Portal de Periódicos da CAPES

Tpn1p, the Plasma Membrane Vitamin B6 Transporter of Saccharomyces cerevisiae

2003; Elsevier BV; Volume: 278; Issue: 21 Linguagem: Inglês

10.1074/jbc.m300949200

1083-351X

Jürgen Stolz, Martin Vielreicher,

Bacterial Genetics and Biotechnology

Pyridoxine (PN) is a metabolic precursor of pyridoxal phosphate that functions as a cofactor of many enzymes in amino acid metabolism. PN, pyridoxal, and pyridoxamine are collectively referred to as vitamin B6, and mammalian organisms depend on its uptake from the diet. In addition to the ability to use extracellular vitamin B6, most unicellular organisms are also capable of synthesizing PN to generate pyridoxal phosphate. Here, we report the isolation of Saccharomyces cerevisiae mutants that have lost the ability to transport PN across the plasma membrane. We used these mutants to isolate TPN1, the first known gene encoding a transport protein for vitamin B6. Tpn1p is a member of the purine-cytosine permease family within the major facilitator superfamily. The protein functions as a proton symporter, localizes to the plasma membrane, and has high affinity for PN. TPN1 mutants lost the ability to utilize extracellular PN, pyridoxal, and pyridoxamine, showing that there is no other transporter for vitamin B6 encoded in the genome. Amino acid substitutions that led to a loss of Tpn1p function localized to transmembrane domain 4 within the 12-transmembrane domain protein. Moreover, expression of TPN1 was regulated and increased with decreasing concentrations of vitamin B6 in the medium. We also provide evidence that of the highly conserved SNZ and SNO genes in S. cerevisiae, only the protein encoded by SNZ1 is required for vitamin B6 biosynthesis. Pyridoxine (PN) is a metabolic precursor of pyridoxal phosphate that functions as a cofactor of many enzymes in amino acid metabolism. PN, pyridoxal, and pyridoxamine are collectively referred to as vitamin B6, and mammalian organisms depend on its uptake from the diet. In addition to the ability to use extracellular vitamin B6, most unicellular organisms are also capable of synthesizing PN to generate pyridoxal phosphate. Here, we report the isolation of Saccharomyces cerevisiae mutants that have lost the ability to transport PN across the plasma membrane. We used these mutants to isolate TPN1, the first known gene encoding a transport protein for vitamin B6. Tpn1p is a member of the purine-cytosine permease family within the major facilitator superfamily. The protein functions as a proton symporter, localizes to the plasma membrane, and has high affinity for PN. TPN1 mutants lost the ability to utilize extracellular PN, pyridoxal, and pyridoxamine, showing that there is no other transporter for vitamin B6 encoded in the genome. Amino acid substitutions that led to a loss of Tpn1p function localized to transmembrane domain 4 within the 12-transmembrane domain protein. Moreover, expression of TPN1 was regulated and increased with decreasing concentrations of vitamin B6 in the medium. We also provide evidence that of the highly conserved SNZ and SNO genes in S. cerevisiae, only the protein encoded by SNZ1 is required for vitamin B6 biosynthesis. Vitamins are essential dietary compounds for many organisms. Most water-soluble vitamins or derivatives thereof function as cofactors in enzymatic reactions. The group of vitamins that is referred to as vitamin B6 consists of pyridoxine (PN), 1The abbreviations used are: PN, pyridoxine; PL, pyridoxal; PM, pyridoxamine; PLP, pyridoxal 5′-phosphate; ORF, open reading frame; GFP, green fluorescent protein; 3-AT, 3-amino-1,2,4-triazole. pyridoxal (PL), and pyridoxamine (PM). The active intracellular form of vitamin B6, pyridoxal 5-phosphate (PLP), has