Uptake, intracellular binding, and excretion of polyamines during growth of Neurospora crassa.

In Neurospora crassa mycelia, the amounts of the main polyamines, putrescine and spermidine, are approximately 0.8 and 18 nmol/mg, dry weight. We wished to know what determines these pool sizes. In the growth medium, externally added polyamines enter cells largely by a nonsaturable, diffusional system. In a mutant unable to polyamines, internal and external spermidine appear to equilibrate across the cell membrane during growth. However, this was true only after an intracellular "sink," with a capacity equal to the amount of spermidine found in wild-type cells, had been saturated. We speculate that internal anionic binding sites, detectable in permeabilized cells, sequester virtually all of the spermidine normally found in exponentially growing N. crassa. Further evidence for this view was that in mature, stationary cultures, excess spermidine is excreted. Putrescine is also excreted if its concentration in the cell is abnormally high. The control of pool size by intracellular binding and excretion may be an advantage in this pathway, because feedback inhibition does not prevail, enzyme regulation is by comparison slow, and excessive polyamines are toxic.

In Neurospora crassa mycelia, the amounts of the main polyamines, putrescine and spermidine, are approximately 0.8 and 18 nmol/mg, dry weight. We wished to know what determines these pool sizes. In the growth medium, externally added polyamines enter cells largely by a nonsaturable, diffusional system. In a mutant unable to synthesize polyamines, internal and external spermidine appear to equilibrate across the cell membrane during growth. However, this was true only after an intracellular "sink," with a capacity equal to the amount of spermidine found in wild-type cells, had been saturated. We speculate that internal anionic binding sites, detectable in permeabilized cells, sequester virtually all of the spermidine normally found in exponentially growing N. crassa. Further evidence for this view was that in mature, stationary cultures, excess spermidine is excreted. Putrescine is also excreted if its concentration in the cell is abnormally high. The control of pool size by intracellular binding and excretion may be an advantage in this pathway, because feedback inhibition does not prevail, enzyme regulation is by comparison slow, and excessive polyamines are toxic. ?. 1989 Academic Press. Inc.
Polyamine biosynthesis in animals and fungi begins with the decarboxylation of ornithine by ornithine decarboxylase (ODC),3 followed by the conversion of the product, putrescine, to spermidine and spermine. These compounds are multivalent cations, with two, three, and four ionizable amine groups, respectively.
The polyamines have several peculiarities not shared by most biosynthetic intermediates. First, they are sequestered in cells (l), and it is likely that they bind to anionic cell constituents such as ribosomes, DNA, polyphosphates, and phospholipids (l-3). Second, while they are required for growth, spermidine and, in ' This work was supported in part by United States Public Health Service Research Grant GM-35120 from the National Institute of General Medical Sci-These peculiarities have led us to the present coordinated study of the management of polyamine pools of N. crassa in vivo, focusing on uptake, intracellular binding, and excretion during growth. MATERIALS AND METHODS ences.
less (a.ya) strain, IC3 (6); a strain lacking both orni-higher organisms, spermine, are "deadend" products: they are needed in few further biochemical reactions, and even catabolism of the polyamines may be restricted. Under these circumstances, it is odd that efficient feedback inhibition of ornithine decarboxylase, a key biosynthetic enzyme, has never evolved in any eucaryote. This is particularly unusual because high levels of polyamines may be toxic, even in an organism like Neurospwa crassa, where little polyamine turnover takes place (1,4,5). thine decarboxyiase and arginase (spe-1, a,ga), IC1894-53a (7); and strain IC2572-4a, carrying the ago mutation and another mutation, LV105, which blocks polyamine biosynthesis between putrescine and spermidine. The alleles of the spe-1 and a.ga genes were LVlO and UM906, respectively, and are available from the Fungal Genetics Stock Center, University of Kansas Medical Center, Kansas City, Kansas. The LV105 mutation has normal levels of the ornithine and S-adenosylmethionine decarboxylases, and is probably blocked in the spermidine synthase reaction. 4 The growth medium used was Vogel's medium N (8). Standing cultures were made in 10 ml medium in 50-ml Erlenmeyer flasks held at 32°C. Exponential cultures were heavily inoculated in medium in boiling flasks, with forced air for aeration and agitation, as described previously (9). The medium was supplemented as indicated under Results section. Dry weights were monitored by taking measured samples of culture, collecting the mycelia by filtration, acetone-drying, and weighing (9). Inoculations were made with conidia thoroughly washed in 0.25 M NaCl to remove external polyamines (5). Similarly, all harvests of mycelial samples, particularly those grown in the presence of polyamines, were washed in 0.25 M NaCl before extraction.
