| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Center for Systems Microbiology, BioCentrum-DTU, Building 301, Technical University of Denmark, DK-2800 Kgs. Lyngby, Denmark
Correspondence
Jan Martinussen
jma{at}biocentrum.dtu.dk
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
), genes encoding proteinases necessary for casein degradation (Efstathiou & McKay, 1976; McKay et al., 1976) and di- and oligopeptide transport genes (Yu et al., 1995a, 1996).
Pyrimidine metabolism has been extensively studied in L. lactis in our laboratory (Andersen et al., 1996; Jorgensen et al., 2003, 2004; Kilstrup & Martinussen, 1998; Martinussen et al., 1994, 1995, 2001, 2003; Martinussen & Hammer, 1994, 1995, 1998; Wadskov-Hansen et al., 2000, 2001), and a review of nucleotide metabolism in lactic acid bacteria has recently been published (Kilstrup et al., 2005). In L. lactis, pyrimidines can be synthesized de novo or can be salvaged from the growth medium (Martinussen et al., 1994). The pyrimidine de novo synthesis pathway consists of six steps leading to the formation of UMP and seems to be universal among all organisms. L. lactis is also able to metabolize various pyrimidine nucleosides and nucleobases, but not cytosine, due to the lack of cytosine deaminase (Martinussen et al., 1994).
Orotate is an intermediate in the pyrimidine biosynthetic pathway. It is formed in the fourth step, in which dihydroorotate is converted into orotate by dihydroorotate dehydrogenase, encoded by the pyrD gene. A surprising discovery in L. lactis is the presence of two genes, pyrDa and pyrDb, both coding for a functional dihydroorotate dehydrogenase (Andersen et al., 1994, 1996) (Fig. 1). Later the same was reported in Enterococcus faecalis (Marcinkeviciene et al., 1999, 2000). Searching for homology revealed that Streptococcus spp. also harbour two pyrD genes, suggesting that this might be a characteristic of these closely related organisms. Milk contains a high concentration of orotate, which disappears during fermentation in the presence of lactic acid bacteria, including L. lactis cultures (Kneifel et al., 1992; Saidi & Warthesen, 1989). This finding suggests that at least some lactic acid bacteria can metabolize orotate. Growth of Lactobacillus bulgaricus is stimulated by orotate but no transporter has been identified yet (Suzuki et al., 1986). Pyrimidine-requiring mutants of both Escherichia coli and Salmonella typhimurium can utilize orotate as their sole pyrimidine source after transport into the cell mediated by the C4-dicarboxylic transport protein DctA (Baker et al., 1996). DctA also mediates the transport of orotate in Sinorhizobium meliloti (Yurgel & Kahn, 2005). However, no homology to the deduced amino acid sequence of dctA was found in the databases of published lactic acid bacteria genomes.
|
; Martinussen & Hammer, 1995). 5-Fluoropyrimidines are only toxic to the cell after being converted by the enzymes of pyrimidine metabolism. The most toxic metabolite is 5-fluorodeoxyuridine monophosphate (FdUMP, Fig. 1
), which prevents formation of dTMP by thymidylate synthase since it covalently binds to the enzyme. Hence, 5-fluoroorotate is transported and metabolized by the same enzymes that metabolize orotate. Our research focuses on the elucidation of orotate metabolism in L. lactis. This article reports the screening of L. lactis strains for their ability to utilize orotate, and the cloning and characterization of a gene, oroP, encoding an orotate transporter. In strains able to utilize orotate, the oroP gene was found on plasmids. Interestingly, homologues of oroP could be identified on the genomes of the L. lactis strains MG1363 and IL1403 despite the fact that these strains were unable to utilize orotate. oroP was found to be functional also in Escherichia coli and Bacillus subtilis. The results show that oroP can be exploited as an efficient selection/counterselection marker.
Plasmids found in lactococci replicate either via the sigma or rolling-circle replication mode, in which single-stranded DNA is produced as a replication intermediate (Hayes et al., 1990), or via the theta mode. Several theta-replicating plasmids can coexist in a single cell, explaining the many plasmids of different sizes found in lactococci. These plasmids are extremely stable, both structurally and segregationally, and exhibit a narrow host range (Frere et al., 1993; Hayes et al., 1991; Seegers et al., 1994).
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
) or in defined SA medium (Jensen & Hammer, 1993) supplemented with 1 % glucose (GSA). Strains DB0410, NCDO712, CHCC336, FHCY-1 and 3107 were propagated in the presence of lactose at 1 % (LSA) instead of glucose. E. coli strains were grown at 37 °C in Luria–Bertani (LB) broth or in minimal ABT glucose medium (Clark & Maaloe, 1967). B. subtilis strains were grown at 37 °C in LB broth or in minimal MM medium (Saxild & Nygaard, 1987). Agar plates were made by adding agar (15 g l–1) to the broth medium. Supplements were added to the different media at the following concentrations unless otherwise indicated: orotate at 20 µg ml–1, uracil at 20 µg ml–1, 5-fluoroorotate at 20 µg ml–1, erythromycin at 5 µg ml–1, chloramphenicol at 5 µg ml–1 for lactococci and 20 µg ml–1 for E. coli, ampicillin at 100 µg ml–1. Transformation of L. lactis was performed by electroporation (Holo & Nes, 1995). For selection of transformants of ED79.1175 by the ability to grow on orotate as sole pyrimidine source, cells were spread on agar plates containing SA medium, 1 % glucose, 250 mM sucrose, 2.5 µM CaCl2, 2.5 µM MgCl2 and 100 µg orotate ml–1. E. coli cells were transformed using the calcium chloride procedure (Mandel & Higa, 1992). Transformation of B. subtilis cells was carried out as previously described (Boylan et al., 1972): competent B. subtilis cells were obtained by adding 3 ml of an overnight culture (not lysed) grown in GM1 medium to 10 ml fresh GM1 medium and incubation for 4.5 h at 37 °C. The bacterial culture was subsequently diluted 1/10 in GM2 medium and incubated for 1 h at 37 °C. For 168MIU6 (trpC2 pyrB) cells, 20 µg uracil ml–1 was added to the GM1 and GM2 media. For the transformation, 1 ml of the competent culture was mixed with 10 or 30 µl of the linearized plasmid solution (100 ng µl–1) and the mixture was incubated for 35 min at 37 °C. Cells were centrifuged for 4 min and the cell pellet was resuspended in 200 µl 0.9 % NaCl solution and plated on selective LB agar plates containing 5 µg neomycin ml–1. The agar plates were incubated overnight at 37 °C.