multiple roles as a versatile cofactor of enzymes that almost exclusively function in the metabolism of amino compounds. PLP-dependent enzymes include α-, β-, and γ-synthetases, racemases, and decarboxylases of amino acids as well as aminotransferases (1John R.A. Biochim. Biophys. Acta. 1995; 1248: 81-96Crossref PubMed Scopus (331) Google Scholar). Unlike mammals, most unicellular organisms and plants are prototrophic for vitamin B6. The pathway that catalyzes PN biosynthesis is best characterized in Escherichia coli (2Dempsey W.B. Neidhart F.C. Ingraham J.L. Low B.L. Magasanik B. Schaechter M. Umbarger H.E. Escherichia coli and Salmonella typhimurium. 1. American Society for Microbiology, Washington, D. C.1987: 539-543Google Scholar, 3Drewke C. Leistner E. Vitam. Horm. 2001; 61: 121-155Crossref PubMed Google Scholar). A recently discovered pathway of vitamin B6 biosynthesis depends on the products of the SNZ (also referred to as PDX1) and SNO (also referred to as PDX2) genes (4Ehrenshaft M. Chung K.R. Jenns A.E. Daub M.E. Curr. Genet. 1999; 34: 478-485Crossref PubMed Scopus (41) Google Scholar, 5Ehrenshaft M. Bilski P. Li M.Y. Chignell C.F. Daub M.E. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 9374-9378Crossref PubMed Scopus (256) Google Scholar, 6Ehrenshaft M. Daub M.E. J. Bacteriol. 2001; 183: 3383-3390Crossref PubMed Scopus (70) Google Scholar, 7Osmani A.H. May G.S. Osmani S.A. J. Biol. Chem. 1999; 274: 23565-23569Abstract Full Text Full Text PDF PubMed Scopus (150) Google Scholar, 8Bean L.E. Dvorachek Jr., W.H. Braun E.L. Errett A. Saenz G.S. Giles M.D. Werner-Washburne M. Nelson M.A. Natvig D.O. Genetics. 2001; 157: 1067-1075Crossref PubMed Google Scholar). Organisms encode either Snz/Sno-like proteins or proteins with homology to E. coli PdxA/PdxJ, but not both (5Ehrenshaft M. Bilski P. Li M.Y. Chignell C.F. Daub M.E. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 9374-9378Crossref PubMed Scopus (256) Google Scholar, 9Mittenhuber G. J. Mol. Microbiol. Biotechnol. 2001; 3: 1-20PubMed Google Scholar). The proteins encoded by the SNZ and SNO genes are highly conserved and have members in all three domains of life (9Mittenhuber G. J. Mol. Microbiol. Biotechnol. 2001; 3: 1-20PubMed Google Scholar). The Cercospora nicotianae SNZ homolog SOR1 is required for resistance to singlet oxygen-generating photosensitizers (10Ehrenshaft M. Jenns A.E. Chung K.R. Daub M.E. Mol. Cell. 1998; 1: 603-609Abstract Full Text Full Text PDF PubMed Scopus (95) Google Scholar). This finding was surprising and led to the discovery that PN, PL, PM, and PLP directly act as chemical quenchers of singlet oxygen and are as effective as well known antioxidants such as vitamins C and E (5Ehrenshaft M. Bilski P. Li M.Y. Chignell C.F. Daub M.E. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 9374-9378Crossref PubMed Scopus (256) Google Scholar, 11Bilski P. Li M.Y. Ehrenshaft M. Daub M.E. Chignell C.F. Photochem. Photobiol. 2000; 71: 129-134Crossref PubMed Scopus (283) Google Scholar). Thus, the pyridoxine biosynthesis pathway serves a dual function to provide PLP as a cofactor and to provide protection against active oxygen species. Three homologs of SNZ genes and three homologs of SNO genes are present in the genome of Saccharomyces cerevisiae. The genes encoding these proteins are arranged into divergently oriented SNZ-SNO gene pairs that are regulated by a common promotor (12Padilla P.A. Fuge E.K. Crawford M.E. Errett A. Werner-Washburne M. J. Bacteriol. 1998; 180: 5718-5726Crossref PubMed Google Scholar). The expression of SNZ1 is induced in stationary phase (13Braun E.L. Fuge E.K. Padilla P.A. Werner-Washburne M. J. Bacteriol. 