Polyamine uptake. Uptake of '"C-polyamines (putrescine and spermidine) was measured as described previously, but in the growth medium; that is, without transfer to the dilute medium used to characterize the saturable uptake systems (5).
Cell permeabilizatim. Retention of polyamines by permeabilized cells was measured by resuspending cells in a Na' 3-[N-morpholinolpropanesulfonic acid buffer (20 mM Nat, pH 7.2) with 0.2% glucose (5). The suspension was split in half, and n-butanol (7.5% by volume, final concentration) was added to one portion. After 5 min, cells were collected from both halves by centrifugation, and the medium and cell pellets were analyzed for putrescine and spermidine. The values for pellets were corrected for the amounts of extracellular polyamine in medium entrained by cell pellets.
Polyamine determinafions. Cellular polyamines were extracted with 0.4 M HCIO, which contained 2 mM EDTA. Extracts (usually 100 ~1 of a l-ml extract) were dansylated and determined by high-performance liquid chromatography as previously described (10). Samples of supernatants from the permeabilization medium, after addition of HClO, and EDTA to 0.4 M and 2 KIM, respectively, were analyzed similarly.
Polyamines of the growth medium were isolated by application to AGSOW cation-exchange columns, removing salts with 1.5 N HCI and eluting polyamines with 6 N HCl. These were evaporated, taken up in HClO,-EDTA, and determined as above. 1,YDiaminoheptane was used as an internal standard. Hydrolysis of medium and cell extracts in 6 N HCI yielded little or no more polyamines than unhydrolyzed samples, and no acetylpolyamines were detected in wild-type cells by direct analysis. The latter was done by thin-layer chromatography as described by Seiler and Knodgen (11).

Uptake of Polyamines from the Growth Medium
Uptake of putrescine and spermidine by wild-type (strain ORS-6a) or aga (strain IC3) cells from Vogel's growth medium is slow because the process lacks a saturable component seen in a dilute buffer (5). This can be attributed to the cations of the medium, particularly Ca'+, which are inhibitory to the saturable system. The remaining uptake activity is largely nonsaturable, suggesting a diffusional mechanism (  buffer, the saturable system contributes rates of about 1.8 and 3.0 nmol/min/mg at 1 mM putrescine and spermidine, respectively (5).) The initial uptake rate in Vogel's medium is short-lived (5 to 30 min), suggesting equilibration across the cell membrane or a negative control of further uptake.
Excretion of Polyumines during Growth for "excretion." Instead, it suggests that spermidine is retained by cells by a mechanism that can be saturated, that synthesis continues after that point, and that the excess is lost to the medium. Cultures that were not swirled showed the same pattern of appearance of polyamines in the medium and in the cells, indicating that the differentiation involved in conidiation had little effect.
If the mechanism of uptake from the growth medium is diffusional, one might expect efflux of polyamines from the cell during growth. This was tested in two situations: either during standing, long-term growth or during logarithmic growth of young cultures (Table I). Standing cultures (10 ml) were inoculated with conidia of the wild-type strain, and the media and cells were analysed at 24, 48, and 72 hr growth at 32°C. The cultures were swirled twice each day to prevent aerial growth and conidiation. As growth proceeded, putrescine and spermidine, particularly the latter, were found in the medium (Table I). The greatest increase was seen as growth stopped. Although rather little putrescine was in the cells at any time, a higher proportion of it was lost to the medium, in keeping with its lower valence. The maintenance of the cellular spermidine pool per milligram dry weight argues against cell breakage or cell death being resnonsible Excretion of polyamines during logarithmic growth of germinating conidia was tested in three cultures (Table II). In the case of wild type mycelia, virtually no polyamines were excreted. The LV105, uga strain (IC2572-4a), which accumulates putrescine, lost over 10% of its putrescine to the medium (Table II). Spermidine was not seen in the medium, in part because little was present in the cell. Finally, the arginase-less aga strain (IC3), grown on arginine and later given ornithine, was tested. During growth on arginine, this strain cannot make ornithine because its biosynthesis is feedback inhibited and the arginase reaction, an alternate source of ornithine, is missing (6). Under these conditions, ornithine decarboxylase reaches about 70-fold its normal activity in the cells. (Growth continues, albeit slowly, because a small amount of the putrescine analog, cadaverine, appears by the decarboxylation of lysine (12).) The arginine-grown -_ cells were given ornithine, leading to a large burst of putrescine synthesis (4). About one-third of the putrescine made appears in the medium. None of the spermidine is excreted, but in fact the spermidine pool never much exceeds that found in wild-type cells.
