Studies on the regulation of ornithine decarboxylase in yeast: Effect of deletion in the MEU1 gene
- Laboratory of Biochemistry and Genetics, National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, Building 8, Room 223, Bethesda, MD 20892
-
Contributed by Herbert Tabor, August 23, 2005
Abstract
Methylthioadenosine is formed during the biosynthesis of spermidine and of spermine and is metabolized by methylthioadenosine phosphorylase, an enzyme missing in several tumor cell lines. In Saccharomyces cerevisiae, this enzyme is coded by the MEU1 gene. We have now studied the effect of the meu1 deletion on polyamine metabolism in yeast. We found that the effects of the meu1Δ mutation mostly depend on the stage of cell growth. As the cell density increases, there is a marked fall in the level of ornithine decarboxylase (ODC) in the MEU1 + cells, which we show is caused by an antizyme-requiring degradation system. In contrast, there is only a small decrease in the ODC level in the meu1Δ cells. The meu1Δ cells have a higher putrescine and a lower spermidine level than MEU1 + cells, suggesting that the decreased spermidine level in the meu1Δ cultures is responsible for the greater apparent stability of ODC in the meu1Δ cells. The lower spermidine level in the meu1Δ cells probably results from an inhibition of spermidine synthase by the methylthioadenosine that presumably accumulates in these mutants. In both MEU1 + and the meu1Δ cultures, the ODC levels were markedly decreased by the addition of spermidine to the media, and thus our results contradict the postulation of Subhi et al. [Subhi, A. L., et al. (2003) J. Biol. Chem. 278, 49868–49873] of a novel regulatory pathway in meu1Δ cells in which ODC is not responsive to spermidine.
Ornithine decarboxylase (ODC) is the key regulatory enzyme of the polyamine biosynthetic pathway and catalyzes the decarboxylation of ornithine to 1,4-diaminobutane (putrescine). Putrescine is then converted to spermidine and spermine by the addition of aminopropyl groups from decarboxylated S-adenosylmethionine (1). Methylthioadenosine is formed as a byproduct of these aminopropyl transferase reactions (spermidine synthase and spermine synthase) (2–4) and is then converted to adenine and to methionine by the “methionine salvage” pathway (5–9). The first step in the methionine salvage pathway is the phosphorolysis of methylthioadenosine to methylthioribose-1-phosphate and adenine by methylthioadenosine phosphorylase (EC 2.4.2.28). The metabolism of methylthioadenosine in mammalian cells has been of special interest because methylthioadenosine phosphorylase is absent in many tumors (10–15).
Thomas et al. (16) and Subhi et al. (17) have shown that, in Saccharomyces cerevisiae, methylthioadenosine phosphorylase is coded by the MEU1 gene and have described meu1Δ mutants. The availability of these yeast mutants deleted in the methylthioadenosine phosphorylase gene permits more extensive studies on the role of this gene and the methionine salvage pathway in polyamine metabolism in yeast, and a comparison of these effects in yeast with those reported in mammalian cells. Therefore, as an extension of our work on the regulation of ornithine decarboxylase and of polyamine pools in yeast (18), we have now studied the effect of the meu1Δ deletion on these parameters. Our findings were different from those reported by Subhi et al. (17) on the effects of this mutation, and data are presented on the probable reasons for some of these differences.
Materials and Methods
Strains, Media, and Growth. The yeast strains used in this study are listed in Table 1. All strains were maintained in YPAD (1% yeast extract/2% peptone/2% agar/2% dextrose) medium, except strains Y622 and Y623, which were grown on SD-Ura-Leu plates, with additions of the required supplements. For most of the experiments, SD medium [6.7 g of yeast nitrogen base/20 g of dextrose/0.75 g of complete supplement mixture, pH 5.2 (Qbiogene, Irvine, CA)] was used. Yeast cultures were inoculated from plate to liquid SD medium to make the primary inocula, and, after growth, these inocula were further diluted into fresh SD medium and incubated at 30°C with shaking for 16–24 h. Cultures were harvested at different optical densities and washed with PBS (for polyamine estimation) or with enzyme extraction buffer (for enzyme assays) and stored at –80°C. Optical density was measured at 600 nm in a Genesis2 spectrophotometer (Thermo Spectronic, Rochester, NY); an optical density of 1.0 represented 1 × 107 cells per ml.