1996; 178: 6865-6872Crossref PubMed Google Scholar), a feature later also found for SNO1 (12Padilla P.A. Fuge E.K. Crawford M.E. Errett A. Werner-Washburne M. J. Bacteriol. 1998; 180: 5718-5726Crossref PubMed Google Scholar). In S. cerevisiae, deletion of SNZ1 causes PN dependence and deletion of SNO1 slows growth, whereas deletion of SNO2, SNO3, SNZ2, or SNZ3 does not lead to PN requirement (14Rodriguez-Navarro S. Llorente B. Rodriguez-Manzaneque M.T. Ramne A. Uber G. Marchesan D. Dujon B. Herrero E. Sunnerhagen P. Perez-Ortin J.E. Yeast. 2002; 19: 1261-1276Crossref PubMed Scopus (81) Google Scholar). Sno1p and Snz1p are present in a high molecular mass complex (12Padilla P.A. Fuge E.K. Crawford M.E. Errett A. Werner-Washburne M. J. Bacteriol. 1998; 180: 5718-5726Crossref PubMed Google Scholar) and show two-hybrid interactions with each other as well as with other proteins, including Snz2/3p and Sno2/3p (12Padilla P.A. Fuge E.K. Crawford M.E. Errett A. Werner-Washburne M. J. Bacteriol. 1998; 180: 5718-5726Crossref PubMed Google Scholar, 14Rodriguez-Navarro S. Llorente B. Rodriguez-Manzaneque M.T. Ramne A. Uber G. Marchesan D. Dujon B. Herrero E. Sunnerhagen P. Perez-Ortin J.E. Yeast. 2002; 19: 1261-1276Crossref PubMed Scopus (81) Google Scholar, 15Ito T. Tashiro K. Muta S. Ozawa R. Chiba T. Nishizawa M. Yamamoto K. Kuhara S. Sakaki Y. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 1143-1147Crossref PubMed Scopus (663) Google Scholar, 16Uetz P. Giot L. Cagney G. Mansfield T.A. Judson R.S. Knight J.R. Lockshon D. Narayan V. Srinivasan M. Pochart P. Qureshi-Emili A. Li Y. Godwin B. Conover D. Kalbfleisch T. Vijayadamodar G. Yang M. Johnston M. Fields S. Rothberg J.M. Nature. 2000; 403: 623-627Crossref PubMed Scopus (3915) Google Scholar). In vitamin B6 auxotrophic mutants, uptake of PN, PL, or PM from the extracellular space is required to generate intracellular PLP via a salvage pathway. The biochemistry of vitamin B6 uptake is best studied in Saccharomyces carlsbergensis (17Shane B. Snell E.E. J. Biol. Chem. 1976; 251: 1042-1051Abstract Full Text PDF PubMed Google Scholar), an allopolyploid hybrid thought to have originated from fusion of S. cerevisiae with an unidentified yeast species (18Casaregola S. Nguyen H.V. Lapathitis G. Kotyk A. Gaillardin C. Int. J. Syst. Evol. Microbiol. 2001; 51: 1607-1618Crossref PubMed Scopus (88) Google Scholar). Based on experiments with labeled substrates, Shane and Snell (17Shane B. Snell E.E. J. Biol. Chem. 1976; 251: 1042-1051Abstract Full Text PDF PubMed Google Scholar) concluded that S. carlsbergensis has two high affinity transport systems for vitamin B6. Consistent with an active transport mechanism for vitamin B6 uptake, S. carlsbergensis accumulates PN up to 3000-fold relative to outside concentrations, and transport is sensitive to metabolic inhibitors (17Shane B. Snell E.E. J. Biol. Chem. 1976; 251: 1042-1051Abstract Full Text PDF PubMed Google Scholar). Accumulation of PN is not a unique feature of S. carlsbergensis, but is commonly found in many yeast species, regardless whether they are prototrophic or auxotrophic for vitamin B6 (19Yagi T. Kim Y. Hiraoka Y. Tanouchi A. Yamamoto T. Yamamoto S. Biosci. Biotechnol. Biochem. 1996; 60: 893-897Crossref PubMed Scopus (6) Google Scholar). Despite much evidence that vitamin B6 uptake across the plasma membrane is a protein-mediated process, not a single sequence of a PN transport protein is known to date. Here, we used S. cerevisiae mutants to identify Tpn1p, a protein from the purine-cytosine permease family that mediates the high affinity uptake of vitamin B6 across the plasma membrane. Yeast Strains and Media—The yeast media used were YPD (1% yeast extract, 2% peptone, and 2% dextrose) and SD (synthetic dextrose; 2% glucose and 0.67% yeast nitrogen base without amino acids). SD