Localization and function of the yeast multidrug transporter Tpo1p.

In Saccharomyces cerevisiae four transporters, Tpo1p-Tpo4p, all members of the major facilitator superfamily, have been shown to confer resistance to polyamines. It was suggested that they act by pumping their respective substrate into the lumen of the vacuole depending on the proton gradient generated by the V-ATPase. Using sucrose gradient ultracentrifugation we found that an hemagglutinin (HA)-tagged Tpo1p as well as its HA-tagged Tpo2p-4p homologues co-localize with plasma membrane markers. Because the HA-tagged Tpo1p carrier protein proved to be functional in conferring resistance to polyamines in TPO1 knockouts, a function of Tpo1p in transport of polyamines across the plasma membrane seemed to be likely. The polyamine transport activity of wild type cells was compared with the respective activity of a TPO1 knockout strain. The results obtained strongly suggest that Tpo1p is a plasma membrane-bound exporter, involved in the detoxification of excess spermidine in yeast. When studying polyamine transport of wild type cells, we furthermore found that S. cerevisiae is excreting putrescine during the fermentative growth phase.

In Saccharomyces cerevisiae four transporters, Tpo1p-Tpo4p, all members of the major facilitator superfamily, have been shown to confer resistance to polyamines. It was suggested that they act by pumping their respective substrate into the lumen of the vacuole depending on the proton gradient generated by the V-ATPase. Using sucrose gradient ultracentrifugation we found that an hemagglutinin (HA)-tagged Tpo1p as well as its HA-tagged Tpo2p-4p homologues co-localize with plasma membrane markers. Because the HA-tagged Tpo1p carrier protein proved to be functional in conferring resistance to polyamines in TPO1 knockouts, a function of Tpo1p in transport of polyamines across the plasma membrane seemed to be likely. The polyamine transport activity of wild type cells was compared with the respective activity of a TPO1 knockout strain. The results obtained strongly suggest that Tpo1p is a plasma membrane-bound exporter, involved in the detoxification of excess spermidine in yeast. When studying polyamine transport of wild type cells, we furthermore found that S. cerevisiae is excreting putrescine during the fermentative growth phase.
Polyamines are essential compounds occurring in virtually all prokaryotic and eukaryotic cells (1). Because these compounds are toxic in higher concentrations, their intracellular content is tightly regulated by controlling biosynthesis and degradation and also by transport into intracellular storage compartments or out of the cell. In the yeast Saccharomyces cerevisiae the polyamines putrescine, spermidine, and spermine are found. All three compounds are actively transported into the vacuole by a substrate/nH ϩ antiport mechanism (2). Four transport proteins, Tpo1p-Tpo4p, have been postulated to be responsible for the polyamine transport capability of the yeast vacuolar membrane (3,4). According to their protein sequences, these four proteins are all members of a family of multidrug resistance transporters within the major facilitator superfamily. Tpo1p was identified by its sequence similarity to the Bacillus subtilis multidrug transporter Blt, which is involved in spermidine excretion (5). The other transporters of this group were found on the basis of sequence similarity to Tpo1p (4). The function of all Tpo proteins was deduced from the phenotype of mutant strains. Their overexpression led to an increased tolerance for polyamines added to the growth medium, whereas a respective deletion rendered the cells more sensitive to polyamines. Evidence for their participation in vacuolar polyamine uptake, however, was largely indirect.
In the course of a project to identify new vacuolar transporters belonging to the class of secondary carriers, we were looking for a well characterized carrier of this type in the yeast vacuolar membrane to use as a control for localization assays. We chose Tpo1p because of its sequence similarity to the transport proteins we were interested in, in particular to Ycr023p. In an analysis of their subcellular distribution, however, we found that all Tpo transporters are actually localized in the cytoplasmic membrane. This observation encouraged us to perform a detailed study on Tpo1p function. The results obtained suggest that at least Tpo1p participates in polyamine export out of the cell.

