ABSTRACT
The tutB gene, which lies just downstream of tpl, has been cloned from Erwinia herbicola, and its product was analyzed. Despite its high sequence similarity to tryptophan transporters, TutB was found to be a tyrosine-specific transporter. Tryptophan acted as a competitive inhibitor of tyrosine transport. Unlike the tryptophanase operon, the tpl and tutB genes do not constitute an operon.
Tyrosine phenol-lyase (Tpl) is a tyrosine-inducible enzyme distributed in some enteric bacteria (8). Tpl catalyzes the α,β-elimination of l-tyrosine to produce pyruvate, ammonia, and phenol (16), thereby allowing bacteria to utilize l-tyrosine as carbon and nitrogen sources. One of the most notable features of Tpl is its ability to synthesize 3,4-dihydroxyphenylalanine (l-DOPA) from pyruvate, ammonia, and catechol through the reversal of α,β-elimination (15). l-DOPA is used in the treatment of Parkinson's disease, which afflicts 1 out of every 1,700 individuals.
During the course of study of tpl, an open reading frame, designated as tutB, was found just downstream of the tpl gene (9). It was predicted to encode a protein of 416 amino acids with many hydrophobic regions. Since the genetic organization of the tpl and tutB loci (Fig. 1A) is quite similar to that of the tryptophanase operon (tna consisting of tnaA [tryptophan indole-lyase] and tnaB [tryptophan-specific permease]) of Escherichia coli (7), and also since the physiological role of Tpl as to tyrosine is considered to be equivalent to that of TnaA as to tryptophan, the tutB gene was assumed to constitute an operon with tpl and to encode a tyrosine-specific transporter.
Genetic organization of the tpl-tutB locus in E. herbicola (A) and structures of the lac fusions used in this study (B). The open reading frames corresponding to Tpl and TutB are depicted as arrows, and the restriction enzyme cleavage sites for BanII (B), BssHII (Bs), HpaI (H), MfeI (M), NcoI (N), PstI (P), and SmaI (S) are also shown. The DNA sequence between the two open reading frames is shown (see text). Bar, 0.5 kb.
In this study, we cloned the tutB gene from Erwinia herbicola and determined the properties of its product by using Escherichia coli cells. In addition, to elucidate the role of the Tpl-TutB system in the tyrosine degradation pathway, lac fusions were constructed and their expression was monitored. The bacterial strains used in this study were derivatives of E. coli K-12 or E. herbicola AJ2985. The strains and plasmids are listed in Table 1 with their characteristics. Standard recombinant DNA procedures were used (23), and the results of in vitro manipulations such as PCR and mutagenesis were confirmed by DNA sequencing (24).
Strains, plasmids, and oligonucleotides used in this work
Cloning and sequencing of tutB of E. herbicola.
An open reading frame situated 155 nucleotides downstream of the coding region of tpl has already been found and designated tutB although its function has not been elucidated. A plasmid, pEBT, obtained in our previous study (31) contains the tpl gene and a C-terminally truncated tutB gene. The intact tutB gene was cloned from an E. herbicola genomic library constructed as described previously (13), and the resulting plasmid was named pTK928. The 0.6-kb MfeI-HpaI fragment excised from pTK928 and the 3.3-kb SmaI-MfeI fragment cut off from pEBT were annealed at their MfeI-protruding ends, and the resulting 3.9-kb fragment containing the entire tpl and tutB genes in their native configurations was cloned into the SmaI site of pUC19 to generate pTK1016. The structure of the tpl-tutB gene is schematically shown in Fig. 1A. The DNA sequence of tutB was determined by the method described by Sanger et al. (24).
Computer analysis of the deduced amino acid sequence suggested that the product is a membrane protein. A homology search using FASTA (21) revealed that the TutB protein belongs to a family of aromatic amino acid transporters with 11 membrane-spanning segments (25, 26). A detailed discussion as to the similarity among them is presented later.
Transport studies.