For experiments with cycloheximide, cultures were grown to an OD600 of 1.5, and cycloheximide (final concentration 200 μg/ml culture) was added from a stock solution in ethanol.
Enzyme Assay and Protein Estimation. ODC was measured by using l-[1-14C]ornithine (Amersham Pharmacia, Cat. no. CFA491, 57 mCi/mmol) (1 Ci = 37 GBq) as a substrate (19, 20). The harvested cells were resuspended in cold extraction buffer [50 mM Tris·HCl, pH 8.0/1 mM DTT/0.1 mM EDTA/0.05% Tween 80/50 μM pyridoxal phosphate/1 mM phenylmethylsulfonyl fluoride (Sigma) and 1× “protease halt,” (Pierce)]. The suspensions were transferred to 2-ml screw cap tubes (Sarstedt) together with equal amounts of glass beads (425–600 μm, Sigma). The cultures were broken in a MiniBead Beater (Biospec Products, Bartlesville, OK) with two 1-min bursts, cooling the suspensions in ice between bursts. The cell lysates were centrifuged at 12,000 rpm in a Sorvall RC5C, SS34 rotor (DuPont) at 4°C for 20 min, and 10 μl of the clarified supernatant was used for each assay in duplicate in a total volume of 125 μl (50 mM Tris·HCl, pH 8.0/400 nCi of l-[1-14C]ornithine/100 μM cold ornithine/0.5 mg/ml BSA/50 μM pyridoxal phosphate/5 mM DTT). ODC activity was expressed as nmol of 14CO2 released from ornithine per mg of protein per h at 37°C.
Spermidine synthase was assayed with some modifications of the method reported previously (4). The yeast cultures were grown to OD600 of 0.8 and were washed with extraction buffer (50 mM Tris·HCl, pH 8.0/1 mM DTT/0.1 mM EDTA/1 mM phenylmethylsulfonyl fluoride). The protein was extracted as above. The spermidine synthase assay mixture contained 0.1 M Hepes-KOH (pH 7.3), 5 mM DTT, 50 nCi of [1,4-14C]putrescine (Amersham Pharmacia, Cat. no. CFA301, 108 mCi/mmol), 100 μM unlabeled putrescine, 100 μM decarboxylated S-adenosylmethionine (a gift from G. Stramentinoli, Bio Research Co., Research Laboratories, Liscate, Milan, Italy), and 3 μl of enzyme extract in a total volume of 10 μl. The mixture was incubated at 37°C for 30 min, and the reaction was terminated by adding 10 μl of ice cold 10% trichloroacetic acid. Five microliters of the trichloroacetic acid extract was spotted on thin layer chromatography (Polygram Ionex-25 SA-Na, Macherey-Nagel, Düren, Germany) along with radioactive standards for putrescine and spermidine. The chromatograms were developed with 2 M KCl and exposed to a Fuji BAS-MS imaging plate; after scanning, the [14C]spermidine formation was quantitated by image gauge software (Fuji). To test its inhibitory effect, methylthioadenosine at different concentrations was added in a total volume of 1 μl per 10 μl of reaction mixture, and incubated as above. Protein concentrations of the cell lysates were measured with the Bradford reagent (Bio-Rad), by using BSA as the standard (21).
Estimation of Polyamines. Polyamines were assayed by a modification of the HPLC chromatography procedure previously described (4). The amines in the cells were extracted with 5 ml of 10% perchloric acid per gram wet weight of cells. We applied 30–60 μl of the extract to a sulfonated cation exchange resin (A5-2806, St. John Associates, Beltsville, MD), packed in a 4.5 × 150-mm column, which was maintained at 70°C. The eluted amines were measured fluorometrically after reaction with o-phthaldehyde. The eluting buffer was modified to contain 112 g of KCl, 24 g of tripotassium citrate·H2O, 24 ml of 2 M HCl, and 29.2 g of NaCl per liter.