EXPERIMENTAL PROCEDURES
Strains and Media-The haploid S. cerevisiae strain 23344c (MAT␣ ura3) was used as genetic background for all of the experiments. This strain is isogenic with ⌺1278b (6) and was kindly provided by Bruno André (Brussels, Belgium). The cells were grown aerobically at 30°C. Preparation of yeast-rich (YPD) and synthetic complete minimal media followed standard recipes (7). Growth assays on solid media were performed using a modified citrate-buffered yeast minimal medium (8) in which the Mg 2ϩ content was limited to 50 M to enhance polyamine sensitivity (9). For polyamine export experiments and for the preparation of vacuolar vesicles, a modified CBS medium described by Verduyn et al. (10) was used, containing 20 g/liter glucose and 5 g/liter NH 4 SO 4 . Uracil auxotrophic strains were grown in the presence of 20 mg/liter uracil. The CBS medium was buffered with either 1% succinic acid adjusted to pH 5.8 with NaOH or 50 mM potassium phthalate at pH 5.5 Gene Modifications-Replacement of the TPO1 gene by lacZ was done via PCR-based gene targeting using the gene disruption cassette encoded by plasmid pUG6lacZ (11). The plasmid contains the lacZ gene from Escherichia coli, followed by the dominant kanMX marker, and is a derivative of pUG6 described by Gü ldener et al. (12). The sequences of the primers used for the amplification of the disruption cassette were 5Ј-TTTTTTTTAGTCAAAGAAGCAAGAGAAAACTAGACAGAGACAA-TGTTCGTACGCTGCAGGTCGAC and 5Ј-AAAAATGCAAATATAGA-AAGAGCATGATTTCTGCTTTTCTTTTTCGCATAGGCCACTAGTGG-ATCTG.
Genomic HA 1 tagging of the four TPO genes and of the open reading frame YCR023c was performed by using the plasmid pUG6-HA (13), which encodes three tandem repeats of a HA epitope followed by kanMX. Via PCR a DNA molecule was generated, consisting of a 3xHA-kanMX marker cassette flanked by short regions homologous to the end of the respective gene. The PCR primers used for this purpose consisted of 45 nucleotides corresponding to the genomic sequence ultimately upstream and downstream, respectively, of the stop codon of the gene to be tagged, followed by 20 nucleotides homologous to the pUG6-HA plasmid to amplify the epitope and the kanMX marker. After transformation of the 1.7-kb PCR product into strain 23344c and selection for resistance to G418 (200 mg/l) on YPD agar plates, the stop codon of the respective gene was usually replaced by 3xHA-kanMX through homologous recombination. All of the gene modifications were verified by diagnostic PCR. A list of the strains used and generated in this study is given in Table I.
Plasmid Constructions-The yeast expression vector pDR199 was obtained from Wolf Frommer (Tü bingen, Germany). It is a derivative of pDR195 (14), which in turn was generated from YEplac195 (15). The plasmid contains a copy of the PMA1 promoter and the terminator region of the ADH1 gene. The TPO1 gene was amplified via PCR using genomic DNA of strain 23344c as template and cloned between promoter and terminator using an XmaI and an XhoI site. The PCR primers for TPO1 were 5Ј-GCGTCCCCGGGATGTCGGATCATTCTC-CCATT and 5Ј-GCGTCCTCGAGTTAAGCGGCGTAAGCATACTT. The underlined sequences indicate the XmaI and XhoI sites, respectively. The TPO1-3xHA fusion gene was amplified via PCR with genomic DNA from strain RK 25 as template using the same 5Ј primer as for the amplification of the unmodified TPO1. The sequence of the 3Ј primer was GCGTCCTCGAGTTAGGCGGCGTAGTCAGGAAC, with the XhoI site underlined. The PCR fragment was inserted into pDR199 by using the XmaI and the XhoI site between PMA1 promoter and ADH1 terminator. Accuracy of the constructs was verified by sequencing. The plasmids were transformed into yeast according to Gietz et al. (16).
