start site of the citM-citCDEFXG transcript at theresidue in boldface, defined as �1. Identity between the promoter regions is indicated with asterisks. Thin arrows indicate putative binding sitesthat are conserved in both regions.


on Novem


.org/D

ownloaded from

http://jb.asm.org/

residue located 37 bp upstream from the ATG codon of citM,and, as expected, the levels of the extended products werehigher in RNA preparations from cultures grown at pH 5.0than in those from cultures grown at pH 7.0. Analysis of thisregion allowed us to identify a standard promoter sequencewith consensus �35 TTGACA and �10 TATAAC boxes (Fig.4B). The presence of these �35 and �10 sequences suggeststhat this promoter could be recognized by the L. lactis �39

transcription factor (1). The region upstream from the �35box (position �35 to �94) contains an unusually high A�Tcontent (near 95% A�T), which may contribute to the activityof the cit promoter (Pcit promoter) due to the intrinsic curva-ture of these A�T-rich sequences (12).

In Fig. 4B we compare the two pH-controlled promoterregions from L. lactis strain CRL264: the chromosomal Pcitregion and the previously described plasmid promoter P1,present in the citQRP operon (8). This alignment shows severalA�T-rich stretches at the same positions with respect to the�35 box in both promoters, which may be involved in theregulation of these operons by external pH.

Expression of the cit operon from L. lactis is enhanced underacidic conditions. L. lactis IL-1403 harboring plasmid pHPB22or pHPB11 was used to analyze the activity of the pH-regu-lated promoters Pcit and P1. Plasmids pHPB22 and pHPB11,derivatives of the pAK80 vector, contain a fusion of the lacLMreporter genes from Leuconostoc to the Pcit and P1 promoters,respectively. The �-galactosidase activities driven by each lac-tococcal promoter region were analyzed as a function of theexternal pH. To this end, cells were grown in batch culture inM17G medium at an initial pH of 5.0 or 7.0. Samples weretaken at different growth stages, and the external pH wasdetermined. As shown in Fig. 5, the �-galactosidase levels ofboth fusions increased about fivefold as the external pH de-creased from 6.5 to 4.4. In order to analyze the influence of the

growth phase of the cultures on the transcriptional activities ofboth promoters, we performed similar experiments in a pH-controlled fermentor. To this end, the cells were cultured at afixed pH of 7.0 or 5.0, and samples were taken at differentstages of growth until the cultures reached stationary phase.The �-galactosidase activities measured in L. lactis IL-1403(pHPB11) at a fixed pH of 5.0 at the early exponential (t1),exponential (t2), and stationary (t3) phases of growth were1.5 � 103, 1.4 � 103, and 1.6 � 103 Miller units, respectively,while the �-galactosidase levels of this strain at pH 7.0 were330, 305, and 215 Miller units at t1, t2, and t3, respectively. The�-galactosidase activities of strain L. (pHPB22) growing at pH5.0 were 5.5 � 104 Miller units at t1, 6 � 104 Miller units at t2,and 5.6 � 104 Miller units at t3, while the �-galactosidase levelsof this strain at pH 7.0 were 1.16 � 104 Miller units at t1, 1.3 �104 Miller units at t2, and 1.35 � 104 Miller units at t3. Thesedata confirm that the transcriptional induction driven from bothpromoters occurs at low pH independently of the growth phase.

To investigate whether the response of the Pcit and P1promoters is specific to the acidic stress, we tested the ability ofseveral environmental stresses to induce the expression of thecit-lacLM fusions in L. lactis. We assayed the �-galactosidaseactivities of strains L. lactis IL-1403(pHPB22) and L. lactisIL-1403(pHPB11) upon exposure to NaCl, H2O2, and heatstress in a medium with an initial pH of 7.0. As shown in Fig.6 none of these stress conditions induced the expression of thePcit and P1 promoters at pH 7.0. Further, these conditions didnot abolish the acid stress induction of the transcriptionalfusions (data not shown). These results confirmed the speci-ficity of the acid stress to induce transcription of two keyoperons involved in citrate fermentation in L. lactis CRL264.