Molecular cloning techniques.
Standard molecular cloning techniques were used according to Sambrook et al. (1989). Chromosomal DNA from lactococci was prepared as previously described (Johansen & Kibenich, 1992). Chromosomal DNA from B. subtilis was prepared according to Saxild & Nygaard (1987). Plasmid DNA was obtained by purification on Qiagen anion-exchange columns using the protocols supplied by the manufacturer. Plasmid DNA from lactococci was purified using the following modifications. Cells were resuspended in P1 buffer containing 20 mg lysozyme ml–1 followed by 20 min incubation at 37 °C. Purification on the Qiagen anion-exchange column was omitted. Isolation of plasmids from DB0410 and 901-1 was carried out by the adapted Qiagen protocol for the isolation of BAC and PAC clones, and of the pDBORO plasmid for sequencing purposes by the Qiagen very-low-copy plasmid purification protocol.
PCR amplification of DNA.
Amplification of DNA was performed on 50 ng DNA in a final volume of 50 µl containing 200 µM of each deoxyribonucleoside triphosphate, 200 nM of each oligonucleotide, and Taq polymerase. AmpliTaq DNA polymerase (2.5 U) was used in the verification experiments. For cloning purposes either Pfu DNA polymerase, TaqPlus Precision polymerase mixture or Elongase (Invitrogen) was used according the manufacturer's recommendations. Amplification was performed with 25–30 cycles of 95 °C 50 s, 50 °C 50 s and 72 °C 1–5 min.
Nucleotide sequences.
Nucleotide sequences were determined by MWG-Biotech. The sequence of the oroP region has been assigned the GenBank accession number DQ089807; the total pDBORO sequence can be accessed by the GenBank number EF210104.
Bacterial strains and plasmids.
The bacterial strains, plasmids and primers used in this work are shown in Tables 1 and 2. All plasmids were analysed by restriction enzyme analysis and sequencing; all recombinant strains were verified by PCR.
|
|
pyrDa mutant ED58.82 was made. To achieve this, plasmids pRL101, pED102 and pED202 were constructed. pRL101 containing pyrDa was constructed in the following way. A PCR fragment was amplified with primers pyrDaBamHI and pyrDaNcoI using pIP61 (Andersen et al., 1994) as template. Both pIP61 and the PCR fragment were digested with BamHI and NcoI, ligated together and transformed into E. coli XL1-Blue at 37 °C with selection for 100 µg ampicillin ml–1, resulting in pRL101. Plasmid pED102 was obtained in the following way. Using pRL101 as template and the primers pyrDaBamHI and pyrDaHindIII the pyrDa allele was amplified by PCR. The fragment was subcloned into the vector pCR2.1-TOPO using the TOPO-TA cloning system. Plasmid pED202 was constructed by ligating pED102 and pGhost+4 (Maguin et al., 1992) together at the XbaI site, which is unique in both plasmids. After transformation into E. coli XL1-Blue, the fusion plasmid was obtained by selection at 37 °C on LB agar plates containing both 100 µg ampicillin ml–1 and 150 µg erythromycin ml–1. Plasmid pED202 was transformed into L. lactis MG1363, selecting for erythromycin resistance at 28 °C. Using the method originally developed by Biswas et al. (1993) and improved by Martinussen & Hammer (1994), the pyrDa allele was exchanged with the wild-type, resulting in the mutant ED58.82.
In the second step, the pyrD double mutant ED79.1175 was created by deletion of a 780 bp internal fragment in the pyrDb gene. This was carried out in the following way. Two PCR fragments were amplified in separate reactions with MG1363 chromosomal DNA as template using pyrDbIF and pyrDbIR in one reaction and pyrDbIIF and pyrDbIIR in the other. Since the primers pyrDbIR and pyrDbIIF are partially complementary, the two PCR products could be used as templates in a new PCR reaction with primers pyrDbIF and pyrDbIIR. The resulting PCR fragment containing the pyrDb deletion was cloned into the vector pCR2.1-TOPO, yielding pED105. Plasmid pED204 was made by ligating pED105 and pGhost+4 together at the XbaI site and transforming into E. coli XL1-Blue as described above for pED202. The wild-type allele pyrDb was exchanged with pyrDb as described above for ED58.82 (pyrDa), thus obtaining the pyrD double mutant ED79.1175.
A 3.8 kb DNA fragment was amplified using pDBORO plasmid DNA and the primer set DBORO2/DBORO8. The PCR fragments were cloned into the vector pCR2.1-TOPO and recombinant plasmids were selected at 37 °C on LB agar plates containing 100 µg ampicillin ml–1. The recombinant plasmid pED112 harboured oroP and ysbBA.
An EcoRI fragment from pED112 was ligated into the unique EcoRI site of pCI3440 (Hayes et al., 1990) and amplified in E. coli XL1-Blue at 37 °C with selection for resistance to 20 µg chloramphenicol ml–1. The resulting plasmid was called pED210.
