INTRODUCTION

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS LITERATURE CITED

Polyamines are low-M_r aliphatic amines that are present in all living organisms. In higher plants, the most prevalent polyamines are spermidine (Spd), spermine (Spm), and their diamine precursor, putrescine (Put). A variety of roles have been proposed for polyamines in the growth, development, and stress responses of plants (for review, see Minocha and Minocha, 1995; Cohen, 1998; Bouchereau et al., 1999). Polyamines also serve as precursors for secondary metabolites such as nicotine, and can be conjugated with phenolic acids to produce plant defense-related compounds (Martin-Tanguy, 1997).

Rates of polyamine biosynthesis and degradation, their conjugation with phenolic acids, and intercellular transport all contribute to cellular levels of free polyamines in plants. Put is synthesized directly from Orn by Orn decarboxylase (ODC; EC 4.1.1.17) and indirectly from Arg by Arg decarboxylase (ADC; EC 4.1.1.9) and two other enzymes (Fig. 1). It appears that the activities of ODC and ADC are regulated in a developmental and tissue-specific manner (Minocha et al., 1995; Walden et al., 1997). Spd is formed from Put by the addition of an aminopropyl group derived from decarboxylated S-adenosyl-Met (SAM) by the enzyme Spd synthase (EC 2.5.1.16). The addition of another aminopropyl group to Spd gives rise to Spm. Decarboxylated SAM is produced from SAM via SAM decarboxylase (SAMDC; EC 4.1.1.50). All three decarboxylases have short half-lives (Cohen, 1998), indicating that they are important metabolic control points in the cell.

View larger version (14K): [in this window] [in a new window]   Figure 1.   Pathway summary of polyamine biosynthesis and catabolism in plants. GDC, Glu decarboxylase; PDH, 1-pyrroline dehydrogenase.

Polyamines are catabolized through the activity of one or more diamine oxidases (DAOs; E.C.1.4.3.6) and polyamine oxidases (PAOs; E.C. 1.5.3.3; Smith, 1985; Cohen, 1998; Bagni and Tassoni, 2001). DAO is believed to be loosely bound to the cell wall and can be released in the apoplast (Angelini et al., 1993; Møller and McPherson, 1998). The best substrates for DAO are Put and cadaverine (Cad); however, it can also act on aromatic and aliphatic monoamines (Smith, 1985). As a result of DAO action, Put is converted into ¹-pyrroline with the release of ammonia and hydrogen peroxide (Fig. 1). Pyrroline dehydrogenase then converts ¹-pyrroline to -amino-butyric acid (GABA), which enters the tricarboxylic acid (TCA) cycle via succinate after transamination and oxidation. An alternate route for GABA production in plants, especially under stress conditions, is directly from Glu through the action of (cytosolic) Glu decarboxylase (Shelp et al., 1999). Degradation of Spd by PAO yields ¹-pyrroline and 1,3 diaminopropane, whereas breakdown of Spm yields 1,3-aminopropylpyrroline (which subsequently gets converted to 1,5-diazabicyclononane), along with diaminopropane and hydrogen peroxide (Smith, 1985; Bouchereau et al., 1999; Bagni and Tassoni, 2001). Diaminopropane is eventually converted into -Ala (Fig. 1). Thus, polyamine catabolism is not simply a degradative process but is also an important link between amino acid and carbon metabolism in plants.

Plant cells take up polyamines by a rapid and active mechanism that shows biphasic kinetics and reaches saturation within a few minutes (Bagni and Torrigiani, 1992). In whole plants, polyamines can be translocated via the xylem (Bagni and Pistocchi, 1991) and are also present in the phloem (Friedman et al., 1986).

Transgenic manipulation of polyamine metabolism has become a valuable tool for studying their physiological roles in plants (Walden et al., 1997; Kumar and Minocha, 1998; Bhatnagar et al., 2001; Roy and Wu, 2001). Cellular content of polyamines has been modulated by overexpression or down regulation of the key genes odc, adc, and samdc. Overexpression of heterologous odc or adc cDNAs generally causes the production of high levels of Put (DeScenzo and Minocha, 1993; Bastola and Minocha, 1995; Burtin and Michael, 1997; Capell et al., 1998; Kumar and Minocha, 1998; Bhatnagar et al., 2001). In most cases, only a small increase in Spd and Spm has been observed despite the elevated levels of Put in the transgenic cells. This observation, combined with the fact that under conditions of stress (Minocha et al., 1997, 2000; Bouchereau et al., 1999) it is also mostly Put that is found to fluctuate widely without major changes in Spd and Spm, suggests that the levels of Spd and Spm in the cells are under a tight homeostatic regulation. The exact mechanism of this homeostasis is not known; it could be achieved through their regulated biosynthesis or increased degradation or both. An obvious question then is: What is the fate of excess Put produced in the transgenic cells overexpressing one of the Put biosynthetic enzymes?