Western Blot Analysis. The same protein extracts (20 μg) that were used for ODC assays were fractionated on 10% NuPAGE Novex Bis-Tris SDS gel (Invitrogen) in a minigel apparatus (Invitrogen). Immunoblotting was performed on poly(vinylidene difluoride) (PVDF) membrane (Invitrogen). Immunodetection was carried out by using an antibody (1:5,000 dilution) raised against recombinant yeast ODC protein (18). Immunospecificity of the antibody was enhanced by prior adsorption of the antibody with an spe1Δ extract that does not contain ODC protein. For chemiluminescence assay, Western Breez immunodetection kit (Invitrogen) was used, and the immunoreactive bands were detected and scanned with a Bio-Rad Scanner (GS-800).
Results
meu1 Deletion Does Not Affect the Growth Rate. We have measured the growth rate of four different isogenic meu1Δ and MEU1 + pairs of strains and have found essentially no effect of the meu1Δ on the growth rate. The results are presented along with polyamine content in Table 2.
Polyamine content of different pairs of isogenic strains of MEU1+ and meu1Δ cells harvested at 0.4 OD 600
Effect of meu1 Deletion on ODC Activity and Polyamine Levels in Logarithmically Growing Cells. As shown in Fig. 1, we found a small effect of the meu1 deletion on the amount of ODC in early logarithmically growing cells (OD = 0.4) when measured either by assay of ODC activity or by Western blot. With either assay, the ODC level of the meu1Δ cells was 30–60% higher than that found with the MEU1 + cells. The same results were obtained with four different isogenic pairs of MEU1 + and meu1Δ cultures (Fig. 1, Sets A–D). Set C contains the results we obtained with the strains used by Subhi et al. (17) and shows a much smaller increase in the ODC level in the meu1Δ strain than the 800% increase reported by these investigators.
Level of ODC in logarithmically growing MEU1 + and meu1Δ cultures (OD = 0.4). Each set represents a different pair of Meu1 + and meu1Δ isogenic strains. (Upper) The assays for ODC activity expressed in nmol of 14CO2 released per mg of protein per h. (Lower) The assays for ODC protein (Western blot) with the same extracts (20 μg).
Polyamine assays on perchloric acid extracts of the above cultures (Table 2) showed a 3- to 5-fold elevation in the putrescine level and a small decrease in the spermidine level, which resulted in an increase in the putrescine to spermidine ratios in meu1Δ cells compared with MEU1 + cells.
Effects of Cell Density on ODC Levels of MEU1+ and meu1Δ Cultures. Because Subhi et al. (17) might have used cells from an overgrown culture, we then repeated the ODC assays with cultures grown to higher optical densities to see whether the cell concentration might account for the differences between our results and theirs. As shown in Fig. 2A, when the ODC measurements were carried out on extracts of cells harvested at higher optical densities, the ODC levels were lower in both the MEU1 + and meu1Δ cells but the decrease was much less pronounced in the meu1Δ cultures. When the measurements were made on nearly stationary phase cultures (OD600 1.5–2.0), the MEU1 + cells showed an 80–90% decrease in activity, whereas the meu1Δ cells showed only a 30–40% decrease in ODC activity. Thus, when measured in overgrown cultures, the residual activity in the meu1Δ cells was 3- to 4-fold higher than found in the MEU1 + cultures. Western blot analysis showed a similar effect of cell density, and indicated that the decrease in ODC was due to a decrease in the ODC protein (Fig. 2B). Completely stationary phase cultures had very little ODC activity both in MEU1 + and meu1Δ cells (OD = 4.4, Fig. 2 A).
Effect of increase in optical density of the cultures on ODC activity (A) and ODC protein (Western blot) (B). For these experiments, Y610 (MEU1 +) and Y607 (meu1Δ) cells were used. The result with an spe1Δ extract (rightmost lane, Western blot) shows the specificity of the ODC antibody.
The recent availability of an oaz1Δ mutant that lacks antizyme (22) permitted us to study whether the loss of ODC activity in cultures grown to a high cell density was the result of an effect on enzyme synthesis or of enzyme degradation via the well-established antizyme system (22–27). For this purpose, we repeated the above experiment on the effect of cell density on ODC activity and protein level in a MEU1 + oaz1Δ strain (isogenic with Y610, except for the oaz1 deletion). As shown in Fig. 3, the cell density did not affect the level of ODC in this strain, indicating that the loss of ODC seen in Fig. 2 was due to the action of the antizyme-mediated degradation system of ODC.