With pDBORO as template and primers DBORO18 and CS105, a PCR product was obtained and cloned into pCR2.1-TOPO. The recombinant plasmid pMBK701 was obtained after transformation into L. lactis ED79.1175 and selection for growth on GSA plates supplied with orotate as sole pyrimidine source.
A DNA fragment containing the oroP gene was amplified with primer set DBORO22BamHI/DBORO23EcoRI using pDBORO plasmid DNA as template. The PCR fragment was digested with BamHI and EcoRI and ligated into the B. subtilis integrative vector pDG268 (Antoniewski et al., 1990) digested with the same restriction endonucleases. The recombinant plasmid pED301 was amplified in E. coli XL1-Blue at 37 °C on LB agar plates containing 100 µg ampicillin ml–1. Recombinant plasmid pED307 was constructed in the same way except that primer set DBORO22BamHI/DBORO24EcoRI was used. Plasmid pED301 was integrated into the B. subtilis 168 amyE gene by a double cross-over recombination through the amyE N-terminal and amyE C-terminal parts located on the pDG268 vector. The plasmid was linearized by digestion with KpnI prior to transformation in order to ensure chromosomal integration by a double cross-over event. The neomycin-resistant strain ED348 was readily obtained. The control strain ED344 was obtained by transforming strain 168 with KpnI-linearized pGD268. The B. subtilis pyrimidine auxotrophic mutant 168MIU5 (trpC2 pyrB) (Potvin et al., 1975) was transformed with linearized pED307 DNA to obtain ED358 [pyrB (oroP-ysbB)+]. As a control strain 168MIU5 was also transformed with KpnI-linearized pGD268, resulting in ED364.
Chromosomal integration by a double cross-over event was identified by the lack of 
-amylase activity of the B. subtilis recombinants. Integration of pED301 or pED307 DNA into the amyE gene was investigated at the chromosomal level by PCR with primer sets 268neo/DBORO20 and 268neo/DBORO04.
Detection of -amylase activity in B. subtilis.
The -amylase phenotype of B. subtilis clones on LB agar plates containing 1 % starch was analysed by adding a solution of 0.5 % iodine and 1 % potassium iodine to the plates on which the strains had been growing overnight. The -amylase-producing colonies formed a clear halo, whereas the -amylase-negative colonies did not.
Growth experiments and orotate uptake.
Starting with fresh colonies from GSA plates with erythromycin, strains were grown for 6–8 h at 30 °C in 10 ml GSA medium with erythromycin. Different dilutions of the growing culture were added to 10 ml medium and grown overnight. From an exponentially growing culture, new medium was inoculated to OD436 0.05. The growth experiments were performed in plastic flasks without stirring. For uptake experiments, 500 µl cells at OD436 0.8 were transferred to an Eppendorf tube containing [14C]orotate at a specific activity of 52 mCi mmol–1 (1924 MBq mmol–1). After 1, 3 and 5 min the medium was removed by filtration through a 0.22 µm filter. The filter was washed twice with 5 ml water. After drying, the radioactivity on the filters was determined in a scintillation counter.
Phylogenetic analysis.
The GenBank program protein–protein BLAST at NCBI (http://www.ncbi.nlm.nih.gov/BLAST) was used to find similar proteins. Phylogenetic analysis was performed by a CLUSTAL W alignment used in a cluster analysis at the TreeTop - Phylogenetic Tree Prediction server (http://www.genebee.msu.su/services/phtree_reduced.html). The corresponding phylogenetic tree was obtained at the Phylodendron Phylogenetic tree printer (http://www.es.embnet.org/Doc/phylodendron/treeprint-sample1.html).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
). Surprisingly, MG1363 was found to be completely resistant. Moreover, the pyrimidine-requiring derivative ED79.1175 was only able to grow in the presence of uracil but not orotate. The uptake of uracil has previously been shown to be facilitated by a dedicated uracil transporter encoded by the chromosomally located pyrP (Martinussen et al., 2001).
To determine whether the inability to utilize orotate is a general trait among L. lactis strains, 11 strains (nos 1–11 in Table 1) were screened for their sensitivity to 5-fluoroorotate. Streaking the strains on defined SA medium containing 50 µg 5-fluoroorotate ml–1 showed that only two strains, DB0410 and 901-1, were affected by 5-fluoroorotate: their growth was inhibited, suggesting that they were able to metabolize orotate in significant amounts. Both strains were further analysed on defined GSA medium with different 5-fluoroorotate concentrations (5, 10, 20 and 50 µg ml–1). No growth of either strain was observed at a concentration of 20 µg ml–1. Growth was severely reduced at 5 µg 5-fluoroorotate ml–1 for both strains.
The genes encoding the enzymes for lactose and casein degradation, required for growth in milk, have been found to be plasmid-borne in lactococci. Since orotate is present in high amounts in cow's milk (Motyl et al., 1991), we examined whether the ability of DB0410 and 901-1 to utilize orotate was plasmid dependent. Plasmids were indeed present in DB0410 and 901-1 as judged by agarose gel electrophoresis of crude extracts (not shown). The plasmid preparations were used to transform the pyrimidine-auxotrophic strain L. lactis ED79.1175 (pyrDa pyrDb), a derivative of L. lactis MG1363, and streaked on defined GSA medium supplemented with 100 µg orotate ml–1 as sole pyrimidine source. MG1363 is unable to metabolize orotate since the strain is resistant to 5-fluoroorotate, and ED79.1175 is unable to utilize orotate as sole pyrimidine source. After 2 days, more than 1000 transformants were obtained with the total plasmid preparation of DB0410, while after 4 days, only 5 transformants were obtained with the plasmid preparation isolated from 901-1. No colonies appeared on GSA plates without orotate, indicating that the transformants obtained had acquired orotate-utilization genes. Thirty-six DB0410 transformants and the five 901-1 transformants were restreaked on GSA agar plates supplemented with 100, 50 and 20 µg orotate ml–1, or with 20 µg uracil ml–1. Strain ED79.1175 transformed with plasmid DNA from DB0410 grew equally well on orotate and uracil. In contrast, ED79.1175 transformed with plasmid DNA from strain 901-1 did not grow efficiently on orotate when supplied at 50 µg ml–1. It was decided to further characterize the strains originating from the transformation with DB0410 DNA.