medium contains 2 μg/liter PN (1.93 μm). For media with varying concentrations of vitamin B6, yeast nitrogen base without amino acids and without vitamins (Bio 101, Inc.) was used. All vitamins were added to the standard concentrations present in yeast nitrogen base, and PN, PL, or PM was added to give the desired concentrations. Only supplements that were required by the strains were added. Growth assays were performed with yeast cell suspensions prepared in water in 96- well plates. Starting from the highest cell density (A600 = 0.6), serial 10-fold dilutions were made and transferred to plates with a replicating tool. The lager brewing yeast S. carlsbergensis 4228 was obtained from Deutsche Sammlung von Mikroorganismen und Zellkulturen (Braun-schweig, Germany). FY1679-08a (MATaura3-52 leu2-Δ1 trp1-Δ63 his3- Δ200) and the isogenic ygl186cΔ strain (MATaura3-52 his3-Δ200 trp1- Δ63 ygl186cΔ::kanMX4) were obtained from EUROSCARF (Frankfurt am Main, Germany). All other yeast strains used in this study were isogenic to W303-1A (MATaleu2-3,112 his3-11,15 trp1-1 ade2-1 ura3-1 can1-100) (20Thomas B.J. Rothstein R. Cell. 1989; 56: 619-630Abstract Full Text PDF PubMed Scopus (1349) Google Scholar). MW980 (MATasnz1-sno1Δ::URA3 snz2-sno2Δ::LEU2 snz3-sno3Δ::LEU2 trp1-1 ade2-1 his3-1 can1-100) (12Padilla P.A. Fuge E.K. Crawford M.E. Errett A. Werner-Washburne M. J. Bacteriol. 1998; 180: 5718-5726Crossref PubMed Google Scholar) was mutagenized with ethyl methanesulfonate. Two mutants (MW980 tpn1-1 and tpn1-3) with slow growth on 0.02 μm PN were crossed with RKY02-1D (MATα ade2-1 ura3-1 trp1-1 leu2-3 his 3-11,15 can1-100 bna1Δ::HIS3) (21Kucharczyk R. Zagulski M. Rytka J. Herbert C.J. FEBS Lett. 1998; 424: 127-130Crossref PubMed Scopus (48) Google Scholar). A haploid strain derived from the cross of MW980 tpn1-3 that showed slow growth on medium containing 0.02 μm PN was then crossed with W303-1A snz1-sno1Δ::his5+ (MATaleu2-3,112 trp1-1 ade2-1 ura3-1 can1-100 snz1-sno1Δ::his5+). This cross was performed to substitute the snz1-sno1Δ::URA3 deletion for a snz1-sno1Δ::his5+ allele. From the progeny, JSX14-3a (MATaleu2-3,112 trp1-1 ade2-1 ura3-1 can1-100 snz1-sno1Δ::his5+tpn1-3) was used for complementation. The linked SNZ1-SNO1 genes were deleted in W303-1A with PCR products using the Schizosaccharomyces pombe his5+ gene as a marker. The ygl186cΔ::kanMX4 deletion was transferred from FY1679 into W303-1A and MW980 after PCR with primers flanking the integration site. All deletions were verified by PCR. Plasmid Constructs—An S. cerevisiae library in the multicopy vector YEp352 (22te Heesen S. Knauer R. Lehle L. Aebi M. EMBO J. 1993; 12: 279-284Crossref PubMed Scopus (106) Google Scholar) was used. It contains genomic DNA that was partially digested with Sau3AI and ligated into the BamHI site. Plasmids that complemented the growth defect of JSX14-3a were characterized by restriction digests to identify SNZ1-SNO1-containing plasmids. All other plasmids were sequenced. To generate a full-length allele of TPN1, the part of the gene that is missing in the library clone was amplified with Vent polymerase and ligated to the part of TPN1 that encodes the C terminus using the HindIII site within the open reading frame (ORF). Overexpression of TPN1 was achieved in pVT100-U, a URA3 multicopy vector with the yeast ADH1 promotor (23Vernet T. Dignard D. Thomas D.Y. Gene (Amst.). 1987; 52: 225-233Crossref PubMed Scopus (465) Google Scholar). All other constructs were made in the YCplac (single copy) and YEplac (multicopy) series of plasmids (24Gietz R.D. Sugino A. Gene (Amst.). 