To investigate whether the insertion sequence IS983, locatedupstream from the citM-citCDEFXG operon, had any influ-ence on its transcription, we compared the �-galactosidase

FIG. 5. Influence of a shift in the external pH on expression of cit-lacLM fusions. L. lactis IL-1403 containing pHPB22 (A) or pHPB11 (B) wasgrown in M17G adjusted to an initial pH of 7.0 or 5.0. Samples were taken at different growth stages, and the external pHs are indicated on thebottom. �-Galactosidase was determined as described in Materials and Methods. Each bar shows the average and standard deviation of the valuesfrom at least three experiments.


on Novem


.org/D

ownloaded from

http://jb.asm.org/

activities from cells transformed with plasmids pHPB21 (con-taining IS983) and pHPB22 (devoid of IS983) (Fig. 2 andTable 1). No difference in the �-galactosidase activities couldbe found between these two fusions when the cells were grownin M17G or M17GC under neutral or acidic conditions (datanot shown). Thus, IS983 does not seem to play any role incitM-citCDEFXG transcription, as is the case for IS982 in plas-mid pCit264 of L. lactis CRL264 (16).

The citI gene is transcribed as a monocistronic mRNA. Asshown in Fig. 3, the L. lactis cit operon contains citI, whichdisplays high homology with the previously described citI genefrom W. paramesenteroides strain J1, coding for a transcrip-tional regulator of the citMCDEFGRP operon (19, 20). How-ever, in L. lactis, citI is oriented reverse to the cit operon (Fig.1 and 2). To investigate the transcriptional pattern of expres-sion of citI in L. lactis, total cellular RNA was isolated fromcultures of strain CRL264 grown in M17G at pH 7.0 or 5.0.The RNA was hybridized with a single-stranded [�-32P]DNA-labeled probe corresponding to citI. As shown in Fig. 7A, a1-kb citI mRNA could be visualized in the Northern blot ex-periment. The amount of mRNA detected by this techniquewas larger in RNA preparations from cultures grown at pH 7.0than in those from cultures grown at pH 5.0. Thus, in contrastto the expression of the cit operon, this gene seems not to beinduced by acidic pH. However, as found for the cit operon,expression of the citI gene was not influenced by addition ofcitrate to the growth medium (data not shown). The transcrip-tional start site of citI was determined by primer extensionexperiments from total RNA of cells of L. lactis grown inM17G at pH 7.0. The transcript starts with an adenosine res-

idue located 117 bp upstream from the ATG of citI. Thepromoter is situated upstream of this sequence, with �35(TTatCc) and �10 (TAacAT) motifs separated by a shortsequence of 17 nt. These hexamers deviate three and tworesidues, respectively, from lactococcal consensus sequences,suggesting that the PcitI promoter could be independent of�39, the major � sigma factor of L. lactis (1) (Fig. 7B). Aputative Rho-independent terminator sequence was found inthe 3� region of citI (cauaGAAAGAGGCauuuaaGCCUUUUUUUUcuguuau; G°, �10.6 kcal/mol) (28). To test whetherthere are divergent promoters in the intergenic fragment lo-cated between citC and citI, we cloned this fragment in thedirection of either citI or citC transcription into the pAK80promoter probe vector, giving plasmids pHPB41 and pHPB42,respectively. The resulting plasmids contain a fusion of thelacLM reporter gene to the PcitI promoter (pHPB41) or theputative PcitC promoter (pHPB42). The �-galactosidase activ-ities of strains bearing these plasmids were determined in cellsgrown in M17G medium at an initial pH of 7.0 or 5.0 (Fig. 7D).Surprisingly, the �-galactosidase activity driven by pHPB41was twofold higher in cells growing at pH 5.0 than in cellsgrowing at pH 7.0, showing that the PcitI promoter is inducedby acidic pH. These results are in conflict with the data ob-tained by Northern blotting (Fig. 7A), which indicated that thecitI transcript was not induced at low pH. These experimentssuggest that the levels of the citI transcript are controlled insome way at a posttranscriptional level. On the other hand,when L. lactis IL-1403 was transformed with plasmid pHPB42,carrying the PcitC-lacLM fusion, the levels of �-galactosidaseactivity were similar to that found in lactococcal cells trans-

FIG. 6. Influence of diverse environmental stresses on expression of cit-lacLM fusions. L. lactis IL-1403 transformed with pHPB22 (A) orpHPB11 (B) was grown to an A660 of 0.7 in M17G at pH 7.0 in the presence of NaCl (300 mM), in the presence of H2O2 (1 mM), or at 37°C, and�-galactosidase activity was determined as described in the legend to Fig. 5.


on Novem


.org/D

ownloaded from

http://jb.asm.org/

formed with the pAK80 vector (Fig. 7D), indicating the ab-sence of promoters in this region.