The plasmid profile of five transformants was analysed and all of them harboured only a single plasmid. It was named pDBORO, and the size was roughly estimated by restriction fragment analysis to be about 16 kbp. The restriction enzyme map is shown in Fig. 2. The complete DNA sequence of the plasmid was determined on both strands with an error rate less then 0.01 % and a redundancy of at least 2.
|
), which is in agreement with the overall G+C content (35.4 mol%) of the L. lactis subsp. lactis strain IL1403 genome sequence (Bolotin et al., 2001). Sequence analysis revealed the presence of 19 ORFs, but ORF prediction carried out with the software programs GENEMARK (Lukashin & Borodovsky, 1998) and EasyGene (Larsen & Krogh, 2003) suggested only 15 ORFs (Table 3). Coding density is 77.8 %, which is lower than the coding density (86 %) of the IL1403 genome (Bolotin et al., 2001). The general organization is shown in Fig. 2(a).
|
), respectively. Similarities between plasmid sequences have been described for other lactococcal plasmids such as pAH90, which indicates a high frequency of horizontal transfer among the lactococci and an overall plasticity in plasmid structures (O'Sullivan et al., 2001).
Table 3 lists the properties of the ORFs and their closest relatives based on BLAST searches. Two of the three IS elements are identical and nearly identical (96 %) to IS946, including the 18 bp inverted repeats (Romero & Klaenhammer, 1990). One IS946 is inserted before ORF8 and the second IS946 is inserted in the end of ORF10. The third IS element is nearly identical (98 %) to IS982 (Yu et al., 1995b) but the 999 bp IS982 on pDBORO is flanked by 16-perfect instead of 18-perfect inverted repeats and has a C to T mutation at position 778. This mutation creates a stop codon in the ORF encoding the transposase, which results in a truncated transposase of 229 instead of 296 aa.
An ORF encoding a protein of 383 aa with high similarity to RepB proteins, an ORF encoding a 258 aa OrfX-like protein and ORF2 are part of the replication module and are described below. The copB gene most probably codes for a copper-potassium-transporting ATPase B first identified in Enterococcus hirae (Odermatt et al., 1993) as they are 98 % identical. The deduced amino acid sequence of copB located in the IL1403 genome contains an HNM repetition in the N-terminal part which is lacking in the deduced amino acid sequence of copB in pDBORO. Three ORFs, ORF10, ORF8 and ORF19, encode proteins with low amino acid sequence identity (42 % or less) to known proteins but in two ORFs a (partial) functional domain could be identified (Table 3). ORF10 encodes a protein of 530 aa showing homology to SunT of the ABC-type bacteriocin/lantibiotic exporter from B. subtilis (Paik et al., 1998). The 50 C-terminal amino acids are however missing, and the last 22 bp of ORF10 overlap with the downstream IS946 element and a new stop codon is created by a read-through into the IS946 sequence. ORF19 encodes a 103 aa protein with a domain found in bacterial nucleoid DNA-binding proteins. The deduced amino acid sequence of ORF8 shows no significant similarity to any protein in the databases searched. The DNA fragment between position 3887 and 5373 contains two ORFs, ORF12 and ORF13, which are nearly identical to ORFB (98 %), with unknown function, and ORFC (98.5 %), a putative resolvase. Both homologues are found on plasmid pCI200.
The replication module of pDBORO
Between positions 5934–7082 and 7082–7855, a non-coding sequence and ORFs were found that are normally involved in the replication of theta-type plasmids (Frere et al., 1993; Gravesen et al., 1995; Hayes et al., 1991). The repB gene codes for a protein of 383 aa with a high amino acid sequence identity to initiator replication proteins of other theta-replicating lactococcal plasmids. The highest identity was found with RepB of the pS7A plasmid (86.4 %; accession no. NC_004652), the food-grade cloning vector pVS40 (85.6 %) (von Wright & Raty, 1993), pCI305 (82 %) (Hayes et al., 1990), pCI605 (82 %; accession no. NC_002138), pAH33 (82 %) (O'Sullivan et al., 2001) and pDR1-1 (82 %; accession no. NC_004163).
Upstream of the repB gene, between positions 5625 and 5933, a region is found that shows high identity at the nucleotide level (80 %) with the cis-acting ori region and the upstream sequence of the repB gene in the lactococcal plasmid pCD4 (Emond et al., 2001). Moreover, this region shows the same organization of secondary structures, including a 41 bp AT-rich box flanked by a GCC and GCGTGG cluster and followed by a tandem of three and a half perfect direct repeats of 22 bp.