1988; 74: 527-534Crossref PubMed Scopus (2522) Google Scholar). A green fluorescent protein (GFP)-tagged version of TPN1 was generated in YEplac195, comprising the TPN1 promotor (429 bp upstream of the ATG start codon), followed by the GFP sequence, the TPN1 ORF, and 742 bp of TPN1 terminator sequence. The plasmid was transformed into W303-1A wild-type cells, and fluorescence was recorded with a Zeiss LSM 510 META confocal microscope. An identical construct in YCplac33 did not yield any fluorescence above background levels. To generate the SNZ1-containing plasmids, a library plasmid harboring the SNZ1-SNO1 gene pair was cut with SphI and BspEI to isolate SNZ1 along with 478 bp of promotor and 691 bp of terminator sequences. The SNZ1 gene was ligated into SphI-XmaI-cut YCplac33 or YEplac195. SNO1 was isolated as a ScaI-HindIII fragment containing 427 bp of promotor and 454 bp of terminator sequences. SNO1 was ligated into SmaI-HindIII-cut YCplac22 or YEplac112. All plasmids were transformed into W303-1A snz1-sno1Δ::his5+tpn1Δ::kanMX4. For the analysis of TPN1 alleles from S. cerevisiae and S. carlsbergensis, PCR with Vent polymerase was performed on genomic DNA as a template and primers that allowed amplification of the entire ORF. The products of two independent PCRs were ligated into pCR4-TOPO (Invitrogen) and sequenced. Uptake Experiments—Cells were grown to mid-logarithmic phase in SD medium, washed with water, and suspended in citric acid/phosphate buffers containing 1% d-glucose to energize the cells. 500 μl of cells (0.25 A600 units/ml for cells overexpressing TPN1 and 10 A units/ml for others) were incubated at 30 °C and stirred with a magnetic stirrer. The substrate, a mixture of tritiated PN (20 Ci/μmol; American Radiolabelled Chemical, Inc., St. Louis, MO) and unlabeled PN, was added to give a final concentration of 10 μm. Competitors or inhibitors were tested at pH 4.0 and added 60 s before starting the experiment. At given times, 60-μl aliquots were withdrawn, diluted with 5 ml of water, and filtered. After washing with excess water, the radioactivity associated with the filters was quantified by scintillation counting. RNA Analyses—MW980 cells were cultured overnight in media with 2, 0.2, and 0.02 μm PN and diluted into fresh media with the same concentrations of PN. Cells from the 0.02 μm PN culture were additionally used to inoculate media with 0, 2, and 0.2 nm PN. RNA was prepared from cells (40 A600 units), separated on formaldehyde gels, and transferred to nitrocellulose membranes. The blots were incubated with 32P-labeled probes corresponding to the entire ORF of TPN1. RNA prepared from strains carrying a tpn1Δ deletion showed no signals with the TPN1 probe (data not shown). Blots were exposed to phosphorimaging screens for quantification. Signals obtained with an actin (ACT1) probe were used for normalization. Generation of Mutants with Defects in PN Uptake—The yeast S. cerevisiae has two pathways for the synthesis of the essential cofactor PLP. One pathway requires the de novo synthesis of pyridoxine. The other pathway is the salvage pathway, which involves the transport of external sources of vitamin B6 across the plasma membrane. Because intracellular PN can derive from these two pathways, we reasoned that mutation of a putative PN transporter alone would not give rise to a phenotype. We therefore used strain MW980, which lacks all three SNZ-SNO gene pairs, for a mutational approach to identify the PN transporter. As expected, S. cerevisiae wild-type strains were capable of growing on a wide variety of PN concentrations and also grew in the absence of PN (Fig. 1A). MW980 grew like the wild-type strain when PN was present at concentrations of 0.02 μm or higher, but grew poorly at lower concentrations or when PN was absent (Fig. 1A). To identify mutants with defects in PN uptake, MW980 was treated with ethyl methanesulfonate and plated