DISCUSSION

Three functions possibly implicated in pH homeostasis havebeen characterized in lactococci to date: (i) the H�-ATPase,(ii) the arginine deiminase pathway, and (iii) a glutamate de-carboxylase. The H�-ATPase expels protons from the cell viaATP hydrolysis. This activity increases as the pH decreases,and it has been demonstrated that it is essential for cell via-bility at low pH (11, 12). The arginine deiminase pathwayconverts arginine to ammonia, ornithine, and CO2. Ammoniaproduction may contribute to survival at low extracellular pHby neutralization of the medium (5, 18). Glutamate decarbox-ylase (GadB) converts glutamate to �-aminobutyrate and CO2

in an H�-consuming reaction contributing to pH homeostasis

(25). These three mechanisms may reduce acidification of theinternal compartment and thus could be important in survivalunder acidic conditions. However, regulation of the expressionof these pH-controlled stress response systems remains essen-tially unknown for lactococci. In this work, we have character-ized at a molecular level a new system involved in pH ho-meostasis in L. lactis, the citrate fermentation pathway.

We have identified an 11.4-kb chromosomal region in L.lactis CRL264 that includes eight genes involved in the citratefermentation pathway, organized in an operon which is tran-scribed as a single polycistronic mRNA of approximately 8.6 kb(the cit operon). Similar large transcripts encompassing all ofthe genes for citrate utilization were also reported for the citoperons from W. paramesenteroides (20), L. mesenteroides (2),and K. pneumoniae (4). Several distinct smaller mRNA specieswere detected in addition to the full-length transcript (Fig. 3).The abundance of the smaller species indicates that they are

FIG. 7. Analysis of the expression of citI in L. lactis CRL264. (A) Northern blot analysis of the citI gene. The probe (probe III) used in thisexperiment is indicated at the top of panel C. (B) Nucleotide sequence of the citC-citI intergenic region. �10 and �35 regions are indicated bygrey boxes; �1 represents the transcription initiation site (the oligonucleotide used in the primer extension experiment is underlined). The ATGcodon is in boldface. (C) Schematic representation of the pAK80-derived plasmids pHPB41 and pHPB42. Pcit and PcitI are the promoters of thecitM-citCDEFXG operon and citI, respectively. T, putative Rho-independent transcriptional terminator. (D) The �-galactosidase activities ofL. lactis IL-1403 cells bearing plasmids were determined as described in Materials and Methods. The initial pHs at which the cells were grown areindicated. Error bars indicate standard deviations.


on Novem


.org/D

ownloaded from

http://jb.asm.org/

more stable than the full-length transcript. A putative cit op-eron mRNA processing could explain the existence of the mi-nor species of mRNA, as described for the cit operons presentin other bacteria (2, 4, 20).

The transcriptional start point (tsp) of the citM-citCDEFXGoperon was identified upstream from citM by primer extensionanalysis. In this region we could find �35 (TTGACA) and �10(TATAAC) boxes, separated by 17 bp, with similarity to thesequences recognized by the L. lactis �39 transcriptional factor(1). In addition, sequence analysis of this promoter regionrevealed an extremely high A�T content, which may cooper-ate with the activity of Pcit promoter due to the intrinsic bend-ing of A�T sequences. However, no �10 extended motif, asdescribed for about half of the lactococcal promoters, could befound in this operon. The alignment of the nucleotide se-quences of the Pcit and P1 promoters allowed us to find directrepeats placed at the same positions from the �35 box (Fig.4B). These direct repeats could be the binding site of a regu-latory protein, suggesting a common regulatory mechanism forthese two operons.