The oroP gene found on pDBORO encodes an orotate transporter
To identify the genes responsible for orotate utilization, pDBORO deletion derivatives were constructed (Fig. 2b). Three deletions were made. The pDBORO DNA was digested with XbaI, EcoRV or ClaI and diluted prior to ligation to enhance self-ligation. Ligation mixtures were transformed into L. lactis ED79.1175 (pyrDa pyrDb) and plated on GSA agar plates containing 20 µg orotate ml–1. Only pDBORO derivatives containing the replication module and the orotate transporter coding genes are able to grow. Very few transformants were obtained from the ClaI deletion, but they all contained both ClaI fragments. As shown above, all the functions required for replication are present on one ClaI fragment(Fig. 2b). Several hundred clones were isolated from the two other deletions. This leads to the conclusion that the DNA between the ClaI site at 8.5 kb and the XbaI site at 10.5 kb (Fig. 2b) is required for growth on orotate as sole pyrimidine source. The sequence of the ClaI–XbaI fragment revealed the presence of two complete ORFs, oroP and ysbB, and two truncated ORFs, orf2 and ysbA (Fig. 2c). As ysbA is truncated in the XbaI deletion, and orotate utilization is unaffected, it can be concluded that ysbA is not required for orotate metabolism.
To verify that this region of pDBORO is indeed responsible for orotate uptake, a PCR fragment containing the oroP and ysbB genes was amplified with primers DBORO2 and DBORO8 and cloned into the pCI3340 vector, resulting in construct pED210. By electroporation pED210 was introduced into ED79.1175 and transformants were selected on rich medium containing chloramphenicol. As control the vector pCI3340 was also established in ED79.1175. Transformants were screened on GSA agar plates containing chloramphenicol in the absence and presence of 20 µg orotate ml–1 as the sole pyrimidine source. Only transformants that harboured pED210 could grow on orotate. This confirmed the results of the plasmid deletion analysis. The plasmid pED210 carries only the oroP and ysbB genes, showing that the ability to metabolize orotate is linked to oroP and/or ysbB.
To investigate whether oroP by itself has the ability to facilitate orotate transport, or the ysbB gene is required, plasmid pMBK701 carrying only the oroP gene and the replication module from pDBORO was constructed by PCR amplification and subsequent cloning in the E. coli pTOPO vector. Transformation into E. coli was unsuccessful. Therefore, the ligation mixture was used to transform L. lactis ED79.1175 with selection for growth on orotate as sole pyrimidine source. Transformants were readily obtained, and a restriction enzyme and PCR analysis revealed that they all harboured a plasmid corresponding to pMBK701. It was verified that the plasmid contained the oroP gene in addition to the pTOPO vector. This finding indicates that the only gene on pDBORO conferring the ability to utilize orotate is oroP. The purified plasmid was transformed into E. coli by selecting for ampicillin resistance, and an analysis of transformants showed that the plasmids suffered from deletions. As pTOPO is a high-copy-number plasmid, it is reasonable to conclude that overexpression of oroP alone is detrimental to E. coli.
To verify that oroP facilitates orotate transport, an uptake assay using [14C]orotate was performed. Fig. 3(a) shows a representative experiment in which the L. lactis strains MG1363, DB0410 and MG1363/pED210 (oroP) were assayed for their ability to facilitate orotate uptake. The presence of plasmid pED210 conferred orotate uptake ability to MG1363. Moreover, the uptake was linear right from the start, indicating that orotate is not converted extracellularly.
|
). During such a condition, it would be expected that orotate would be phosphoribosylated to OMP, but not further decarboxylated to UMP (Fig. 1), and hence unable to support growth. In the experiment shown in Fig. 3(b) strain ED79.1175 (pyrDa pyrdDb) harbouring pDBORO was grown with uridine or orotate as sole pyrimidine source in the presence and absence of 6-azauracil. Only the utilization of orotate, and not that of uridine, was impaired by 6-azauracil. Since OMP decarboxylase is inhibited by 6-azaUMP, this result strongly suggests that orotate is indeed taken up by the cell and not converted prior to transport. It could be argued that OMP is formed outside the cell and subsequently transported, but this is extremely unlikely, since 5-phospho-D-ribosyl 1-pyrophosphate (PRPP) is required for the reaction, and phosphorylated compounds with a few exotic exceptions are unable to cross the cytoplasmic membrane.
Based on the DNA sequence, OroP was determined to consist of 207 aa. The deduced sequence was analysed using the TMHMM Server (http://www.cbs.dtu.dk/services/TMHMM), and the overall structure suggests that OroP is a membrane protein with nine transmembrane regions (Fig. 4a). Using the SignalP 3.0 server (http://www.cbs.dtu.dk/services/SignalP) the signal peptide cleavage site in OroP was predicted to be between Ala23 and Asp24 (Fig. 4b). A BLASTP search showed that OroP belongs to a group of conserved hypothetical membrane proteins. Thus, these observations indicate that oroP encodes an orotate transporter, as found in the uptake experiments.
|
As mentioned previously, homology searches using the ysbA, ysbB and oroP reading frames as queries surprisingly revealed that homologues of these genes are present on the chromosome of L. lactis IL1403. The genes are annotated as ysbA, ysbB and ysbC, coding for proteins with unknown functions (Bolotin et al., 2001). The oroP gene was originally annotated as ysbC, but is here renamed according to its function. The gene order oroP-ysbB-ysbA is conserved in both the plasmid pDBORO and the IL1403 chromosome. Compared to IL1403, the deduced oroP amino acid sequence has five amino acid substitutions. Four substitutions conserve a hydrophobic side-chain, whereas the fifth changes an arginine to a glutamine residue. An oroP homologue exhibiting 89 % identity with the oroP ORF was also found in the chromosome of L. lactis MG1363 by searching the genome sequence at GenBank (accession no. AM406671). The gene order oroP (llmg_1937 in the MG1363 sequence)-ysbB (llmg_1936) was conserved in MG1363 whereas ysbA seems to be absent. An oroP allele was also found in the chromosome of L. lactis SK11 by a protein–protein BLAST search at NCBI (http://www.ncbi.nlm.nih.gov/BLAST).
oroP homologues are primarily found in Gram-positives
In order to determine whether putative orotate transporters are widespread, a bioinformatics approach was used to analyse the presence of oroP homologues in the databases. A protein–protein BLAST search using the server at NCBI revealed a number of putative proteins with homology to OroP from pDBORO. Two kinds of protein sequences showed homology to OroP: a long form with similar length to OroP and a short version. Interestingly, the long form seems to have emerged from a duplication of a conserved domain, DUF606, found in ORFs with unknown function. The long version was only found in Gram-positives.
