Excretion of Putrescine and Spermidine by the Protein Encoded by YKL174c (TPO5) in Saccharomyces cerevisiae*

The properties of the protein encoded by YKL174c (TPO5) were studied. It was found that TPO5 excretes putrescine effectively and spermidine less effectively. γ-Aminobutyric acid slightly inhibited the excretion of putrescine, but basic amino acids did not affect excretion, suggesting that TPO5 preferentially recognizes polyamines. Accordingly, yeast cells transformed with the plasmid encoding YKL174c (TPO5) were resistant to toxicity caused by 120 mm putrescine or by 3 mm spermidine, and a mutant with a disrupted YKL174c (TPO5) gene was sensitive to toxicity by 90 mm putrescine. The growth of this mutant was faster than that of the wild-type strain. In parallel, there was an increase in putrescine and spermidine content of the YKL174c (TPO5) mutant compared with wild-type. It is noted that TPO5 functions as a suppressor of cell growth by excreting polyamines. The level of YKL174c (TPO5) mRNA was increased by the addition of polyamines to the medium. The degree of induction of the mRNA was spermine > spermidine > putrescine. The subcellular localization of TPO5 was determined by immunostaining of hemagglutinin-tagged TPO5, and it was found on Golgi or post-Golgi secretory vesicles. Excretion of putrescine and spermidine by TPO5 was reduced in cells that have mutations in the secretory or endocytic pathways, indicating that both processes are involved in the excretion of polyamines.

Polyamines, aliphatic cations present in almost all living organisms, are necessary for normal cell growth (1,2). Intracellular polyamine levels are elaborately regulated by biosynthesis, degradation, and transport (3). With regard to transport, the properties of three polyamine transport systems were characterized in Escherichia coli (4 -6). They include spermidine-preferential and putrescine-specific uptake systems, which belong to ATP-binding cassette transporters, and a protein, PotE, involved in the excretion of putrescine by a putrescine-ornithine antiporter activity. In Saccharomyces cerevisiae, we identified four genes which encode polyamine transport proteins TPO1-TPO4 (7,8). Among the four polyamine transporters, those encoded by TPO2 and TPO3 were specific for spermine, whereas those encoded by TPO1 and TPO4 recognized putrescine, spermidine, and spermine. Furthermore, we recently reported that UGA4, which catalyzes the transport of ␥-aminobutyric acid (9), is located on vacuoles and also catalyzes the transport of putrescine into vacuoles (10). UGA4 is classified into the family of amino acid-polyamineorganocation transporters (11). The name "polyamine" of amino acid-polyamine-organocation transporters is originated from polyamine transporter PotE. Among the family, there is only one gene (YKL174c) whose function has not yet been identified. In this study, we tried to clarify properties of the protein encoded by YKL174c. We found that YKL174c excretes putrescine and spermidine from cells and that the protein is located on Golgi or post-Golgi secretory vesicles. Thus, we termed the YKL174c gene as TPO5, which is involved in transport of polyamines.

EXPERIMENTAL PROCEDURES
Plasmids-For construction of YEpYKL174c, the gene for the YKL174c open reading frame and its upstream region was amplified by PCR from yeast X2180-1A (MATa SUC2 mal0 gal2 CUP1) genomic DNA as template, using primers HindIII-YKL174cF (5Ј-CCCAAGCT-TCAATGACACCAACATTTATTATCC-3Ј) and BamHI-YKL174cR (5Ј-CGCGGATCCCATCATCATCAAGGAGAAGG-3Ј). The resulting 2.5-kb fragment was digested with HindIII and BamHI, and inserted into the same restriction sites of the plasmid YEp352 (12). For construction of YEpYKL174c-HA 3 , which encodes five glycine residues and three copies of Haemophilus influenzae HA 1 epitope at the C terminus of YKL174c lacking the termination codon, PCR was performed as described above using primers HindIII-YKL174cF and SalI-YKL174c-StopR (5Ј-TA-ACGCGTCGACTATATCATATCTACGATCATCGGCA-3Ј). The product was digested with HindIII and SalI and inserted into the same sites of YEpUGA4-HA 3 (10). Plasmids were introduced into yeast cells by the lithium acetate method of Ito et al. (13).