To compare the transcriptional activities of Pcit and P1, the�-galactosidase activities shown in Fig. 5 were corrected on thebasis of the number of copies of each promoter in L. lactisCRL264. The number of copies of plasmid pCit264 in thisstrain was estimated to be about seven (10), while there is onlyone chromosomal copy of the citM-citCDEFXG operon in thisstrain. Thus, the transcriptional activity of the Pcit chromo-somal promoter seems to be about fivefold higher than that ofthe plasmid P1 promoter.

We have evidence to prove that both plasmid operon(citQRP) encoding the citrate transporter and the chromo-somal operon (citM-citCDEFXG) encoding the citrate lyasecomplex have similar responses to acid stress (8) (Fig. 3). Infact, the transcription of both operons is induced at acidic pHand it is not significantly influenced by the presence of citrate(Fig. 3). In agreement with this observation, the citrate lyaseactivity is increased in extracts obtained from cells grown at pH5.0 compared with cells grown at pH 7.0. We have also shownthat exposure to different stress conditions failed to induce theexpression of the citM-citCDEFXG and citQRP operons. More-

FIG. 8. Response of the cit operons to acidic stress and its contribution to pH homeostasis in L. lactis CRL264. The fermentation of glucoseresults in accumulation of lactic acid and acidification of the medium. Under these conditions, P1 and Pcit are activated, resulting in increasedsynthesis of CitP and the citrate lyase complex. The CitP transporter actively and specifically removes lactate from the cytoplasm, relieving thestress imposed by this weak acid. In addition, the increased metabolization of citrate contributes to alkalinization of the cytoplasm by enhancingthe proton consumption. The regulator protein CitI could be involved in the regulation of both cit operons of L. lactis CRL264.


on Novem


.org/D

ownloaded from

http://jb.asm.org/

over, the induction is independent of the growth phase, asdemonstrated by experiments performed in a pH-controlledfermentor. Thus, these two key operons involved in citratefermentation in L. lactis CRL264 are specifically induced byacidic stress.

How could the transcription of the L. lactis cit operon beregulated by pH? In this report we have shown that the activityof the citI promoter, whose putative gene product is highlyhomologous to the citrate-regulated CitI transcriptional acti-vator from W. paramesenteroides, is increased twofold at acidicpH. Thus, it could be possible that when L. lactis cells detect adecrease in external pH, transcription of citI is induced, result-ing in an increase in the cellular levels of CitI. If this is the case,a small increase in citI transcription could account for a largerincrease in the cit mRNA detected in L. lactis under acidicconditions. The levels of citI then could be down regulatedthrough a not-yet-identified posttranscriptional mechanism,resulting in a decrease of the cit transcript. It is worth notingthat the 8.6-kb cit transcript includes a citI sequence comple-mentary to the citI mRNA. In fact both, the mRNA sensetranscript and the citI antisense transcript located in the citoperon were confirmed by reverse transcription-PCR experi-ments with primers CitI264U (Table 1) and CitI264L (5�-TAGGATCCTTATGAATAATCATGAACTTCTCG-3� [theBamHI site is underlined]), respectively (data not shown).Therefore, it is tempting to speculate that citI translation couldalso be down regulated by an antisense mechanism. Clearly,more experiments are necessary to elucidate the role, if any, ofCitI in transcriptional induction of the cit operon by acidic pH.

The scheme shown in Fig. 8 represents the contribution ofthe citrate fermentation to pH homeostasis in L. lactis. Thisbacterium produces lactic acid during glucose fermentation,implying that these cells are normally exposed to acid stress. Atlow pH, lactic acid (a weak organic acid) is not charged and caneasily pass through the cell membrane in the protonated form.Inside the cell, it dissociates to the lactate1� form, producing astrong stress to the cell. As described here, when the externalpH decreases to 5.0, lactococcal cells growing in M17G areable to sense the acidic stimulus and trigger the coordinatedexpression of the operons controlled by the P1 and Pcit pro-moters (Fig. 7). In the presence of citrate, the plasmid-encodedCitP permease is responsible for the specific removal of lactatefrom the cytosol (through the CitP citrateH2�/lactate1� anti-port) (12). Cytoplasmic (chromosomally encoded) componentsof the citrate fermentation pathway then contribute to pHhomeostasis, consuming scalar protons and generating a pH(17). The pyruvate produced for the citrate fermentation isused for the production of less acidic compounds such asdiacetyl and acetoin (Fig. 7). In this way L. lactis can survive atlow pH during cometabolism of glucose and citrate. Thus, wepropose that the pH-controlled expression of proteins requiredfor the exchange of citrate for lactate coordinated with decar-boxylation of citrate constitutes an important mechanism foracidic stress resistance in L. lactis.