Sucrose Density Gradient Centrifugation-Subcellular localization of triple HA-tagged proteins was determined following the protocol of Sorin et al. (17). The cells were grown in 600 ml of YPD medium to A 600 of 2-3. NaN 3 (10 mM) was added prior to harvesting, and the culture was chilled on ice. The cells were converted to spheroplasts by adding lysing enzymes (Sigma) in a concentration of 1 mg/ml in S/K buffer (1.2 M sorbitol, 100 mM potassium phosphate, pH 7.5) and incubating for 45 min at 30°C. The spheroplasts were suspended in 4 ml of lysis buffer (0.3 M sorbitol, 20 mM triethanolamine acetate, pH 7.2, 1 mM EDTA, supplemented with a commercially available protease inhibitor mixture (Complete, EDTA-free; Roche Molecular Biochemicals) and homogenized by 30 strokes of a Wheaton A Dounce homogenizer. Unlysed cells were removed by centrifugation (800 ϫ g for 3 min at 4°C), and the supernatant was layered on top of a noncontinuous gradient ranging from 18 to 54% (w/v) sucrose in 10 mM Hepes, pH 7.1, 1 mM EDTA (10 steps of 4% difference each). The gradients were centrifuged for 2 h at 40,000 rpm (4°C) in a Beckmann SW41 Ti rotor and fractionated manually from top to bottom (12 fractions of 1 ml each). Isolated fractions were diluted (1:5) in 100 mM Tris-Cl, pH 7.5, 150 mM NaCl, 5 mM EDTA, and the membranes were pelleted by ultracentrifugation (100,000 ϫ g for 2 h at 4°C). The resulting pellets were incubated on ice (30 min) in 400 l of Tris buffer (see above) containing 5 M urea. They were again centrifuged (17,000 ϫ g for 45 min at 4°C) and finally resuspended in 100 l of SDS sample buffer.
Immunoblotting-The proteins were separated by SDS-PAGE (12%) and electroblotted onto nitrocellulose membranes using a semidry transfer system. For immunodetection a monoclonal anti-HA (Roche Molecular Biochemicals), a monoclonal anti-Vph1p (kindly provided by Patricia M. Kane, Syracuse, NY), and a polyclonal anti-Pma1p (kindly provided by Bruno André, Brussels, Belgium) were used as first antibody. The respective secondary antibody coupled to horseradish peroxidase and the enhanced chemiluminescence detection system (Invitrogen) was used for visualization.
Polyamine Transport in Intact Yeast Cells-The cells were grown overnight in synthetic complete minimal medium, harvested at A 600 ϭ 1.0, washed twice in the same volume of 20 mM Na/Hepes buffer, pH 7.2, containing 10 mM glucose and resuspended at A 600 ϭ 1.0 in the same buffer. Transport assays were done at 30°C and started by adding [ 14 C]spermine to a final concentration of 20 or 100 M. 100-l aliquots were filtered at defined time points through nitrocellulose filters that were preincubated in 100 mM LiCl containing 1 mM spermine. The radioactivity trapped on the filters was determined in a liquid scintillation counter.