Polyamine Transport Assay-Yeast cells in a 20-ml culture were harvested during the exponential growth phase (A 540 ϭ 0.5), washed three times with 20 mM Na-Hepes buffer (pH 7.2) containing 10 mM glucose, suspended at 0.2 mg protein/ml in the same buffer, and incubated at 30°C for 5 min. The reaction was started by the addition of labeled substrates. After incubation for 20, 40, or 60 min, a 0.5-ml aliquot of reaction mixture was filtered through cellulose acetate filter (pore size 0.45 m, Advantec) and washed twice with the buffer described above containing a 10-fold concentration of non-labeled substrates. The radioactivity of the filters was counted in a liquid scintillation counter. The concentrations and specific activities of substrates were as follows: 0. After cells were washed as described above, [ 14 C]putrescineloaded cells were incubated in the buffer containing 20 mM Na-MES buffer, pH 5.0, and 10 mM glucose. After incubation for 30, 60, 90, or 120 min at 30°C, a 0.5-ml aliquot of the reaction mixture was centrifuged (17,000 ϫ g, for 5 min at 4°C), and the radioactivity of the supernatant was counted in a liquid scintillation counter. The protein concentration was determined by the method of Bradford (21).
Measurement of Polyamine Contents in Whole Cells-Yeast cells were grown in the presence and absence of putrescine in 5 ml of medium and harvested at the exponential growth phase (A 540 ϭ 0.5). Polyamines were extracted by the treatment with 10% (w/v) trichloroacetic acid at 70°C for 1 h with occasional shaking. Polyamine contents were determined by high pressure liquid chromatography as described previously (22). Protein was determined by the method of Bradford (21).
Indirect Immunofluorescence Microscopy-This was carried out according to the methods of Pringle et al. (23) using mouse monoclonal anti-HA as a primary antibody and goat anti-mouse IgG (HϩL) antibody conjugated with Alexa Fluor® 488 as a secondary antibody, or using rabbit anti-SEC7 as a primary antibody and goat anti-rabbit IgG (HϩL) antibody conjugated with Alexa Fluor® 546 as a second antibody. For the staining of DNA, fixed and permeabilized cells were treated with 200 g/ml RNase A for 1 h and stained with 100 g/ml propidium iodide for 1 h. The slides were immersed in ProLong antifade solution (Molecular Probes) and fluorescence was analyzed with a confocal microscope (ZEISS, LSM510 Laser Scanning Microscope) with an argon laser ( ex ϭ 488 nm and em ϭ 505 nm) for YKL174c-HA 3 or with HeNe laser ( ex ϭ 543 nm and em ϭ 560 nm) for SEC7 and DNA.
Subcellular Fractionation and Western Blot Analysis-Yeast cells carrying the plasmid encoding the HA-tagged YKL174c in a 600-ml culture were harvested at A 540 ϭ 0.5. The subcellular distribution of HA-tagged YKL174c protein was determined according to the method of Sorin et al. (24). Briefly, after incubation with 40 g/ml zymolyase 20T (Seikagaku Corporation, Tokyo) at 30°C for 30 min, cells were lysed in a hypotonic buffer (0.3 M sorbitol, 20 mM Tris acetate, pH 7.2, 1 mM EDTA, 10 M E64C, a thiol proteinase inhibitor (Peptide Institute, Inc.), and 20 M FUT-175, a serine protease inhibitor (25)) and fractionated by a discontinuous sucrose gradient ranging from 18 to 54% (w/v) in 10 steps of 4% difference each. The fractionated membrane was suspended in 100 l of a SDS sample buffer (25 mM Tris-HCl, pH 6.8, 5% glycerol, 1% sodium dodecyl sulfate, 0.2% sodium deoxycholic acid, and 0.05% bromphenol blue), and a 50-l aliquot of samples was separated on a 8.5% SDS-polyacrylamide gel. After proteins were blotted onto a polyvinylidene fluoride membrane (Immobilon P, Millipore), HA-tagged TPO5, SEC7 (a marker of Golgi complex), VPH1 (vacuolar H ϩ -ATPase) and PMA1 (plasma membrane ATPase) were detected using ECL Western blotting analysis system (Amersham Biosciences) according to the manufacturer's protocol.