Extracts of standing and logarithmic cultures have only traces of acetylpolyamines or other polyamine conjugates, and very little of the polyamines are found conjugated to macromolecuIes (data not shown). Little polyamine catabolism has been detected in cultures of this type, and polyamines do not serve as a nitrogen source (5). Only in strains that take up excessive amounts of putrescine does acetylputrescine appear, and it is not a large fraction of the putrescine absorbed.5 5 R. H. Davis and J. L. Ristow, unpublished observations Therefore, polyamine excretion is the only known means of disposing of a large excess of polyamines.

Relation of Internal and External Spermidine
Steady-state cultures of the wild-type and the spe-1 mutant strains were tested for the effect of spermidine supplementation upon the internal spermidine pool (Fig. 2). Because little spermidine is synthesized, spermidine is almost a nonmetabolizable intermediate, as noted above. In varying the spermidine concentration in the medium, the tests of the two strains differed at the lowest concentration: wild type was grown in minimal medium, while the spe-1 mutant was grown necessarily with 0.5 mM spermidine. The slope of the internal vs external spermidine concentra- Open symbols, abscissa represents putrescine-supplemented cultures; closed symbols, abscissa represents spermidine-supplemented cultures. The dotted line represents equal concentrations in the medium and in cell water, assuming 2.5 ml cell water per gram cells, dry weight (right ordinate). tion was consistent with equilibration of spermidine across the membrane. However, the y-intercept of this function in the spe-1 mutant, in which all spermidine comes from the medium, is about 18 nmol/ mg dry weight. This reflects very efficient uptake of spermidine from the medium at low concentrations.
It is very likely that this efficient uptake at low external levels of spermidine is driven by intracellular binding to anionic cell constituents.
At higher concentrations of extracellular spermidine, these sites would be saturated, and free equilibration of diffusible, intracellular polyamines with those in the medium takes place.
It is noteworthy that the y-intercept of the spe-1 spermidine curve, when spermidine is varied, is equal to the amount of this metabolite found in wild-type cells grown in minimal medium (Fig. 2). This suggests that the same intracellular binding capacity largely determines the spermidine pool size of wild-type cells, and that there is little diffusible spermidine, in keeping with previous tracer work (1).

Relation. of External Putrescine and
Internal Polyamines When external putrescine is varied in wild-type cultures, the cellular putrescine level roughly follows what is expected of equilibration across the cell membrane, with the curve passing close to the origin (Fig. 2). However, because the wild-type strain continues to make putrescine and spermidine even in the presence of external putrescine, the contribution of uptake to the internal pool is not clear. Ih the spe-1 mutant, the synthesis of spermidine is clearly limited by putrescine uptake at low putrescine concentrations, and spermidine achieves its normal pool size only when the medium has over 2 mM putrescine (Fig. 2). Only at this concentration of external putrescine, in fact, does equilibration across the cell membrane keep pace with demand. Intracellular binding of putrescine (a divalent amine) cannot be seen in this experiment, probably because the trivalent spermidine competes successfully for binding sites.
There is a significant discrepancy between the polyamine pools of wild type and the spe-1 mutant. Whereas wild type can maintain a normal pool of spermidine (18 nmol/mg, dry weight) in minimal medium with 0.8 nmol of putrescine, the spe-i mutant is unable to do so until the internal putrescine pool reaches 3 to 4 nmol/mg dry weight (Fig. 2). Because the putrescine in wild type is drawn from the ODC reaction, and that of the spe-1 mutant is transported from the medium, ornithine decarboxylase appears to deliver putrescine to the spermidine synthase reaction more efficiently than the transport system does. Whether this reflects an organization of enzymes in wild type or uptake and sequestration at unfavorable locations in the mutant cannot be decided.

Polyamine Retention by Permeabiilized
CQllS Cells of wild type, grown in minimal medium or in medium supplemented with 5 mM putrescine or 5 mM spermidine, were tested for their ability to retain their polyamine pools after permeabilization with nbutanol (Table III). The technique is known to remove all soluble arginine, 99% of which is in the vacuoles (13), from cells (5). The permeabilization medium has rather low ionic strength, and the degree of binding of polyamines by permeabilized cells does not necessarily reflect the state of polyamines in viva. Nevertheless, all permeabilized cells appeared to retain over 90% their spermidine, and over half their putrescine. This was true even of cells preloaded during growth with putrescine or spermidine. Only in the case of spermidine-loaded cells was putrescine retained poorly, but the values are too low to be really meaningful. The experiment demonstrates that there is significant anionic material in cells capable of binding polyamines, the strength of binding being related to the charge of the polyamine, as predicted from the observations of living cultures. It is significant that the retention of total polyamines in this experiment was much better than that of isotopically labeled polyamines introduced into cells just before permeabilization in previous experiments (5). DISCUSSION We have shown that in the growth medium, the saturable polyamine uptake systems seen in dilute buffer are virtually inactive, and that entry of polyamines is largely diffusional. The nonconcentrative nature of this system was shown by analysis of steady-state cultures in which the external polyamine concentration was varied. Over most of the concentration range tested, the slope of internal vs external concentration had the relationship expected of equilibration across the cell membrane. Moreover, excess polyamines are excreted into the medium, although not necessarily at a rate that maintains a normal pool size.