Preparation of Vacuole-derived Vesicles and Transport Measurements-The preparation of vacuolar vesicles was done following a slightly modified protocol of Ohsumi and Anraku (18). The cells were grown in CBS medium buffered with 50 mM potassium phthalate, pH 5.5, and supplemented with 10 mM putrescine and 1 mM of each spermidine and spermine to an A 600 of 2-3. The cells were harvested and resuspended in 100 mM Tris-sulfate, pH 9.4, 10 mM dithiothreitol to yield a concentration of 0.5 g of cell fresh weight/ml. The cells were shaken for 10 min at 30°C, centrifuged, and resuspended to 0.15 g of cell fresh weight/ml in SOB (1.2 M sorbitol, 5 mM MES-Tris, pH 6.9). The spheroplasts were generated by incubation with lysing enzymes (Sigma) (1 mg/5 ϫ 10 8 cells) for 30 min at 30°C. The spheroplasts were washed twice in SOB and resuspended in buffer A (12% Ficoll, 10 mM MES-Tris, pH 6.9, 100 M MgCl 2 ), protease inhibitors (Complete; Roche Molecular Biochemicals) were added, and the suspension was homogenized by seven strokes of a Wheaton A Dounce homogenizer. The resulting suspension was transferred to a centrifuge tube, overlaid with half the volume of buffer A, and centrifuged at 60,000 ϫ g for 1 h and 15 min. The white floating layer that appeared after centrifugation was transferred to an ultracentrifuge tube, adjusted with buffer A (see above) to 6 ml, and overlaid with 6 ml of buffer B (8% Ficoll, 10 mM MES-Tris, pH 6.9, 100 M MgCl 2 ). After further centrifugation (60,000 ϫ g for 1 h and 15 min), vacuolar vesicles formed a white layer on top of the solution. They were aspirated, suspended in 100 mM MES-Tris, pH 6.9, 100 mM KCl, 20 M MgCl 2 , frozen under liquid nitrogen, and stored for further use at Ϫ80°C. The enzyme assays were done according to Roberts et al. (19). Phenylalanine transport measurements were performed according to Sato et al. (20), and spermine import was determined after Kakinuma et al. (2).
Export Measurements of Accumulated Spermidine-The cells were grown overnight in succinate-buffered CBS medium in the presence of 10 mM spermidine and harvested at A 600 ϭ 1.3-1.5 by centrifugation. The cell pellets were washed three times in medium without polyamines but containing only 1 g/liter NH 4 SO 4 and finally resuspended in the same medium. The cell suspension was incubated at 30°C with agitation, and aliquots were taken at defined time points. To analyze spermidine efflux, the cells were immediately separated from the medium by centrifugation (11,000 ϫ g for 3 min), and released spermidine was quantified using high performance liquid chromatography according to Price et al. (21).
Determination of Total Polyamine Content-Extraction of total polyamines was performed according to Tomitori et al. (4). Yeast cells were incubated in 10% trichloroacetic acid at 65°C for 1 h. Derivatization and quantification by high performance liquid chromatography were done as described above.

RESULTS
Tpo1p Is Localized in the Cytoplasmic Membrane-In the course of an approach to identify new vacuolar transporters, we determined the subcellular localization of various putative transport proteins that have been fused C-terminally to a triple HA epitope tag. A version of Tpo1p with the same tag was generated as bona fide control, because we considered it to be an established vacuolar transporter (3,4). The intracellular localization was determined by sucrose density gradient centrifugation of membrane preparations isolated from yeast cells carrying the respective fusion construct as the only gene copy. Defined fractions were separated by SDS-PAGE and subsequently analyzed by immunoblotting (Fig. 1). Although the 3xHA-tagged version of Ycr023p co-localized with the 100-kDa subunit of the vacuolar ATPase (Vph1p), the Tpo1p-3xHA fusion co-localized with the plasma membrane ATPase (Pma1p), thus raising the question of whether tagging of Tpo1p leads to mislocalization, although the HA tag is not known to direct proteins to the plasma membrane.
To get further evidence we tested the functionality of the HA  (3). This phenotype should be rescued by expression of the HA-modified version of the transporter if this fusion is functional. We thus transformed a plasmid encoded version of TPO1-3xHA into a strain in which TPO1 had been replaced by a lacZ-kanMX deletion cassette. The fused gene on the plasmid was expressed under the control of the PMA1 promoter. The resulting yeast cells were tested for spermidine sensitivity in comparison with the wild type and the tpo1::lacZ strain, each transformed with the empty vector (Fig. 2). We found that synthesis of the HA-tagged Tpo1p leads to markedly increased spermidine tolerance. Because the functionality of the HA fusion of Tpo1p could be demonstrated, we tried to get further evidence for the Tpo1 protein being located in the plasma membrane, where it should contribute to polyamine transport.