In order to establish their relationship, a cluster analysis based on a CLUSTAL W alignment was conducted. Only the long versions were included. The resulting phylogenetic tree is shown in Fig. 5. OroP homologues from L. lactis are tightly clustered, and in general, the relationships among the different OroP homologues are clustered according to the relationship of the different strains. The OroP homologue from Lactobacillus plantarum is of special interest. The original sequence is most likely wrongly annotated, since it is elongated by 47 aa compared to the OroP homologues from other organisms, and a much better Shine–Dalgarno sequence is present in front of the second methionine codon (data not shown). The oroP homologue from Lb. plantarum (annotated as lp_2696 in the genome sequence) is located immediately downstream of the pyrE gene of the pyrimidine biosynthetic operon, and since no terminators are found between pyrE and lp_2696 they could be co-transcribed, making the oroP homologue a member of the pyr regulon of L. plantarum. This finding, and the membrane protein characteristics of lp_2696, suggests that it is involved in pyrimidine transport.
|
The orotate transporter is functional in both E. coli and B. subtilis
In order to investigate whether the orotate transport system is functional in other organisms, E. coli and B. subtilis strains harbouring the oroP gene from pDBORO were analysed either by the ability of orotate to fulfil a pyrimidine requirement of an auxotrophic mutant or by sensitivity to 5-fluoroorotate of a wild-type strain. Both methods probe for the presence of a functional orotate-uptake system.
Plasmid pED301 was transformed into E. coli KUR1351 (pyrD dctA) and E. coli XL1-Blue. The KUR1351 transformants were plated on selective ABTG medium supplemented with orotate as sole pyrimidine source, whereas XLI-Blue transformants were screened on 5-fluoroorotate (Table 4). B. subtilis strain ED358 (pyrB oroP+) was streaked on selective MM agar plates containing 5 µg neomycin ml–1 without or with 20 µg orotate ml–1 or 20 µg uracil ml–1 as sole pyrimidine source. B. subtilis strain ED348 (oroP+) was streaked on selective MM agar plates containing 5 µg neomycin ml–1 with and without 20 µg 5-fluoroorotate ml–1. Results are shown in Table 4.
|
). Since XL1-Blue is sensitive to 5-fluoroorotate only in the presence of oroP, it is reasonable to believe that this highly manipulated strain has an additional mutation in dctA. Moreover, oroP is required for the ability of the pyrimidine-auxotrophic derivative of B. subtilis, ED358, and of E. coli KUR1351, to grow on orotate as sole pyrimidine source. These results show that OroP is also capable of facilitating the uptake of orotate and 5-fluoroorotate in both E. coli and B. subtilis and, consequently, must be correctly folded and inserted in the plasma membrane. |
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
and Saidi & Warthesen (1989) observed that the orotate concentration in milk decreased during fermentation with lactic acid bacteria, including L. lactis cultures, suggesting that at least some L. lactis strains are able to metabolize orotate. As orotate and 5-fluoroorotate are both metabolized by the same enzymes in the pyrimidine biosynthesis pathway (Fig. 1), sensitivity to 5-fluoroorotate is a powerful method to demonstrate orotate uptake. Among 11 tested strains only two, DB0410 and 901-1, were 5-fluoroorotate sensitive, meaning that these two strains can utilize orotate; this implies that orotate utilization is not a general property of L. lactis. Both strains were sensitive at a low concentration, suggesting the presence of a high-affinity transporter for orotate. The two sensitive strains belong to different subspecies: DB0410 is a strain of L. lactis subsp. lactis biovar diacetylactis, used as a starter culture for production of cheese and cultured buttermilk (Curic et al., 1999), while 901-1 belongs to the cremoris subspecies (Braun et al., 1989). The property to utilize orotate seems therefore to be strain-dependent rather than species-dependent.
Many functions related to growth in milk are found on plasmids in L. lactis, presumably acquired during their adaptation process, as the original niche of lactococci is believed to be plants (Efstathiou & McKay, 1976; McKay et al., 1976; Yu et al., 1995a, 1996). It was therefore interesting to investigate if the orotate transporter genes were also plasmid-borne. For both L. lactis DB0410 and 901-1, single plasmids were isolated after transferring total plasmid preparations into the pyrimidine-auxotrophic strain ED79.1175 and analysing the transformants for their orotate utilization as sole pyrimidine source, but only pDBORO from L. lactis DB0410 was further characterized.
The oroP gene responsible for orotate uptake in L. lactis was shown to code for a protein of 307 aa with the overall structure of a membrane protein consisting of nine transmembrane regions. To our knowledge, this is the first time an orotate transporter has been identified and cloned in a Gram-positive bacterium. The C4-dicarboxylate transport protein DctA was identified to mediate orotate uptake in the Gram-negative bacteria E. coli, Salmonella typhimurium and Sinorhizobium meliloti (Baker et al., 1996; Yurgel & Kahn, 2005). OroP shows no homology to DctA.
An oroP homologue is present in the chromosome of L. lactis strains
Surprisingly, an oroP homologue was found in the sequenced genome of IL1403 but IL1403 is resistant to 5-fluoroorotate, indicating that this homologue does not confer the ability to utilize orotate in significant amounts.