Of greatest interest was the finding that a considerable, apparently concentrative uptake of spermidine into spe-1 cells took place when the medium had a low concentration (co.5 mM) of spermidine. Above this concentration, the internal and external concentrations varied as expected of equilibration.
We interpret the apparent concentrative uptake at low concentration as evidence for spermidine sequestration by anionic binding sites within the cell, such as nucleic acids (especially of ribosomes (2)), vacuolar polyphosphate (l), and phospholipids (3). This is consistent with our previous tracer work showing that at least 70-85% of the spermidine of N. crassa cells is sequestered from biosyn-thetic reactions producing and consuming nisms. In addition, intracellular binding of it (1). It also fits with the finding that buta-the potential effecters (e.g., spermidine nol-permeabilized, wild-type cells release and spermine) may make them erratic alhardly any endogenous spermidine. losteric signals. Quantitatively, the amount of spermi-We may speculate that during periods of dine trapped by spe-1 cells at low external polyamine insufficiency, the needs of cells concentration approximates that found in for polyamines can be met for a time by wild-type cells grown in minimal medium.
drawing on the sequestered fraction. This This implies that in exponential cultures would allow time for derepression or enof N. crassa, the spermidine synthetic rate zyme stabilization to restore a normal rate saturates spermidine binding sites, with-of synthesis. When polyamines are in exout much excess. The small excess, in fact, cess, N. crassa appears to dispose of excess is the pool used as an intermediate be-polyamines by excretion. This is seen most tween putrescine and spermidine synthe-clearly in spermidine excretion by normal sis, and as the metabolic signal for regulat-stationary cultures, but putrescine can ing ornithine decarboxylase synthesis and also be excreted when it is made in great turnover (10,14). Because the diffusible excess, as noted above. In the absence of a pool is so small, it is highly responsive to substantial polyamine catabolic route, exthe rate of spermidine synthesis, and cretion of polyamines prevents their accuis well suited to the role of a regulatory mulation to toxic levels. signal.
The mechanism of control of the pools in It was not possible to decide whether pu-N. crassa is very similar to that for the putrescine might bind to intracellular sites in trescine pool of Escherichia coli proposed +uo. Under all conditions in which inter-recently by Kashiwagi and Igarashi (15). nal and external concentrations were com-They found that excess putrescine was pared, the anionic binding sites inferred readily excreted, especially in strains that from the spermidine experiments were ei-overproduced it. Spermidine was not exther occupied with the stronger cation, creted by E. coli, unlike N. crassa, in which spermidine, or, if not, little putrescine was it is lost to the medium when growth stops. available owing to its rapid conversion to The difference is probably due to the spermidine.
The behavior of putrescine difference in the efficiency of feedback conwas that expected of equilibration with trol of S-adenosylmethionine decarboxylcell water, as though no binding sites rease in E. coli, which readily prevents excess mained. Only in the logarithmic cultures spermidine biosynthesis, even when the of the LV105, aga strain, having little spergene for the enzyme is present in multiple midine, was there a large amount of intracopies. It is very likely, however, that polycellular putrescine which might have been amine excretion is part of the way in which bound internally (Table II). Moreover, many organisms and cell types manage the some putrescine is bound by permeabilized size of their polyamine pools (16), and that cells, even those that contain normal this comes into play when intracellular amounts of spermidine.
binding sites for the higher polyamines are The control of polyamine pool size insaturated. volves a number of factors. In terms of polyamine synthesis, the control of the synthesis and the turnover of ornithine de-ACKNOWLEDGMENTS carboxylase are the most important (14). It We thank Glenn Barnett, John Pitkin, and Laura is noteworthy that these mechanisms are Williams for extensive discussion and critical reading the sole known mechanisms of controlling of the manuscript. this enzyme in N. crassa, and that feedback inhibition does not prevail. The latter mechanism has not been found in any or-