The Polyamine Uptake Rate of Intact Cells Is Not Influenced by TPO1 Deletion-Because the localization of Tpo1p in the plasma membrane was contradictory to the proposed function in vacuolar polyamine transport, we tested first whether polyamine import into cells is influenced by TPO1 deletion. An earlier report had indicated that a TPO1 deletion strain shows spermine import rates similar to wild type cells (3), whereas a decreased import rate in TPO1 deletions was reported later (4). We found, however, no difference in spermine uptake activity between the tpo1::lacZ strain and wild type using 100 M (Fig.  3) or 20 M spermine (not shown), suggesting that Tpo1p is not involved in polyamine import of yeast cells and/or that Tpo2p-Tpo4 may complement the knockout of Tpo1p.
Vacuolar Vesicles from a TPO1 Disruption Strain Do Not Show Any Impairment in Polyamine Uptake-In view of several reports describing Tpo1p as vacuolar polyamine importer, we tried to account for an impairment in vacuolar polyamine uptake after TPO1 deletion. Consequently, we prepared vacuolar vesicles from wild type cells and tpo1::lacZ mutants, respectively, grown in the presence of 10 mM putrescine, 1 mM spermidine, and 1 mM spermine. Polyamines were added to the growth medium because a stimulatory impact of externally added polyamines on TPO1 expression had been reported recently (4). The polyamine concentrations employed led to a markedly increased intracellular polyamine level (about 4-fold in the case of putrescine and spermine, and about 2-fold for spermidine), which should be high enough to enhance the expression of TPO1.
We optimized the protocol for the isolation of vacuolar vesicles, which led to preparations of high purity (Table II). The ATPase activity of the vesicle preparation was completely inhibited by the addition of the V-ATPase-specific inhibitor concanamycin A (not shown), proving that significant contaminations of plasma membrane and mitochondria were virtually absent. The vesicles were used to determine the uptake of amino acids (20) and spermine (2). Because phenylalanine uptake measurements in particular were highly reproducible in isolated vacuolar vesicles, we used this carrier activity to normalize the spermine uptake rate and thus to eliminate possible variations caused by the vesicle preparation, a strategy that was not employed in the original report of Tomitori et al. (4), who reported a slightly decreased import activity for vacuolar vesicles prepared from a TPO1 deletion strain compared with wild type. But this correction might be crucial, because we found that although the absolute values varied slightly for each preparation, the relative uptake rates measured were essen- Cell growth was assayed on solid synthetic medium in which magnesium content has been limited to 50 M. Spermidine was added at a concentration of 6 mM (right half ); the same strains have also been applied to a plate without spermidine to ensure uniform growth (left half ). The plate without spermidine was incubated at 30°C for 2 days, and the plate with spermidine was incubated for 6 days at the same temperature. The growth of wild type and tpo1::lacZ cells each transformed with the empty vector (pDR199) is shown together with the tpo1::lacZ strain transformed with a plasmid encoding a Tpo1p-3xHA fusion (pDR199 TPO1-3xHA). tially the same for the two strains (Fig. 4). Furthermore, the maximal amount of accumulated spermine was very similar for the two vesicular preparations when normalized to the maximum of accumulated phenylalanine. We thus conclude that the TPO1 deletion had no effect on spermine import into vacuolar vesicles.
Polyamine Export of Growing Yeast Cells-The absence of the TPO1 gene causes increased polyamine sensitivity, indicating a function for Tpo1p in the detoxification of polyamines ( Fig. 2 and Ref. 3). Detoxification in general can be mediated either by transport into the vacuole or out of the cell across the plasma membrane. The results presented so far favor the latter possibility. Thus, we analyzed the putative polyamine efflux mediated by Tpo1p.