The question to be answered is why one homologue codes for a transporter protein that is able to mediate orotate uptake and the other does not. Both deduced amino acid sequences have the overall structure of a membrane protein consisting of nine transmembrane regions, and they differ by only five conservative substitutions. This suggests that the changes in the amino acid sequence do not account for the inability of the IL1403 oroP homologue to facilitate significant orotate uptake. More likely, it could be a matter of gene expression; both a higher promoter strength and the fact that oroP is present in multiple copies when encoded on pDBORO may account for its capability to facilitate transport of sufficient 5-fluoroorotate for the cell to become sensitive to the drug. We are currently investigating whether a low expression of the chromosomally located gene is responsible for the inability to mediate orotate transport, and whether an increased expression of the chromosomal allele could confer sufficient orotate uptake capability.
The potential of oroP as a selection/counterselection marker
The functionality of oroP in other organisms has been assessed and we showed that the L. lactis orotate transporter is able to mediate 5-fluoroorotate sensitivity in B. subtilis 168, E. coli XL1-Blue, and 5-fluoroorotate sensitive lactococci. Moreover, oroP is required for the ability of the pyrimidine-auxotrophic derivative of B. subtilis, ED358, E. coli KUR1351 and L. lactis ED1175 to grow on orotate as sole pyrimidine source.
Based on the results obtained, the oroP orotate transporter system has potential as a selection/counterselection marker. For application in food, vectors should contain selection markers that are acceptable in the food industry. In general, selection is mostly carried out with the use of antibiotic markers, and food-grade alternatives are rare (Pedersen et al., 2005). The oroP gene originates from the L. lactis plasmid pDBORO and can be applied as a food-grade selective marker. In order for oroP to function as a selection marker, the host must have a pyrimidine requirement caused by a mutation in a gene encoding an enzyme in the first part of the biosynthetic pathway, since the pathway from orotate to UMP encoded by pyrE and pyrF must be intact for orotate taken up by OroP to be metabolized (Fig. 1). Mutants in the first part of the pyrimidine biosynthetic pathway can be obtained either by genetic engineering or by mutagenesis followed by a penicillin counterselection against growth in the absence of uracil. Both methods have successfully been utilized to obtain uracil auxotrophic mutants in L. lactis (Martinussen et al., 2001).
The presence of oroP in the cell results in sensitivity to 5-fluoroorotate. The oroP gene can be exploited as a counterselection marker to select for the loss of a function/plasmid simply by streaking the strains on defined medium containing 5-fluoroorotate. Whereas the use of oroP as a selection marker requires a specific genetic background, its utilization as a counterselection marker only requires that the background strain is unable to transport 5-fluoroorotate. Many organisms are unable to utilize orotate, and those strains that do encode an orotate transporter can easily be modified simply by selecting a 5-fluoroorotate-resistant derivative. Whereas different selection markers are abundantly available, counterselection markers are rare. Therefore, the counterselection property of the oroP system is particularly useful, especially when taking into account that the system may be functional in a large number of organisms (Martinussen & Defoor, 2005).
Previously, an alternative counterselection system has been developed in our lab. It exploits the upp gene, whose presence results in sensitivity to 5-fluorouracil (Breuner et al., 1999; Martinussen & Hammer, 1994, 1995). This system requires that the background strain is mutated in the upp gene. Since the upp gene is very abundant, and not always very easy to obtain, the exploitation of oroP as a counterselection marker is more straightforward.
Interestingly, oroP could only be maintained in E. coli in the presence of the downstream gene ysbB, whereas oroP could be established in L. lactis in the absence of ysbB. The reason for the toxic effect of oroP in the E. coli host in the absence of ysbB could be that ysbB encodes either a regulatory protein, or a chaperone involved in assembly and/or translocation of OroP to the membrane. We are currently trying to solve this problem, making the utilization of the gene easier in different organisms.
|
ACKNOWLEDGEMENTS |
|---|
Edited by: D. A. Mills
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Andersen, P. S., Martinussen, J. & Hammer, K. (1996). Sequence analysis and identification of the pyrKDbF operon from Lactococcus lactis including a novel gene, pyrK, involved in pyrimidine biosynthesis. J Bacteriol 178, 5005–5012.
Antoniewski, C., Savelli, B. & Stragier, P. (1990). The spoIIJ gene, which regulates early developmental steps in Bacillus subtilis, belongs to a class of environmentally responsive genes. J Bacteriol 172, 86–93.
Baker, K. E., Ditullio, K. P., Neuhard, J. & Kelln, R. A. (1996). Utilization of orotate as a pyrimidine source by Salmonella typhimurium and Escherichia coli requires the dicarboxylate transport protein encoded by dctA. J Bacteriol 178, 7099–7105.
Biswas, I., Gruss, A., Ehrlich, S. D. & Maguin, E. (1993). High-efficiency gene inactivation and replacement system for gram-positive bacteria. J Bacteriol 175, 3628–3635.
Bolotin, A., Wincker, P., Mauger, S., Jaillon, O., Malarme, K., Weissenbach, J., Ehrlich, S. D. & Sorokin, A. (2001). The complete genome sequence of the lactic acid bacterium Lactococcus lactis ssp. lactis IL1403. Genome Res 11, 731–753.
Boylan, R. J., Mendelson, N. H., Brooks, D. & Young, F. E. (1972). Regulation of the bacterial cell wall: analysis of a mutant of Bacillus subtilis defective in biosynthesis of teichoic acid. J Bacteriol 110, 281–290.
Braun, V., Jr, Hertwig, S., Neve, H., Geis, A. & Teuber, M. (1989). Taxonomic differentiation of bacteriophages of Lactococcus lactis by electron microscopy, DNA-DNA hybridization, and protein profiles. J Gen Microbiol 135, 2551–2560.
Breuner, A., Brondsted, L. & Hammer, K. (1999). Novel organization of genes involved in prophage excision identified in the temperate lactococcal bacteriophage TP901-1. J Bacteriol 181, 7291–7297.