Northern Blot Analysis of YKL174c mRNA-Yeast cells in 100-ml culture were harvested at A 540 ϭ 0.5, washed with a buffer containing 10 mM Tris-HCl (pH 7.5) and 1 M sorbitol, and then resuspended in the same buffer at 5 ϫ 10 8 cells/ml. After incubation with 40 g/ml zymolyase 20T (Seikagaku Corporation, Tokyo) at 30°C for 30 min, RNA was extracted from the spheroplast with an RNAqueous-MiDi Kit (Ambion).
Preparation of Phylogenetic Tree-The phylogenetic tree was calculated from profile-derived multiple alignments by the ClustalW V1.8 program including bootstrapping (27) with the use of the neighborjoining algorithm (28). Fig. 1 shows a phylogenetic tree of the family of amino acid-polyamine-organocation transporters in yeast. Among these proteins, there is one, termed YKL174c, whose function is unknown but which is highly homologous with UGA4, a ␥-aminobutyric acid and putrescine transport protein on the vacuolar membrane (10). In light of this, we carried out experiments to determine whether YKL174c has polyamine transport activity. We looked for activity using YPH499 cells transformed with YEpYKL174c (or the vector YEp352 for control cells) and cells in which the native YKL174c gene is disrupted (⌬ykl174c). As shown in Fig. 2, uptake of putrescine and spermidine was lower in YPH499/YEpYKL174c cells than in control YPH499/YEp352 cells. Uptake of spermine was not affected by transformation with YEpYKL174c. Furthermore, uptake of putrescine was enhanced in ⌬ykl174c cells compared with wild type (Fig. 2B). The results suggest that YKL174c catalyzes the excretion of putrescine and, to a lesser degree, of spermidine. Thus, the YKL174c gene is hereafter termed TPO5. Excretion of putrescine by TPO5 was examined directly by using cells preloaded with [ 14 C]putrescine. As shown in Fig. 3, the excretion of putrescine was enhanced by TPO5; that is, excretion of [ 14 C]putrescine was faster in wild-type cells than ⌬ykl174c cells. The excretion of paraquat, a polyamine analogue (8), was not influenced by TPO5 (data not shown). The results indicate that TPO5 preferentially excretes putrescine. The increased putrescine uptake seen in cells with a disrupted YKL174c gene (equal to putrescine excretion activity by TPO5) was slightly inhibited by non-labeled ␥-aminobutyric acid (1 mM), but not by lysine, arginine, histidine, and ornithine (Fig. 4), indicating that TPO5 recognizes polyamines preferentially.

Excretion of Putrescine and Spermidine by YKL174c (TPO5)-
The effects of TPO5 on polyamine toxicity were then measured using cells transformed with YEpYKL174c (TPO5). As shown in Fig. 5, cells transformed with YEpTPO5 grew faster than control cells (containing the vector) in the presence of 120 mM putrescine or 3 mM spermidine. In addition, the ⌬ykl174c mutant showed strong sensitivity to 90 mM putrescine (Fig. 5) but showed no difference in sensitivity to spermidine compared with the wild-type strain (data not shown). The results are in accordance with the finding that spermidine is not effectively effluxed. An effect of polyamines on cell growth was clearly observed, because the time course for cell growth was followed for more than 150 h, whereas polyamine uptake activity was followed for only 60 or 120 min.