Because little is known about polyamine export in budding yeast, we investigated first whether polyamines might be released into the medium during growth, as has been reported for other microorganisms (22,23). Wild type and tpo1::lacZ cells were grown from A 600 ϭ 0.5 to stationary phase in succinatebuffered CBS medium. Putrescine and spermidine excreted into the medium were quantified. Spermine could not be determined as a contaminant in the synthetic growth medium and interfered with the quantitative HPLC analysis. We found that wild type and mutant cells in fact excrete putrescine but not spermidine until they reach the diauxic shift (Fig. 5) after ϳ24 h. The putrescine release of the two strains was indistinguishable, suggesting that Tpo1p is not involved in this process.
Release of Accumulated Spermidine Is Impaired in a tpo1 Mutant-Although spermidine is the most abundant polyamine in yeast cells (Ref. 9; see also Fig. 6B), we did not observe release of this solute under normal growth conditions. To challenge yeast cells for spermidine export, we increased the intracellular spermidine content by growing cells in the presence of this compound. The cells were cultured in the presence of 10 mM spermidine in succinate-buffered CBS medium under nonlimiting Mg 2ϩ -conditions, where both wild type and mutants are more tolerant to polyamine stress, and the spermidine concentrations applied are thus not toxic for both strains. During the incubation the total intracellular spermidine content of yeast cells in the logarithmic growth phase increased significantly from about 8 to 25 nmol/mg cell dry mass (Fig. 6B). After washing and resuspending in polyamine-free medium, the accumulated spermidine was released. We found a significant difference between wild type and tpo1 deletion mutant under these conditions. Spermidine efflux of the tpo1::lacZ cells was significantly decreased as compared with the wild type (Fig.  6A). The observed difference in efflux rates was not due to a different preloading of cells with spermidine, as can be seen in Fig. 6B. To provide a solid basis for the observed difference in polyamine excretion, the initial efflux rates of 15 independent cultures of wild type and mutant cells, respectively, were determined. It could be shown that the spermidine efflux rate of the TPO1 deletion strain drops to 49 Ϯ 28% of wild type rates (Fig. 6C). These experiments indicate a function of Tpo1p in spermidine detoxification. Preloading yeast cells with putrescine unfortunately led to widely varying results; however, in a series of experiments we did not observe a statistically significant difference between the wild type and the TPO1 deletion strain in terms of putrescine efflux activity (not shown).
Complementation of the Impaired Export Activity-If the observed impairment of spermidine efflux was indeed a consequence of the TPO1 deletion, the introduction of plasmid-borne Tpo1p should lead to an increase of the spermidine export rate. Therefore, we constructed a plasmid expressing TPO1 under control of the PMA1 promoter and transformed this construct into the tpo1::lacZ strain. Expression of the plasmid encoded gene was controlled by analyzing polyamine sensitivity of the resulting yeast, which was in fact highly decreased (not shown).
The initial spermidine export rates of this strain were determined and gave values in the same range as observed for the  4. Relative spermine uptake of isolated vacuolar vesicles. Vacuolar vesicles from wild type (filled bars) and tpo1::lacZ cells (open bars) have been prepared. The cells have been grown in phthalatebuffered CBS medium supplemented with 10 mM putrescine and 1 mM of both spermidine and spermine. Total uptake and the uptake rate of the vesicles were determined for spermine and phenylalanine. The ratio of spermine and phenylalanine uptake and uptake rate, respectively (relative uptake), is shown at the ordinate. wild type (Fig. 6C). When the tpo1::lacZ strain was transformed with the empty vector, the export rates were in the same range as observed for the TPO1 deletion without any plasmid (not shown). The initially accumulated spermidine concentration was unchanged in both experiments. Because TPO1 was found to complement the described phenotype, we consider Tpo1p to be a plasma membrane-bound exporter involved in spermidine export.