on YPD medium. Colonies were replicated on medium containing 2 or 0.02 μm PN. Two mutant strains carrying recessive and allelic mutations (tpn1-1 and tpn1-3) were isolated. Both mutants were growth-restricted on plates containing 20 nm PN or less, but grew like wild-type cells on plates with higher concentrations of PN (Fig. 1A). Compared with MW980, both mutants also had growth deficits on plates containing PL or PM (data not shown). Using [3H]PN as a substrate in a whole cell uptake experiment, we found that both mutants were defective in PN transport compared with W303-1A (wild-type) or MW980 (Fig. 1B). The amount of PN taken up by cells (1 A600 unit) in the 4-min experiment was 2.4 pmol for tpn1-1, 4.2 pmol for tpn1-3, 12.3 pmol for the wild-type strain, and 13.7 pmol for MW980. The higher activity of tpn1-3 relative to tpn1-1 is consistent with its slightly better growth (Fig. 1A). Moreover, the vitamin B6 auxotrophic mutant MW980 possessed higher activity for PN transport compared with wild-type cells. We conclude that both tpn1 mutants are defective in growth on low concentrations of vitamin B6 because of a defect in PN uptake. Complementation Cloning of YGL186C—To eliminate unwanted mutations, we next crossed both mutants with a wild-type strain and tested the progeny from this cross for the presence of the mutation. All strains that showed growth defects possessed the snz1-sno1Δ::URA3 deletion, ruling out a role of SNZ2-SNO2 and SNZ3-SNO3 in PN biosynthesis. Strain JSX14-3a, which carries the tpn1-3 mutation, resulted after a second cross that exchanged the snz1-sno1Δ::URA3 deletion for a snz1-sno1Δ::his5+ allele. JSX14-3a was transformed with a yeast genomic library generated in a multicopy vector and plated on medium containing 20 nm PN. Library transformants were tested on plates containing no vitamin B6 to identify clones that were able to synthesize PN. All of these transformants contained plasmids with the linked SNZ1-SNO1 genes. We did not isolate a single plasmid with the SNZ2-SNO2 or SNZ3-SNO3 gene pair from >20 PN prototrophic strains. Plasmids that did not confer vitamin B6 prototrophy were sequenced, and eight of them contained a gene coding for a protein with homology to plasma membrane transporters. The identified ORF, YGL186C, codes for a protein with homology to Fcy2p, the S. cerevisiae transporter for adenine, guanine, and cytosine (25Weber E. Rodriguez C. Chevallier M.R. Jund R. Mol. Microbiol. 1990; 4: 585-596Crossref PubMed Scopus (63) Google Scholar, 26Brethes D. Napias C. Torchut E. Chevallier J. Eur. J. Biochem. 1992; 210: 785-791Crossref PubMed Scopus (9) Google Scholar). In S. cerevisiae, Ygl186cp, Fcy2p, and the Fcy21 and Fcy22 proteins that are similar to Fcy2p form the family of purine-cytosine permeases within the major facilitator superfamily (27Nelissen B. De Wachter R. Goffeau A. FEMS Microbiol. Rev. 1997; 21: 113-134Crossref PubMed Google Scholar). Ygl186cp is 29% identical to Fcy2p and 58% identical to Pcpl3p, a protein from Kluyveromyces marxianus whose function is unknown (28Ball M.M. Raynal A. Guerineau M. Iborra F. J. Mol. Microbiol. Biotechnol. 1999; 1: 347-353PubMed Google Scholar). An alignment of these three proteins is presented in Fig. 5. Interestingly, cytosine, one of the substrates of Fcy2p, and vitamin B6 compounds show similarities in their chemical structures (Fig. 2A).Fig. 2Structural similarity between vitamin B6 and cytosine and hydropathy analysis of Tpn1p.A, chemical structures of pyridoxal, pyridoxine, pyridoxamine, and cytosine. Note that the structures shown are those for the neutral compounds, whereas the forms of vitamin B6 are a mixture of ionic species whose charge depends on the pH of the medium. B, hydrophobicity analysis of the Tpn1p primary structure with TMpred (29Hofmann K. Stoffel W. Biol. Chem. Hoppe-Seyler. 