In a preliminary experiment, we observed that the introduction of pACYC177 carrying the tutB gene into an E. coli strain with the Φ(tpl′-′lac) gene (14) enhanced the tyrosine-induction ratio of tpl by twofold, but neither the phenylalanine- nor the tryptophan-mediated induction ratio was elevated. Since the expression of tpl is positively regulated by the TyrR protein (22) and its coeffectors (three aromatic amino acids), even though the magnitude of activation is quite different among the three ligands (13, 28), the above observation implied that the tutB gene might encode a tyrosine transporter. In order to characterize the tutB gene product, an E. coli strain deficient in the uptake of all aromatic amino acids (5) was constructed (MG1655 ΔaroP mtr24 ΔpheP Δtna ΔtyrP) and used for transport assays. AroP is a general aromatic transporter with high affinity for the three aromatic amino acids (22), whereas Mtr and TyrP are high-affinity and tryptophan- and tyrosine-specific transporters, respectively (10, 33). The tnaB gene, which constitutes the tna operon with tnaA, encodes a low-affinity, tryptophan-specific transporter (7). The PheP protein, which was initially described as a phenylalanine-specific transporter, has been recently shown to cause the accumulation of tyrosine in cells, although Km for tyrosine was more than 10-fold higher than that for phenylalanine (5). The aroP, pheP, tna, and tyrP genes were disrupted by the method described by Datsenko and Wanner (6). The primers used (aroP-3, aroP-4, pheP-1, pheP-2, tna-1, tna-2, tyrP-3, and tyrP-4) are shown in Table 1. The mtr24 transductant was selected based on resistance to 5-methyltryptophan (10, 11).
The transport assays were performed by the method described by Wookey et al. (33) with l-(U-14C)-tyrosine (456 mCi/mmol, 50 μCi/ml) purchased from Amersham. The rate of nonspecific diffusion of tyrosine was determined using energy-starved cells that had been prepared by incubating cells in the presence of 50 μM carbonylcyanide-m-chlorophenylhydrazone (CCCP) for 30 min prior to starting the assays. The uptake of tyrosine was expressed as picomoles per milligram of dry cells as a function of time.
The E. coli ΔaroP mtr24 ΔpheP Δtna ΔtyrP strain, TK1135, was incapable of accumulating tyrosine (Fig. 2A). Strain TK1135 was transformed with either a low-copy-number (pTK974) or medium-copy-number (pTK1022) plasmid carrying the tutB gene and then examined for the ability to accumulate tyrosine. As shown in Fig. 2A, with an increasing copy number of tutB in the cells, the initial velocity of tyrosine uptake as well as the steady-state level of tyrosine in the cells increased. Under the conditions of 1 to 120 μM tyrosine in the assay mixture, Km for tyrosine in the TutB system was determined to be 38 μM, which is about 10- and 100-fold higher than the Kms for tyrosine in the TyrP and AroP systems (22, 33), respectively, but comparable to Km for tryptophan in the TnaB system (70 μM) (1) and Km for tyrosine in the PheP system (30 μM) (5).
Uptake studies with TutB-expressing E. coli cells. (A) The gene dosage effect on the accumulation of l-(U-14C)-tyrosine in cells. An aromatic amino acid transport-negative strain, TK1135 (▵), was transformed with either a low-copy-number plasmid (pTK974 [□]) or a medium-copy-number plasmid (pTK1022 [▪]) carrying the E. herbicola tutB gene. Cell suspensions were incubated in the presence of 1 μM labeled tyrosine and samples were withdrawn at the indicated times. (B) Uptake and efflux of tyrosine. Strain TK1135 with pTK974 was used for assaying in the presence of 1 μM labeled tyrosine. At the time indicated by the arrow (350 s), CCCP was added to the final concentration of 50 μM. Samples were withdrawn at the indicated times (•). (C) Inability of the strain carrying the mutant tutB (ATG to CTG) allele to accumulate tyrosine. The assay was carried out in the presence of 1 μM labeled tyrosine and at the times indicated samples were removed. The accumulation of tyrosine in the cells carrying the mutant allele (pTK1061 [⋄]) was compared to that in the cells carrying the wild-type allele (pTK1022 [▪]). (D) Inhibition of tyrosine transport by tryptophan. Strain TK1135 transformed with pTK974 was used for assaying. The tyrosine concentration was varied from 5 to 80 μM in the absence (□) or presence (○) of 0.2 mM or 0.5 mM (◊) cold tryptophan. All experiments were repeated twice with essentially the same results, and data for a representative experiment being shown.
The addition of 1 mM sodium ions did not enhance the uptake of tyrosine, and a 30-s preincubation of the cells in the presence of 50 μM CCCP completely abolished the ability to accumulate tyrosine (data not shown). Figure 2B shows the rapid expulsion of accumulated tyrosine from the cells upon the addition of CCCP. These results indicated that the active transport of tyrosine by the TutB system depends on the proton-motive force.