Cell density does not affect the ODC activity or ODC protein (Western blot) in the absence of antizyme (MEU1 + oaz1Δ strain, Y651).
ODC Is More Stable in meu1Δ Cultures than in MEU1+ Cultures. The above experiments (Fig. 2) showed that the ODC level in the meu1Δ cultures did not decrease as rapidly as the enzyme level in the MEU1 + cultures, as the cell density increased. To test whether this difference in the behavior of the two strains was the result of increased stability of the ODC in the meu1Δ cultures, we treated both MEU1 + and meu1Δ cultures (OD = 1.5) with cycloheximide to inhibit protein synthesis, and followed the decay in ODC. We chose the culture with an OD600 of 1.5, because we found that, at this optical density, the ODC activities of MEU1+ and meu1Δ cells showed the largest difference (Fig. 2 A). As shown in Fig. 4, in MEU1 + cultures, ODC decreased with a half-life of 1.5 h. In contrast in meu1Δ cells, there was almost no decrease in ODC activity or protein in 3 hours.
ODC is more stable in meu1Δ (Y607) cells than in MEU1 + (Y610) cells. ODC was measured in MEU1 + and meu1Δ cultures after cycloheximide (CHX) treatment. Cultures were grown to OD600 of 1.5, and, at time zero, CHX was added at 200 μg/ml. Aliquots of the cultures were harvested at 1-hour intervals and assayed for ODC activity (A) and ODC protein (Western blot) (B).
Because we know that the level of ODC is markedly affected by the spermidine concentration in the yeast cell (18, 22, 28, 29), we next determined the level of polyamines in the meu1Δ and MEU1 + cultures grown to the same cell density as in the above experiment. These assays (Fig. 5) showed that the spermidine level was at least 50% less in meu1Δ cells than in MEU1 + cells and showed a 3.5-fold increase in putrescine content in meu1Δ cells compared with MEU1 + cells.
Polyamine levels of MEU1 + (Y610) and meu1Δ (Y607) cultures, harvested at an OD600 = 1.5.
We then tested whether the higher spermidine level in MEU1 + cells was sufficient to explain the decreased level of ODC seen in these cells compared with the meu1Δ cells. We found that this was indeed the case. Thus, if the meu1Δ cells were grown in 5 × 10–6 M spermidine, the spermidine level of the cells increased from 1.4 to 2.6 nanomoles per milligram wet weight, and this increase in the intracellular spermidine level resulted in a 90% decrease in ODC and a 91% decrease in the putrescine level.
The decrease in spermidine concentration found in the meu1Δ cells could have been due to inhibition of spermidine synthase by methylthioadenosine as has been described for spermidine synthase from other sources (30–32). By using crude extracts from MEU1 + and meu1Δ cultures, we found that this was indeed the case (Fig. 6). Even 0.0025 mM methylthioadenosine showed 30–40% inhibition; higher concentrations (0.25–2.5 mM) caused nearly complete inhibition of spermidine synthase activity.
Effect of methylthioadenosine on spermidine synthase activity. Spermidine synthase assays were carried out in extracts of MEU1 + (Y610) and meu1Δ (607) by using [14C]putrescine and decarboxylated S-adenosylmethionine in the presence or absence of different concentrations of methylthioadenosine in vitro. Formation of [14C]spermidine from [14C]putrescine was measured, and 100% activity was considered as the amount of spermidine formed in the absence of methylthioadenosine.
Spermidine Addition Causes a Rapid Decrease in ODC Levels in both MEU1+ and meu1Δ Cells in both Logarithmically Growing and Overgrown Cultures. In contrast to the results reported by Subhi et al. (17), we found that spermidine addition to the cultures caused a rapid fall in the ODC activities in both the MEU1 + and the meu1Δ cells (Fig. 7). Incubation of logarithmically growing cultures (final OD = 0.4) with 100 μM of spermidine for 4 h resulted in 95% decrease of ODC activity in both strains. A similar decrease in ODC activity was found when spermidine was added to a more dense culture (final OD = 2.5). The decrease in enzyme activity was due to a decrease in ODC protein as measured by Western blot (result from a logarithmically growing culture is shown in Fig. 7 Lower).