1993; 374: 166Google Scholar). Default settings were used to calculate the hydropathy profile. The position of Met66 (M66), the putative start site of the protein encoded by the truncated TPN1 allele, is indicated. Roman numerals are given for hydrophobic protein segments that are likely to represent transmembrane domains. aa, amino acids.View Large Image Figure ViewerDownload Hi-res image Download (PPT) Analysis with the TMpred program (29Hofmann K. Stoffel W. Biol. Chem. Hoppe-Seyler. 1993; 374: 166Google Scholar) showed that the 579-amino acid protein encoded by YGL186C is a membrane protein with 12 transmembrane domains (Fig. 2B), a feature typical of plasma membrane permeases for various substrates (30Henderson P.J. Curr. Opin. Cell Biol. 1993; 5: 708-721Crossref PubMed Scopus (130) Google Scholar). However, the YGL186C ORF was contained only partially on all eight plasmids we isolated. The promotor and N terminus of the protein were missing due to the presence of a Sau3AI restriction site that had been cut during the generation of the library. The protein encoded by the truncated gene putatively starts at methionine 66 within the hydrophilic N-terminal part of the protein (Fig. 2B), thus leaving intact the 12 hydrophobic domains that presumably are essential for activity. Although tpn1-3 mutants carrying multicopy plasmids with the truncated gene did not display elevated levels of plasma membrane PN transport relative to controls (data not shown), the plasmids reproducibly complemented the growth phenotype of the mutants. It seemed possible that the truncated gene was expressed at levels that were undetectable by transport assays, but were sufficient to rescue the growth phenotype of tpn1-3 cells. Moreover, we were encouraged by the structural similarity of PN, PL, and PM to cytosine and the homology of Ygl186cp to Fcy2p and characterized YGL186C in more detail. YGL186C Encodes a Plasma Membrane Transport Protein for Pyridoxine—We performed uptake experiments with FY1679 wild-type cells and the isogenic ygl186cΔ deletion mutant to determine whether Ygl186cp functions in PN transport (Fig. 3A). Deletion of YGL186C caused a strong reduction in plasma membrane transport of PN, with the activity dropping from 27 pmol of PN taken up within 4 min in the wild-type cells to 3 pmol of PN in the ygl186cΔ mutant (Fig. 3A). This low level of PN uptake is comparable to the activity of the tpn1-1 and tpn1-3 mutants (Fig. 1B) and shows that YGL186C is involved in the plasma membrane transport of PN. In light of this observation and also data described below, we will refer to YGL186C as TPN1 (transport of pyridoxine-1). A full-length allele of TPN1 was generated and overexpressed using pVT100-U. Overexpression of TPN1 increased PN uptake activity to 700 pmol of PN/A600 units of cells within 5 min (Fig. 3B). This corresponds to a >20-fold increase in PN transport relative to wild-type cells. The increase in PN transport upon overexpression and the decrease in activity following deletion of TPN1, as well as the homology to Fcy2p, made it very likely that Tpn1p is a PN permease. This was further substantiated by the localization of the N-terminally GFP-tagged protein (Fig. 3C). Whereas W303-1A control cells produced only weak vacuolar fluorescence (presumably due to the accumulation of a red pigment in ade2 mutants), the GFP-TPN1-expressing cells showed fluorescence at the plasma membrane (Fig. 3C, right panel). Some intensely fluorescent spots were seen in cells that had high expression levels, and these might correspond to artifacts originating from overexpression. The uptake experiments and data on the localization of