The rate of tyrosine uptake at the saturating concentration (100 μM) was not affected by the addition of a 20-fold molar excess of l-phenylalanine (data not shown), but it decreased in the presence of l-tryptophan in a concentration-dependent manner (Fig. 2D). We examined the ability of TutB-expressing TK1135 cells to accumulate labeled tryptophan (58.1 mCi/mmol, 0.02 mCi/ml; NEN Life Science Products, Inc.); however, no accumulation was observed within the tested range (10 μM to 1 mM). A double-reciprocal plot of the values in Fig. 2D was found to well-fit the case of competitive inhibition, in which the K[infi]i value for tryptophan was determined to be 300 μM. As an alternative approach for examining whether the TutB protein transports tryptophan or not, we used generalized transduction involving a P1 phage lysate prepared on an E. coli trp mutant. If the TutB protein transports tryptophan, the presence of a plasmid carrying tutB in aromatic transport-negative strain TK1135 should allow us to obtain trp transductants of the cells on the minimal medium supplemented with tryptophan. However, this approach did not work well because such transductants were easily obtained when strain TK1135 itself was infected with a P1 phage prepared on a trp mutant. Since the amount of labeled tryptophan that accumulated in the aromatic amino acid transport-negative cells (TK1135) was exactly the same as that accumulated in energy-starved cells (data not shown), the above finding implies that the nonspecific diffusion of tryptophan can support the growth of a tryptophan-auxotroph strain. In the minimal medium supplemented with tryptophan, no difference in the growth rate was observed between TK1135 with the trp mutation and the strain carrying the tutB gene on a plasmid (data not shown). On the basis of these results, we concluded that the TutB protein is a tyrosine-specific transporter and that its activity is inhibited by a competitive inhibitor, tryptophan.
Determination of translation start codon of tutB.
Before constructing the Φ(tutB′-′lac) gene, the translation start site was determined. Multiple alignment of the deduced amino acid sequences of TutB and other transporters in the same family (Fig. 3) suggested that the N-terminal amino acid of TutB is methionine followed by valine, isoleucine, and two consecutive lysine residues. To verify this assumption, site-directed mutagenesis, in which ATG for this methionine was replaced with CTG, was carried out by the method of Kunkel et al. (17). tutB-N in Table 1 was used as a mutagenic primer. The aromatic amino acid transport-negative E. coli strain, TK1135, transformed with pACYC177 carrying either the wild-type (pTK1022) or mutant (pTK1061) tutB allele, was subjected to the transport assay. As shown in Fig. 2C, the cells carrying the mutant allele were incapable of accumulating tyrosine, demonstrating that this methionine is an actual initiation codon. A possible ribosome-binding site was found 6 nucleotides upstream, which is indicated by a wavy line in Fig. 1A.
Comparison of amino acid sequences of TutB and E. coli aromatic amino acid transporters in the same family. TyrP, a high-affinity, tyrosine-specific transporter (33); TnaB, a low-affinity, tryptophan-specific transporter of the tna operon (26); and Mtr, a high-affinity, tryptophan-specific transporter (10). Alignment was constructed with the use of the ClustalW 1.6 program (32), and identical residues (∗) and conservative amino acid changes (•) in the four sequences are shown. The amino acid residues distinctively conserved in the tyrosine or tryptophan transporters at the same positions on alignment are highlighted. Eleven membrane-spanning regions were deduced according to the model proposed for Mtr by Sarsero and Pittard (25), and are enclosed by boxes.
Structural comparison of TutB with other aromatic amino acid transporters.
One unexpected finding was that, when the primary structure of TutB was compared to those of other transporters in the same family, the TutB protein showed higher similarity to tryptophan transporters (Mtr and TnaB) (7, 10) than a tyrosine transporter (TyrP) (33). The TutB protein exhibited 48 and 46% identity with Mtr and TnaB in amino acid sequence, respectively; however, it showed only 31% identity with TyrP. The fact that the TutB protein is capable of recognizing both tryptophan and tyrosine but is active only in tyrosine transport might indicate that this protein should be a powerful tool for investigating the molecular mechanism underlying the active transport of aromatic amino acids.
The membrane topology of the TutB protein was deduced from the model proposed for the Mtr protein (25), the 11 membrane-spanning segments enclosed by boxes in Fig. 3. Scrutiny of the four sequences revealed residues distinctively conserved in the tyrosine- or tryptophan-specific transporters at the same positions on alignment, which are highlighted in Fig. 3. Some of these residues were predicted to exist in loop regions, but this is not so surprising because, as has been reported previously, in the case of AroP-PheP chimeras, replacement of a cytoplasmic loop significantly influenced their transport activity (5).