ODC is markedly decreased by addition of spermidine to both MEU1 + (Y610) and meu1Δ (Y607) cultures. At time zero (OD600 = 0.1 or OD600 = 1.0), yeast cultures were divided into two parts, and spermidine at 100 μM final concentration was added in one part. Samples were harvested after 4 h and assayed for ODC. (Upper) In culture A, the initial OD = 0.1 and the final OD = 0.4. In culture B, the initial OD = 1.0 and the final OD = 2.5. (Lower) Western blot on extracts from culture A.
Effect of 4-Methylthio-2-Oxobutanoic Acid (MTOB) Addition on the Level of ODC. We have found a difference in the effect of MTOB when tested in MET15 + or met15Δ strains. When tested with MET15 + strains, we did not find any repression of ODC by either 1 mM or 0.1 mM MTOB in either MEU1 + (Y610) or meu1Δ (Y607) cultures (data not shown).
However, we did find a small effect of MTOB when added for 4 h to met15Δ cells. These cells were grown with methionine and washed twice in methionine-free medium before incubation with MTOB. When incubated in the absence of methionine, additions of 0.1 mM or 1 mM MTOB for 4 h or 20 h caused a 30–40% inhibition of ODC in MEU1 + cells (Y534) and 20–25% inhibition in meu1Δ cells (Y606). Additions of MTOB to cells supplemented with methionine showed no effect of MTOB. Interpretation of these results is complicated by slower growth rate of cells supplemented with MTOB alone compared with cells supplemented with both methionine and MTOB. These results are thus somewhat different from the findings of Subhi et al. (17) of a complete repression of ODC in meu1Δ met15Δ cells by MTOB. In contrast to the results with MTOB, we found that spermidine added to a culture for 4 h showed 90–95% inhibition of ODC level (Fig. 8).
Effect of MTOB on ODC. MEU1 + (Y534) and meu1Δ (Y606) methionine auxotrophic yeast cells were grown to OD600 of 1.0 in the presence of methionine, washed twice, and diluted in the presence of either methionine, or methionine plus 100 μM spermidine, or 100 μM MTOB, or 1 mM MTOB in the presence or absence of methionine, and harvested after 4 h. Error bars indicate the variations of three individual experiments assayed in duplicate.
Discussion
The first step in the methionine salvage pathway is carried out by methylthioadenosine phosphorylase. The role of methylthioadenosine and of methylthioadenosine phosphorylase in controlling growth and of polyamine biosynthesis has been of particular interest because a number of human and murine tumors have been associated with the loss of this gene (10–15). Overexpression of methylthioadenosine phosphorylase in some tumor cell lines has been reported to result in tumor suppression (33), and addition of methylthioadenosine to animal cell culture has been reported to increase ODC and putrescine levels (34). The gene for methylthioadenosine phosphorylase has been identified in several species; in humans it is located on chromosome 9p21 (35). In S. cerevisiae, it is designated as MEU1 and is located on chromosome XII (36).
In our current work, we have used yeast strains containing the meu1Δ mutation to study the role of methylthioadenosine and the methionine salvage pathway on ODC and on polyamine regulation in yeast, and to compare our results with those reported by Subhi et al. (17). We have found a number of differences in our results and the results presented by Subhi et al. (17) on the effect of the meu1Δ mutation in yeast. The major differences are as follows.
First, Subhi et al. (17) found an 800% increase in ODC activity in the meu1Δ cells compared with MEU1 + cells whereas we found only less than a 30–50% increase (Fig. 1). We attribute this difference to our use of cells harvested in early logarithmic growth. We have found that, with increased cell density, the ODC level of wild-type (MEU1 +) cells falls markedly (Fig. 2) as first reported by Kay et al. (37), and that this decrease does not occur if the strain lacks antizyme activity (Fig. 3). This finding proves that the decrease is due to degradation of the enzyme by ODC antizyme system and not due to inhibition of ODC synthesis. In contrast, we found that there is a much smaller decline in the ODC levels in the meu1Δ cells during increase in optical density. Our studies with cycloheximide confirmed the greater stability of the ODC in the meu1Δ cultures (Fig. 4). As a result in cultures with a higher optical density, we have found a 4-fold higher ODC level in the meu1Δ cells compared with the MEU1 + cells at the same optical density. These differences cannot be explained by an intrinsic difference in stability of the enzyme protein because, as discussed later in this section, the ODC levels from both MEU1 + and meu1Δ cells were both decreased by spermidine addition to the medium. We suggest that the apparent stability is related to the decreased spermidine level in the meu1Δ cells resulting in a lower concentration of antizyme. Further work including a direct measurement of antizyme activity or antizyme protein is necessary to confirm this postulation.