Triple HA Fusions of Tpo2p-4p Are Also Located in the Plasma Membrane-The fact that Tpo1p was localized in the plasma membrane raises the question on the location of the other described polyamine transporters in S. cerevisiae, i.e. whether Tpo2p, Tpo3p, and Tpo4p are located in the plasma membrane as well (4). We therefore generated three strains bearing 3xHA-tagged versions of TPO2, TPO3, and TPO4, respectively, as the only gene copy. Using sucrose density gradient centrifugation, we fractionated membrane preparations of these strains and found that all of the TPO gene products localize in the same fraction, which was identified as plasma membrane fraction before ( Fig. 7; compare Fig. 1). An involvement of the three yeast polyamine transporters in transport of these solutes across the plasma membrane seems to be likely and will be analyzed in the future.

Subcellular Localization of Tpo1p-4p
Revised-The expression and subsequent detection of epitope-tagged proteins has turned out to be a powerful method in unraveling the subcellular localization of proteins. By applying this approach we found that a triple HA-tagged version of Ycr023p co-localized with the vacuolar ATPase, using sucrose gradient centrifugation as the analytical method. Ycr023p is a member of the multidrug resistance family and most probably functions in the detoxification of hazardous compounds into the vacuolar lumen. It has been shown that its deletion confers resistance to allylglycine (25). Unexpectedly, triple HA-tagged versions of all four polyamine transporters, Tpo1p-4p, that have been described in S. cerevisiae so far co-localized with the plasma membrane ATPase. Despite the fact that tags in general might cause mislocalization of proteins, we decided to carefully reevaluate whether these transporters were in fact plasma membrane bound and thus not involved in vacuolar transport of polyamines as had been described before (4). Direct localization studies of these proteins had not been performed so far; their intracellular localization had been deduced indirectly from physiological experiments with Tpo1p being the best studied protein of this group (3,4,24). We provide experimental evidence for a role of Tpo1p in transport across the plasma membrane, which is based on the observations that the HA-tagged protein is functional and that a TPO1 knockout is not impaired in uptake of polyamines into vacuolar vesicles. The observation of Tomitori et al. (3) that vesicles prepared from TPO1-overexpressing strains show an increased spermine import activity is most probably due to mislocalization of the overproduced protein. In this study we could show that a strain deleted in TPO1 shows a significant defect in spermidine secretion under conditions in which the cells are challenged by incubation in medium containing high levels of polyamines. We thus conclude that Tpo1p is a plasma membrane-embedded carrier protein.
The arguments that Tpo2p, Tpo3p, and Tpo4p have the same subcellular location are so far exclusively based on results of FIG. 6. TPO1 disruption leads to impaired release of spermidine into the medium. Wild type and tpo1::lacZ cells were grown in succinate-buffered CBS medium in the presence of 10 mM spermidine. After resuspension in amine-free medium, spermidine efflux was followed by reverse phase HPLC. A, time course of spermidine efflux in wild type (q) and tpo1::lacZ cells (E). The graph shows data from three independent experiments. The error bars represent standard deviation. B, polyamine content of wild type cells grown in the absence of added spermidine (filled bars) as well as of wild type cells (light gray bars) and tpo1::lacZ cells (dark gray bars) grown in the presence of 10 mM spermidine. The bars represent the mean value from three independent cell cultures. C, initial spermidine efflux rates of wild type, of tpo1::lacZ cells, and of the same TPO1 disruption strain transformed with a plasmid expressing TPO1 under control of the PMA1 promoter (pDR199 TPO1). In the case of the two former strains, the bars represent the means from 15 independent experiment. The spermidine efflux rate of the mutant strain yeast was determined four times. The error bars represent standard deviation. sucrose gradient fractionation and need further experimental proof.