It should be noted that the ⌬ykl174c mutant grows faster than the wild-type strain in the absence of polyamines (Fig.  5B). Such a growth difference was observed even in the presence of 0.3 or 1 mM putrescine (Fig. 6A). Under these conditions, cellular polyamine content was measured (Fig. 6B). Levels of polyamines in the ⌬ykl174c mutant cultured without putrescine increased from 2.97 and 30.4 (wild-type) to 21.5 and 60.0 nmol/mg of protein (mutant) for putrescine and spermidine, respectively. The amount of putrescine and spermidine accumulated in the ⌬ykl174c mutant corresponds to 1.84 and 5.13 mM, respectively, if the intracellular water space is estimated as 11.7 l of cell volume/mg of protein (15). Even if 0.3 or 1 mM putrescine was added to the medium, the levels of putrescine and (for 0.3 mM putrescine) spermidine were still higher in the ⌬ykl174c mutant than in wild-type. These results suggest that wild-type yeast having TPO5 gene do not maintain an optimal level of polyamines necessary for cell growth under these conditions. Subcellular Localization of TPO5-The subcellular localization of TPO5 was determined by indirect immunofluorescence microscopy using HA-tagged TPO5 and an anti-HA antibody. As shown in Fig. 7A, TPO5 was located on small vesicles existing in the vicinity of plasma membrane. The vesicles were different from vacuoles, which were observed by differential interference contrast and from nuclei judged by DNA staining with propidium iodide. Accordingly, colocalization of TPO5 with SEC7, a marker of the Golgi complex (20), was examined. As shown in Fig. 7B, most of TPO5 was colocalized with SEC7, indicating that TPO5 is located on Golgi or post-Golgi secretory vesicles. Subcellular localization of TPO5 was then analyzed by sucrose density gradient centrifugation. As shown in Fig. 8, most of TPO5-HA 3 was localized in the low density position similar to Golgi complex, estimated by SEC7, and vacuoles, estimated by VPH1 (a subunit of vacuolar proton ATPase), and very small amount of TPO5 was localized in the high density position similar to plasma membrane, estimated by PMA1 (plasma membrane ATPase 1). Some portion of SEC7 was also located in the high density portion. This was probably because of the coagulation of SEC7 during the preparation of cell lysate, because SEC7 was not observed on plasma membrane (see Fig.  7). These results are in accordance with an idea that TPO5 is located mainly on Golgi or post-Golgi secretory vesicles.
To confirm the location of TPO5 on Golgi or post-Golgi secretory vesicles, polyamine transport activity was measured using a temperature sensitive mutant deficient in the process of secretion (sec6) (17) and a mutant deficient in endocytosis (⌬end4) (14). In the sec6 mutant, fusion of the secretory vesicles with the plasma membrane during the final exocytosis process is inhibited at non-permissive temperatures (17), and endocytosis of a zinc transporter ZRT1 is deficient in ⌬end4 mutant (14). As shown in Fig. 9A, putrescine and spermidine excretion activity was lower with a sec6 mutant containing YEpYKL174c (TPO5) than with its parent strain containing YEpYKL174c (TPO5) at a non-permissive temperature (37°C), but the activity of these two strains containing YEpYKL174c (TPO5) was nearly equal at a permissive temperature (23°C). Excretion activity of putrescine and spermidine was also lower with the endocytosis mutant containing YEpYKL174c (TPO5) than its parent strain containing YEpYKL174c (TPO5) (Fig. 9B). These results indicate that excretion of putrescine and spermidine by TPO5 was not observed in the exocytosis and endocytosis mutants and that both exocytosis of polyamines by TPO5 and endocytosis of TPO5 for recycling are involved in the activity of TPO5.
Induction of TPO5 mRNA by Polyamines-We next tested whether TPO5 mRNA is induced by polyamines. Northern blot analysis was carried out using total RNA isolated from YPH499/YEpYKL174c cultured in the absence and presence of polyamines. As shown in Fig. 10, TPO5 mRNA was most strongly induced by 0.2 mM spermine, even though spermine is not a substrate for TPO5. Spermidine and putrescine also caused induction of TPO5 mRNA. As a control, the level of ACT1 mRNA encoding actin 1 was measured. It was not influenced by polyamines. Because the effective concentration is in the order putrescine Ͼ spermidine Ͼ spermine in terms of the stimulation of protein synthesis (29), it is thought that induction of TPO5 mRNA depends on the effective polyamine concentration in cells. DISCUSSION We studied the function of a protein encoded by an unidentified gene YKL174c, which has high homology with UGA4, a member of the family of amino acid-polyamine-organocation transporters in S. cerevisiae. We found that YKL174c catalyzes the excretion of putrescine and spermidine, and thus termed this protein as TPO5. The homology of an amino acid sequence between TPO5 and UGA4 was 42%, and the major difference was observed in the C-terminal regions of the two proteins. Thus, the cellular localization of these two proteins (TPO5 is found on Golgi or post-Golgi secretory vesicles, and UGA4 is found on vacuoles) may be determined by the C termini of these proteins.