Second, Subhi et al. (17) reported the absence of detectable putrescine in MEU1 + cells and a >20-fold increase in the putrescine level in meu1Δ cells as opposed to our result of 3- to 5-fold increased putrescine level in meu1Δ cells (Table 2). They also found a marked increase in the spermidine level in the meu1Δ cultures, whereas we found a small decrease in logarithmically growing cells and a 50% decrease in cultures grown to an optical density of 1.5 (Fig. 5). We cannot explain the differences in polyamine content between these two studies.
It is possible that the decreased spermidine level in the meu1Δ cultures might be the result of inhibition of spermidine synthase by methylthioadenosine, because we have found that, in in vitro assays, methylthioadenosine inhibited spermidine synthase activity (Fig. 6). Spermidine synthases from rat prostate and bovine brain were also inhibited in an in vitro system by methylthioadenosine (30–32). However, we have not been able to show any change in the ODC activity or the putrescine or spermidine content when either MEU1 + or meu1Δ cells were grown with added methylthioadenosine. This finding contrasts with experiments in mammalian cells showing that addition of methylthioadenosine does indeed increase ODC activity and accumulation of putrescine (34, 38).
Third, a most important and conceptual difference between our results and those of Subhi et al. (17) is their finding that addition of spermidine to the culture medium did not affect the increased level of ODC in meu1Δ cells. In contrast, in our experiments, the addition of spermidine to the culture medium in both meu1Δ and MEU1 + cells resulted in a marked decrease in the level of ODC (Fig. 7). This result is what would be expected from the many previous studies on the regulation of ODC by spermidine in yeast (18, 22, 28, 29) and other systems (23–25). In murine lymphoblasts deficient in methylthioadenosine phosphorylase (MTAP), spermidine addition also suppressed ODC activity (34). Thus, our data and the data from murine lymphoblast cells do not support the postulation of Subhi et al. of a novel control mechanism of ODC that is insensitive to spermidine in meu1Δ cells. At this point, we have no explanation for the differences between our results and those of Subhi et al. (17).
Fourth, Subhi et al. (17) reported that, when 1 mM MTOB was added to a meu1Δ met15Δ culture, the ODC level was almost completely repressed. We found only a small effect in the meu1Δ met15Δ cells (20–25% decrease) and a somewhat larger effect in the MEU1 + met15Δ cells (30–40% decrease). We did not find any effect of MTOB when tested on MET15 + MEU1 + or MET15 + meu1Δ cells.
We conclude that the major regulatory system for polyamine biosynthesis in yeast in both MEU1 + and meu1Δ cells is the well known action of spermidine on the regulation of ODC via the antizyme system. In the absence of endogenous or exogenous methionine, a much smaller regulatory system might involve the methionine salvage pathway, as indicated by the small inhibition of ODC by MTOB (Fig. 8). The effects of the meu1 deletion on ODC levels are probably due to a decrease in the synthesis of spermidine resulting from the inhibition of spermidine synthase by methylthioadenosine, resulting in increased stability of ODC. It is possible that comparable changes might also account for some of the findings reported for the changes in ODC activity in the tumor cell lines that lack methylthioadenosine phosphorylase.
Acknowledgments
We thank Dr. Yolande Surdin-Kerjan (Centre de Génétique Moléculaire, Gif-sur-Yvette, France) and Dr. Warren D. Kruger (Fox Chase Cancer Center, Philadelphia, PA) for the yeast strains. This research was supported by the Intramural Research Program of the National Institutes of Health (National Institute of Diabetes and Digestive and Kidney Diseases).
Footnotes
-
↵ * To whom correspondence should be addressed. E-mail: tabor{at}helix.nih.gov.
-
Author contributions: M.K.C., C.W.T., and H.T. designed research, performed research, analyzed data, and wrote the paper.
-
Abbreviations: ODC, ornithine decarboxylase; MTOB, 4-methylthio-2-oxobutanoic acid.