Polyamine Excretion in S. cerevisiae-Export of polyamines, especially of putrescine as the key metabolite in polyamine biosynthesis, had been described for some prokaryotic (22) and eukaryotic organisms (23), as well as for cultured mammalian cells (26). The biological significance of this process is most probably related to the fact that although polyamines are pivotal for many physiological processes, they are toxic when present in high levels, so that cells had to develop mechanisms for controlling the intracellular pools (27). Although the regulation of biosynthesis is certainly a possibility to control nontoxic levels (28), excretion provides an additional rationale. Taking into account that re-uptake is guaranteed by the presence of importer systems, a possible loss of precursors can be avoided. Putrescine export had not been studied in detail in S. cerevisiae before. When trying to establish a function of Tpo1p in polyamine export, we therefore asked whether yeast had developed mechanisms for putrescine homoeostasis similar to other microorganisms and whether Tpo1p might be involved in that process. Although we were able to detect significant putrescine export which, interestingly, was restricted to the fermentative growth phase, Tpo1p is obviously not involved in this process to a significant extent, because wild type and the TPO1 disruption strain had equal export activity. We cannot rule out, however, that redundant transporters may adjust their putrescine export activities in the deletion strain to compensate for the missing TPO1 function.
Polyamine Detoxification-When cells are stressed under conditions of growth in the presence of high levels of polyamines, a detoxification mechanism has to exist to overcome toxic intracellular levels that may arise because of the presence of polyamine importers. This could in principle be achieved by regulating the uptake activity and/or the activity of exporters for these solutes. We therefore preloaded yeast cells with polyamines and followed the efflux after transfer of cells into polyamine-free medium. Tpo1p was shown not to contribute significantly to putrescine tolerance (3); thus we concentrated on the analysis of spermidine export, which was in fact affected by TPO1 deletion. The decrease in excretion activity was significant and could be reverted by introducing a plasmid-encoded copy of TPO1, proving that the phenotype observed was a direct consequence of the missing TPO1 gene. The rescued strain showed spermidine export rates comparable with the wild type (Fig. 6C), which was also the reason for the polyamine tolerance that we found when testing the functionality of the TPO1-3xHA construct (Fig. 2). In the latter experiment we even observed an increased tolerance of the tpo1::lacZ strain containing TPO1-3xHA on a plasmid, compared with wild type, which was most probably due to the experimental conditions applied. Spermidine tolerance was assayed under magnesium limitation, a condition that leads to enhanced polyamine uptake (9), thus generating higher intracellular levels than the 4-fold increase we observed when challenging the cells for export measurements. Furthermore, it has been shown that polyamine stress in combination with Mg 2ϩ limitation leads to increased TPO1 mRNA abundance, even in a TPO1-overexpressing strain (4). This might explain the overcompensation of spermidine sensitivity observed with the plate assay in contrast to our observation in the spermidine export measurements, which were performed under Mg 2ϩ excess. Unfortunately, we were not able to provide appropriate growth medium completely devoid of a contaminant, which interfered with spermine analysis by HPLC. The contribution of Tpo1p in spermine export thus remains an unsettled question.
TPO1 Is Related to Multidrug Resistance-The data presented so far favor a function of Tpo1p in detoxification of high intracellular polyamine levels. This observation is interesting in the context of other recent publications on the properties of Tpo1p. This transporter has been reported to confer resistance to a variety of structurally nonrelated toxic compounds like quinidine and cycloheximide (24), mycophenolic acid (29), or 2-methyl-4-chlorophenoxyacetic acid and 2,4-dichlorophenoxyacetic acid (30), suggesting an broad substrate specificity, which is characteristic for multidrug resistance proteins (5). Furthermore, TPO1 was recently shown to be a target for the regulators Pdr1p (24) and Pdr3p (30), both of which mediate multidrug resistance. This further emphasizes the concept that the primary function of Tpo1p in S. cerevisiae is the detoxification of hazardous compounds, which includes excess spermidine.