Expression of the Φ(tpl′-′lac), Φ(tpl+-tutB′-′lac), and Φ(tutB′-′lac) genes in E. coli.
For a better understanding of the role of the tpl-tutB gene in the tyrosine degradation pathway, we constructed the Φ(tpl′-′lac), Φ(tpl+-tutB′-′lac), and Φ(tutB′-′lac) translation fusions (Fig. 1B) and monitored their expression in E. coli cells. Construction of the Φ(tpl′-′lac) gene was described elsewhere (14), and the Φ(tutB′-′lac) gene was created by connecting the initiation codon of the tutB gene with the seventh codon of the ′lac gene on pRS552 (27). The primers used (tutB-F and tutB-R) are shown in Table 1. A plasmid, pTK1144, carrying the truncated ′lac gene was also constructed and used as a control for β-galactosidase assaying. An E. coli strain, TK596, was transformed with two compatible plasmids; one was a low-copy-number plasmid containing one of the lac reporter genes (pTK1005, pTK1031, pTK1032, or pTK1144) and the other was a single-copy plasmid carrying either the tpl or tpl-tutB gene (pTK913 or pTK935). Cells were grown in M63-glucose minimal medium (19) in the absence and presence of 1 mM l-tyrosine, and then subjected to the β-galactosidase assay (19). The results are presented in Table 2.
Specific β-galactosidase activity expressed from the Φ(tpl′-′lac), Φ(tpl+-tutB′-′lac), and Φ(tutB′-′lac) genes in E. coli cells grown under tyrosine-induced or noninduced conditions
As for the Φ(tpl′-′lac) gene, its expression was increased by 7-fold upon the addition of tyrosine in the absence of the tutB gene (460 to 3,300) and by 10-fold in the presence of the tutB gene (440 to 4,400). Under tyrosine-induced conditions, the β-galactosidase activity of the cells carrying tutB was 1.3-fold as high as that of the cells not carrying tutB (4,400 against 3,300). The mechanism underlying TyrR-mediated tyrosine induction of tpl has been well established (14, 28). These results indicated that introduction of the TutB system actually increased the intracellular level of tyrosine. It is known that the expression of two major tyrosine transporters, AroP and TyrP, is severely repressed by tyrosine (22); however, both systems played significant roles in the accumulation of tyrosine in the cells grown in the medium supplemented with tyrosine, which resulted in a sevenfold induction of tpl in the absence of tutB. In our assay system, AroP-expressing E. coli cells (MG1655 mtr24 ΔpheP Δtna ΔtyrP) grown in the presence of 1 mM tyrosine accumulated as much tyrosine as was observed in TutB-expressing cells (pMBO131 carrying tutB/TK1135) grown under the same conditions (data not shown), which may account for the relatively low effect of tutB on the tyrosine-induction of tpl.
The β-galactosidase activity of the strain carrying the Φ(tpl+-tutB′-′lac) gene was extremely low but significant compared to that of the strain carrying the ′lac gene (control), indicating that the tutB gene is actually expressed in E. coli cells. The expression of tutB was unaffected by the addition of tyrosine to the medium. Changing the carbon source from glucose to glycerol had no effect on expression (data not shown). Introduction of the tutB gene did not have any effect on the expression of the fusion either. The low-level expression of tutB was not due to the lowered concentration of tyrosine in the cells on the introduction of extra copies of the tpl (tyrosine-degrading enzyme) gene because introduction of pUC19 carrying tpl into the cells had only a partial effect on the tyrosine induction of the Φ(tpl′-′lac) gene (data not shown). β-Galactosidase activity as well as the mode of expression of the Φ(tutB′-′lac) gene were essentially the same as those in the case of the Φ(tpl+-tutB′-′lac) gene. These results demonstrate that the expression of tutB is very low and constant, and unlike the tna operon of E. coli (7, 29), the tpl and tutB genes are transcribed separately. This notion was supported by the results of Northern hybridization analysis (23), with total RNA extracted from E. herbicola cells grown under tyrosine-induced conditions, where an intense signal corresponding to 1.6 kb in length was observed when the tpl-specific probe was used (30), but no distinct band was observed when the tutB-specific probe was used (data not shown). The DNA sequence between the stop codon of tpl and the start codon of tutB is shown in Fig. 1A, and a possible promoter of the tutB gene is indicated by underlining.