ACKNOWLEDGMENTS

We thank Hans Israelsen for generously providing pAK80, MonicaArevalo and Alicia Bruzzone for technical assistance, and two anony-mous reviewers for their comments on a previous version of the manu-script.

This work was supported by grants from Fundacion Antorchas (Ar-gentina) and Agencia Nacional de Promocion Científica y Tecnologica(Argentina) (contract no. 01-09596-B and QLK12002-2388). M.G.M.and P.D.S. are fellows of CONICET (Argentina), and C.M. andD.D.M. are Career Investigators from the same institution.

REFERENCES

1. Araya, T., N. Ishibashi, S. Shimamura, K. Tanaka, and H. Takahashi. 1993.Genetic and molecular analysis of the rpoD gene from Lactococcus lactis.Biosci. Biotechnol. Biochem. 57:88–92.

2. Bekal-Si Ali, S., C. Divies, and H. Prevost. 1999. Genetic organization of thecitCDEF locus and identification of mae and clyR genes from Leuconostocmesenteroides. J. Bacteriol. 181:4411–4416.

3. Bolotin, A., P. Wincker, S. Mauger, O. Jaillon, K. Malarme, J. Weissenbach,S. D. Ehrlich, and A. Sorokin. 2001. The complete genome sequence of thelactic acid bacterium Lactococcus lactis ssp. lactis IL1403. Genome Res.11:731–753.

4. Bott, M. 1997. Anaerobic citrate metabolism and its regulation in enterobac-teria. Arch. Microbiol. 167:78–88.

5. Casiano-Colon, A., and R. E. Marquis. 1988. Role of the arginine deiminasesystem in protecting oral bacteria and an enzymatic basis for acid tolerance.Appl. Environ. Microbiol. 54:1318–1324.

6. Chopin, A., M. C. Chopin, A. Moillo-Batt, and P. Langella. 1984. Twoplasmid-determined restriction and modification systems in Streptococcuslactis. Plasmid 11:260–263.

7. Dornan, S., and M. A. Collins. 1990. High efficiency electroporation ofLactococcus lactis subsp. lactis LM0230 with plasmid pGB301. Lett. Appl.Microbiol. 11:62–64.

8. García-Quintans, N., C. Magni, D. de Mendoza, and P. Lopez. 1998. Thecitrate transport system of Lactococcus lactis subsp. lactis biovar diacetylactisis induced by acid stress. Appl. Environ. Microbiol. 64:850–857.

9. Hugenholtz, J. 1993. Citrate metabolism in lactic acid bacteria. FEMS Mi-crobiol. Rev. 12:165–178.

10. Israelsen, H., S. M. Madsen, A. Vrang, E. B. Hansen, and E. Johansen. 1995.Cloning and partial characterization of regulated promoters from Lactococ-cus lactis Tn917-lacZ integrants with the new promoter probe vector, pAK80.Appl. Environ. Microbiol. 61:2540–2547.

11. Kobayashi, H., T. Suzuki, N. Kinoshita, and T. Unemoto. 1984. Amplifica-tion of the Streptococcus faecalis proton-translocating ATPase by a decreasein cytoplasmic pH. J. Bacteriol. 158:1157–1160.

12. Koebmann, B. J., D. Nilsson, O. P. Kuipers, and P. R. Jensen. 2000. Themembrane-bound H�-ATPase complex is essential for growth of Lactococ-cus lactis. J. Bacteriol. 182:4738–4743.

13. Lopez de Felipe, F., C. Magni, D. de Mendoza, and P. Lopez. 1995. Citrateutilization gene cluster of the Lactococcus lactis biovar diacetylactis: organi-zation and regulation of expression. Mol. Gen. Genet. 246:590–599.