Chopin, A., Chopin, M. C., Moillo-Batt, A. & Langella, P. (1984). Two plasmid-determined restriction and modification systems in Streptococcus lactis. Plasmid 11, 260–263.[CrossRef][Medline]
Clark, D. J. & Maaloe, O. (1967). DNA replication and the cell cycle in Escherichia coli. J Mol Biol 23, 99–112.[CrossRef]
Curic, M., Stuer-Lauridsen, B., Renault, P. & Nilsson, D. (1999). A general method for selection of alpha-acetolactate decarboxylase-deficient Lactococcus lactis mutants to improve diacetyl formation. Appl Environ Microbiol 65, 1202–1206.
Efstathiou, J. D. & McKay, L. L. (1976). Plasmids in Streptococcus lactis: evidence that lactose metabolism and proteinase activity are plasmid linked. Appl Environ Microbiol 32, 38–44.
Emond, E., Lavallee, R., Drolet, G., Moineau, S. & LaPointe, G. (2001). Molecular characterization of a theta replication plasmid and its use for development of a two-component food-grade cloning system for Lactococcus lactis. Appl Environ Microbiol 67, 1700–1709.
Frere, J., Novel, M. & Novel, G. (1993). Molecular analysis of the Lactococcus lactis subspecies lactis CNRZ270 bidirectional theta replicating lactose plasmid pUCL22. Mol Microbiol 10, 1113–1124.[CrossRef][Medline]
Gasson, M. J. (1983). Plasmid complements of Streptococcus lactis NCDO 712 and other lactic streptococci after protoplast-induced curing. J Bacteriol 154, 1–9.
Gravesen, A., Josephsen, J., von Wright, A. & Vogensen, F. K. (1995). Characterization of the replicon from the lactococcal theta-replicating plasmid pJW563. Plasmid 34, 105–118.[CrossRef][Medline]
Hayes, F., Daly, C. & Fitzgerald, G. F. (1990). Identification of the minimal replicon of Lactococcus lactis subsp. lactis UC317 plasmid pCI305. Appl Environ Microbiol 56, 202–209.
Hayes, F., Vos, P., Fitzgerald, G. F., de Vos, W. M. & Daly, C. (1991). Molecular organization of the minimal replicon of novel, narrow-host-range, lactococcal plasmid pCI305. Plasmid 25, 16–26.[CrossRef][Medline]
Holo, H. & Nes, I. F. (1995). Transformation of Lactococcus by electroporation. Methods Mol Biol 47, 195–199.[Medline]
Huggins, A. R. & Sandine, W. E. (1977). Incidence and properties of temperate bacteriophages induced from lactic streptococci. Appl Environ Microbiol 33, 184–191.
Jensen, P. R. & Hammer, K. (1993). Minimal requirements for exponential growth of Lactococcus lactis. Appl Environ Microbiol 59, 4363–4366.
Johansen, E. & Kibenich, A. (1992). Characterization of Leuconostoc isolates from commercial mixed strain mesophilic starter cultures. J Dairy Sci 75, 1186–1191.[Abstract]
Jorgensen, C. M., Hammer, K. & Martinussen, J. (2003). CTP limitation increases expression of CTP synthase in Lactococcus lactis. J Bacteriol 185, 6562–6574.
Jorgensen, C. M., Hammer, K., Jensen, P. R. & Martinussen, J. (2004). Expression of the pyrG gene determines the pool sizes of CTP and dCTP in Lactococcus lactis. Eur J Biochem 271, 2438–2445.[Medline]
Kilstrup, M. & Martinussen, J. (1998). A transcriptional activator, homologous to the Bacillus subtilis PurR repressor, is required for expression of purine biosynthetic genes in Lactococcus lactis. J Bacteriol 180, 3907–3916.
Kilstrup, M., Hammer, K., Ruhdal, J. P. & Martinussen, J. (2005). Nucleotide metabolism and its control in lactic acid bacteria. FEMS Microbiol Rev 29, 555–590.[CrossRef][Medline]
Kneifel, W., Kaufmann, M., Fleischer, A. & Ulberth, F. (1992). Screening of commercially available mesophilic dairy starter cultures: biochemical, sensory, and microbiological properties. J Dairy Sci 75, 3158–3166.[Abstract]
Larsen, T. S. & Krogh, A. (2003). EasyGene – a prokaryotic gene finder that ranks ORFs by statistical significance. BMC Bioinformatics 4, 21[Medline]
Lukashin, A. V. & Borodovsky, M. (1998). GeneMark.hmm: new solutions for gene finding. Nucleic Acids Res 26, 1107–1115.
Maguin, E., Duwat, P., Hege, T., Ehrlich, D. & Gruss, A. (1992). New thermosensitive plasmid for gram-positive bacteria. J Bacteriol 174, 5633–5638.
Mandel, M. & Higa, A. (1992). Calcium-dependent bacteriophage DNA infection. 1970. Biotechnology 24, 198–201.[Medline]
Marcinkeviciene, J., Tinney, L. M., Wang, K. H., Rogers, M. J. & Copeland, R. A. (1999). Dihydroorotate dehydrogenase B of Enterococcus faecalis. Characterization and insights into chemical mechanism. Biochemistry 38, 13129–13137.
), L. lactis MG1363/pED210 (
) and L. lactis DB0410 (+). The radioactivity in 500 µl filtered culture was determined in a scintillation counter after 1, 3 and 5 min incubation with 5 µM [14C]orotate. (b) Growth of L. lactis ED79.1175/pDBORO in GSA medium at 30 °C with uridine or orotate as sole pyrimidine source in the absence and presence of 10 µM azauracil. 
, Uridine; x, uridine and azauracil; 