These results revealed that in contrast to the physiological equivalence of Tpl-TutB and TnaA-TnaB, both of which generate pyruvate (carbon source) and ammonia (nitrogen source) from aromatic amino acids, the mode of expression was quite different between them. At present, we do not have a clear answer as to why the expression level of tutB remained extremely low even in the presence of tyrosine. However, too much accumulation and degradation of tyrosine via the Tpl-TutB system may be unfavorable for cells because of the toxic effect of phenol liberated from tyrosine.
Effect of TutB on expression of Tpl in E. herbicola.
The chromosomal tpl-tutB locus in E. herbiocola was deleted and replaced with the kanamycin resistance gene (kan+) by means of a homologous recombination event using linearized pTK967, and the resulting Δ(tpl-tutB)::kan+ strain was transformed with a mini-F-derived plasmid (pMBO131) or pMBO131 carrying either the tpl-tutB gene (pTK932) or the tpl-tutB(ATG to CTG) gene (pTK1143). It has been shown that the F plasmid is stably maintained in Erwinia species (3). The Tpl contents of the cells grown under tyrosine-induced and non-induced conditions were determined by immunoblotting (13), and the results are presented in Fig. 4. Apparently, the induction of Tpl by tyrosine was observed regardless of the presence or absence of the active TutB system (lanes 3, 4, 5, and 6); however, the presence of the functional tutB gene actually enhanced the tyrosine-induction ratio of Tpl (twofold; lane 4 versus lane 6). These results indicated that, as in the case of the lac reporter assay results for E. coli (Table 2), the TutB system is necessary for full induction of Tpl and also revealed the existence of other tyrosine transport system(s) in E. herbicola.
The Tpl contents of various E. herbicola strains. E. herbicola Δ(tpl-tutB)::kan+ cells transformed with pMBO131 (lanes 1 and 2) or pMBO131 carrying either the tpl-tutB gene (pTK932, lanes 3 and 4) or the tpl-tutB(ATG to CTG) gene (pTK1143, lanes 5 and 6) were grown in M63 minimal medium containing 0.02% yeast extract in the presence and absence of 1 mM tyrosine. When OD600 reached 0.5, the cells were harvested from 1 ml of culture and then suspended in 100 μl of cracking buffer (13). After boiling for 5 min, 10 μl of each whole-cell extract was applied and separated on a sodium dodecyl sulfate-12.5% polyacrylamide gel (18). Immunoblotting with anti-Tpl antibodies was performed as described previously (13). The image on X-ray film was analyzed with a Fujifilm ImageGauge program, and the densitometric values were estimated within the linear range.
In conclusion, we cloned the tutB gene from E. herbicola and analyzed its product in E. coli cells. Although the amino acid sequence of TutB exhibited significant similarity to those of tryptophan transporters, it specifically transported l-tyrosine. Tryptophan acted as a competitive inhibitor of tyrosine transport. Unlike the tna operon, the tpl and tutB genes did not constitute an operon.
Nucleotide sequence accession number.
The DNA sequence of tutB was deposited in GenBank under accession number AF418598 .
ACKNOWLEDGMENTS
We are very grateful to W. Wackernagel for providing pCP20 and B. L. Wanner for providing pKD46. We also thank R. J. Kadner for his helpful comments on the manuscript.
This work was partly supported by a Grant-in-Aid for Scientific Research (A), no. 10306007, from the Ministry of Education, Culture, Sports, Science, and Technology, Japan, and by a Grant-in-Aid for Fine Enzymatic Synthesis of Useful Compounds from Research for the Future (RFTF) of the Japan Society for the Promotion of Science.
FOOTNOTES
- Received 20 November 2001.
- Accepted 14 March 2002.
- Copyright © 2002 American Society for Microbiology
REFERENCES
- Top
- Article
- ABSTRACT
- Cloning and sequencing of tutB of E. herbicola.
- Transport studies.
- Determination of translation start codon of tutB.
- Structural comparison of TutB with other aromatic amino acid transporters.
- Expression of the Φ(tpl′-′lac), Φ(tpl+-tutB′-′lac), and Φ(tutB′-′lac) genes in E. coli.
- Effect of TutB on expression of Tpl in E. herbicola.
- Nucleotide sequence accession number.
- ACKNOWLEDGMENTS
- FOOTNOTES
- REFERENCES
- Figures & Data
- Info & Metrics