Tpn1p thus provide strong evidence that TPN1 encodes a plasma membrane-localized transporter for PN. Tpn1p Is a Proton-PN Symporter with a Broad Substrate Specificity—Cells overexpressing TPN1 from a multicopy plasmid were used for a detailed characterization of the PN transport activity. We found that PN transport was maximal when the uptake experiments were performed at pH 4.0, and 50% of this activity was detectable at pH 7.0. The activity decreased more rapidly toward acidic pH values, falling to 45% at pH 3.0. The Km value for PN uptake is 0.55 μm (data not shown), classifying Tpn1p as a high affinity transporter for PN. We next performed uptake experiments at pH 4.0 to determine whether other vitamin B6 compounds compete with PN uptake (Fig. 4A). We found that PM had only a weak effect when present at a 10-fold molar excess, but PN uptake was reduced to 50% at a 50-fold excess of PM. PL was more potent in competing PN uptake when present at a 10-fold excess. PN and 4-deoxypyridoxine both were equally effective competitors and reduced the PN uptake to 5% of control levels. It is possible that the weak competition of PM relative to PL, PN, and 4-deoxypyridoxine is caused by its amino group, which introduces an additional positive charge. PN uptake was also strongly reduced in the presence of the protonophore carbonyl cyanide m-chlorophenylhydrazone, and the remaining activity amounted to 9% of uninhibited cells. This indicates that, similar to the homologous Fcy2 protein (31Polak A. Grenson M. Eur. J. Biochem. 1973; 32: 276-282Crossref PubMed Scopus (75) Google Scholar, 32Reichert U. Winter M. Biochim. Biophys. Acta. 1974; 356: 108-116Crossref PubMed Scopus (29) Google Scholar, 33Chevallier M.R. Jund R. Lacroute F. J. Bacteriol. 1975; 122: 629-641Crossref PubMed Google Scholar, 34Eddy A.A. Hopkins P. Microbiology. 1996; 142: 449-457Crossref PubMed Scopus (3) Google Scholar), Tpn1p acts as a proton symporter. Thus, the PN transporter Tpn1p has a broad specificity for all three forms of vitamin B6. Tpn1p Is the Only Pyridoxine Transporter in S. cerevisiae— Experiments with S. carlsbergensis had provided evidence that two activities catalyze PN uptake (17Shane B. Snell E.E. J. Biol. Chem. 1976; 251: 1042-1051Abstract Full Text PDF PubMed Google Scholar). We investigated whether the situation in S. cerevisiae is similar and analyzed the effect of deletion of TPN1 in a wild-type strain and in MW980 (Fig. 4B). A tpn1Δ strain grew like the wild-type strain under all conditions tested (Fig. 4B), although it possessed only residual levels of PN transport activity (Fig. 3A). This shows that wild-type cells derive most of their PN from biosynthesis. In contrast, growth was severely compromised when TPN1 was deleted in MW980. The concentrations that were necessary to support growth of MW980 tpn1Δ were 2 μm for PN and PL, whereas PM was at least 10-fold less potent. At high outside concentrations, sufficient amounts of PL and PN might enter the cells by passive diffusion, whereas PM might diffuse more slowly because it is positively charged. Because MW980 tpn1Δ failed to grow on low concentrations of PN, PL, or PM and because PM and PL acted as competitors of PN transport, we conclude that S. cerevisiae has only one transport protein for all three forms of vitamin B6 and that this protein is Tpn1p. Sequencing of tpn1-1 and tpn1-3 and the TPN1 Gene from S. carlsbergensis—A diploid strain constructed by mating tpn1-3 and tpn1Δ strains possessed growth and PN transport characteristics that were indistinguishable from those of the haploid tpn1-3 strain (data not shown). This indicated that the TPN1 gene was mutated in MW980 tpn1-3 and, by inference, also in MW980 tpn1-1. To identify these mutations, the gene was am