14. Lutgens, M., and G. Gottschalk. 1980. Why a co-substrate is required foranaerobic growth of Escherichia coli on citrate. J. Gen. Microbiol. 119:63–70.

15. Magni, C., F. Lopez de Felipe, F. Sesma, P. Lopez, and D. de Mendoza.1994. Citrate transport in Lactococcus lactis subsp. lactis biovar diacetylactis.Expression of the citrate permease P. FEMS Microbiol. Lett. 118:75–82.

16. Magni, C., F. L. de Felipe, P. Lopez, and D. de Mendoza. 1996. Characterizationof an insertion-like element identified in plasmid pCIT264 from Lactococcuslactis subsp. lactis biovar. diacetylactis. FEMS Microbiol. Lett. 136:289–295.

17. Magni, C., D. de Mendoza, W. N. Konings, and J. Lolkema. 1999. Mecha-nism of citrate metabolism in Lactococcus lactis: resistance against lactatetoxicity at low pH. J. Bacteriol. 181:1451–1457.

18. Marquis, R. E., G. R. Bender, D. R. Murray, and A. Wong. 1987. Argininedeiminase system and bacterial adaptation to acid environments. Appl. En-viron. Microbiol. 53:198–200.

19. Martín, M., M. A. Corrales, D. de Mendoza, P. Lopez, and Ch. Magni.1999. Cloning and molecular characterization of the citrate utilizationcitMCDEFGRP cluster of Leuconostoc paramesenteroides. FEMS Microbiol.Lett. 174:231–238.

20. Martín, M., C. Magni, P. Lopez, and D. de Mendoza. 2000. Transcriptionalcontrol of the citrate-inducible citMCDEFGRP operon, encoding genes in-volved in citrate fermentation in Leuconostoc paramesenteroides. J. Bacteriol.182:3904–3912.

21. O’Sullivan, D. J., and T. R. Klaenhammer. 1993. Rapid mini-prep isolationof high-quality plasmid DNA from Lactococcus and Lactobacillus spp. Appl.Environ. Microbiol. 59:2730–2733.

22. Pos, K. M., P. Dimroth, and M. Bott. 1998. The Escherichia coli citratecarrier CitT: a member of the novel eubacterial transporter family related tothe 2-oxoglutarate/malate translocator from spinach chloroplasts. J. Bacte-riol. 180:4160–4165.

23. Raya, R., J. Bardowski, P. S. Andersen, S. D. Ehrlich, and A. Chopin. 1998.Multiple transcriptional control of the Lactococcus lactis trp operon. J. Bac-teriol. 180:3174–3180.

24. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: alaboratory manual, 2nd ed. Cold Spring Harbor Laboratory Press, ColdSpring Harbor, N.Y.


on Novem


.org/D

ownloaded from

http://jb.asm.org/

25. Sanders, J. W., K. Leenhouts, J. Burghoorn, J. R. Brands, G. Venema, andJ. Kok. 1998. A chloride-inducible acid resistance mechanism in Lactococcuslactis and its regulation. Mol. Microbiol. 27:299–310.

26. Schneider, K., C. N. Kastner, M. Meyer, M. Wessel, P. Dimroth, and M.Bott. 2002. Identification of a gene cluster in Klebsiella pneumoniae whichincludes citX, a gene required for biosynthesis of the citrate lyase prostheticgroup. J. Bacteriol. 184:2439–2446.

27. Sesma, F., D. Gardiol, A. P. Ruiz Holgado, and D. de Mendoza. 1990.Cloning of the citrate permease gene of Lactococcus lactis subsp. lactis biovardiacetylactis and expression in Escherichia coli. Appl. Environ. Microbiol.56:2099–2103.

28. Zuker, M., and P. Stiegler. 1981. Optimal computer folding of large RNAsequences using thermodynamics and auxiliary information. Nucleic AcidsRes. 9:133–148.


on Novem


.org/D

ownloaded from

http://jb.asm.org/

Acid-Inducible Transcription of the Operon Encoding the Citrate Lyase Complex of Lactococcus lactis...

Documents

Transcript of Acid-Inducible Transcription of the Operon Encoding the Citrate Lyase Complex of Lactococcus lactis...