Our data clearly indicate that TPO5 is mainly located on Golgi or post-Golgi secretory vesicles and that both processes of exocytosis and endocytosis are involved in the secretion of putrescine and spermidine. This is the first report that shows the necessity of these two processes for secretion of small molecules in yeast. In yeast, it has been reported that the process of exocytosis is necessary for the sorting of tryptophan permease to the plasma membrane (17), and also the process of endocytosis is necessary for degradation in vacuoles of the zinc transporter ZRT1 (14), a general amino acid permease (30) and a uracil permease (31) on plasma membrane. However, there is no report thus far stating that both processes are necessary for the function of a specific protein. It has been reported that both exocytosis and endocytosis are necessary for neurotransmitter release from synaptic vesicles in mammalian cells (32). At present, we do not know the exact reason why TPO5 mainly exists on Golgi or post-Golgi secretory vesicles. It may be better for TPO5 to exist on Golgi or post-Golgi secretory vesicles to excrete polyamines more effectively once polyamines are accumulated in secretory vesicles like neurotransmitters accumulated in synaptic vesicles.
Polyamine transporters previously characterized were not specific for polyamines in S. cerevisiae. TPO1 can excrete other substances that are not related to polyamines (33)(34)(35), and UGA4 has higher affinity for ␥-aminobutyric acid rather than putrescine (10). Thus, TPO5 is the only protein studied to date that recognizes putrescine and spermidine preferentially. Because TPO2 and TPO3 only recognize spermine among polyamines, experiments are in progress to test whether TPO2 and TPO3 specifically recognize spermine. The existence of spermine (or acetylspermine) oxidase has been reported recently in yeast (36). If the enzyme also catalyzes the conversion of spermidine to putrescine, excess polyamines may be excreted effectively by TPO5.
It is surprising that cells with a disrupted TPO5 gene (⌬ykl174c) grow more rapidly than wild-type cells. Under these conditions, the accumulation of putrescine and spermidine in ⌬ykl174c cells was observed. The results suggest that the existence of TPO5 slows down cell growth through excretion of putrescine. It has been reported that excess spermidine inhibits protein synthesis by binding to ribosomes in E. coli (37). Acetylation of polyamines decreases their ability to stimulate protein synthesis and their toxicity (38). Acetyltransferases of polyamines are present in both mammalian cells and E. coli (1). However, the existence of acetyltransferase(s) of polyamines has not been reported in S. cerevisiae, although the whole genome sequence was determined. TPO5 together with spermine oxidase (36) may function for detoxification of polyamines in yeast.
When cells were cultured in the presence of high concentrations of putrescine (120 mM) or spermidine (3 mM), it took 50 -100 h until reaching the logarithmic phase of cell growth (see Fig. 5). It may take a long time to induce the expression of the YKL174c (TPO5) gene under these experimental conditions. Induction of TPO5 mRNA was measured in the presence of 50 mM putrescine or 2 mM spermidine (Fig. 10). Under these conditions, the lag time was within 24 h, and the difference in cell growth between cells transformed with YEpYKL174c (TPO5) and YEp352 in the presence of polyamines was not clearly observed (data not shown). FIG. 7. Subcellular localization of TPO5 determined by immunofluorescence microscopy. Yeast BY4741 cells carrying the plasmid encoding HA-tagged YKL174c (TPO5) gene were cultured in synthetic complete medium and harvested at A 540 ϭ 0.5. Indirect immunofluorescence microscopy was performed as described under "Experimental Procedures." A, differential interference contrast (DIC) and fluorescence images of HA-tagged TPO5 and DNA together with their merged images are shown. B, differential interference contrast and fluorescence images of HA-tagged TPO5 and SEC7 together with their merged image are shown. Scale bar ϭ 10 m. , and harvested at A 540 ϭ 0.5. Northern blot analysis of TPO5 and ACT1 mRNAs was performed as described under "Experimental Procedures." RNA from cells containing the vector YEp352 cultured in the absence of polyamines (Vector/None) was also analyzed. Similar results were obtained when cells were harvested at A 540 ϭ 0.3.