Studies with animal cells in which increased production of polyamines has been induced either chemically or by transgenic manipulation show a concomitant increase in polyamine catabolism and also increased excretion and/or transport (Halmekytö et al., 1991, 1993; Seiler et al., 1996). Although Put in these cells/tissues is largely excreted, increased catabolic breakdown of Spd and Spm occurs via the induction of Spd acetyltransferase and Spm acetyltransferase and associated PAOs (Cohen, 1998). There is also an activation of the ODC antizyme gene, a unique mechanism by which animal ODC levels are regulated, which results in increased turnover of the native ODC protein (Cohen, 1998).

In plants, the relationship between increased Put production and its catabolism is not clear. In bean (Phaseolus vulgaris) seedlings, elevated levels of Put or Cad during development were accompanied by a concomitant increase in the activity of DAO. Thus, it was postulated that faster oxidation of Put in this tissue prevented its level from rising further during periods of accelerated biosynthesis (Scoccianti et al., 1990). A possible function of DAO in regulating polyamine levels during the cell cycle of in vitro-cultured Helianthus tuberosus tuber explants has also been suggested (Torrigiani et al., 1989). A positive correlation between DAO activity and the content of Cad in hypocotyls of soybean seedling was also seen by Scoccianti et al. (1990); however, there was no correlation in the root, where Cad dropped sharply and the DAO activity remained unaltered.

Although a number of laboratories have reported the production of transgenic cell lines and plants expressing genes for polyamine biosynthetic enzymes, and the resulting increases in cellular polyamine contents, none have investigated the effect of polyamine overproduction on their turnover and catabolism (Kumar and Minocha, 1998). Preliminary data from our laboratory suggested that there was an increase in Put degradation in transgenic cells overexpressing a mammalian odc, and the amount of Put being converted into Spd and Spm was relatively small (Andersen et al., 1998; Bhatnagar et al., 2001). Because the expression of odc transgene used in these studies was controlled by a constitutive promoter, and the transgenic cells still reached a plateau for Put accumulation, it can be hypothesized that the excess Put being produced in the transgenic cells must be metabolized or excreted at a fast rate to keep pace with its production.

In this paper, we present experimental evidence on the effects of Put overproduction on the rates of Put catabolism in transgenic poplar (Populus nigra × maximowiczii) cells overexpressing a heterologous odc cDNA. Data are presented on the rates of Put turnover, its conversion into Spd, and on the activity of DAO (a key enzyme involved in Put catabolism) in non-transgenic (NT) and transgenic cells. The results show that there is an increased turnover of Put, as well as its conversion into Spd, in transgenic cells as compared with the NT cells, the increase being proportionate to the cellular content of this diamine. Furthermore, it is demonstrated that the increase in Put catabolism in the transgenic cells is achieved without major changes in the activity of DAO, or the apparent half-life of Put.

	RESULTS

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS LITERATURE CITED

Polyamine Levels in Transgenic and NT Cells

Data on cellular contents of the three polyamines in the two cell lines (NT and a transgenic cell line 2E) over the 7-d culture period have been published earlier (Bhatnagar et al., 2001). Over the period of the present study, the cellular Put content fluctuated somewhat in both cell lines but remained within a narrow range. The 2E cells used in the experiments reported here contained 3 to 4 times higher Put content than the NT cells (Table I). The cellular content of Spd in the 2E cells was also generally higher than the NT cells, the overall amount of Spd being substantially lower than Put at any given time. Spm, which constituted less than 5% of the total soluble polyamines at any given time for both types of cells, showed no significant differences between the two cell lines.

View this table: [in this window] [in a new window]   Table I.   Cellular content of PCA-soluble polyamines (nmol g1 FW) in the non-transgenic (NT) and transgenic (2E) cell lines of poplar at different time intervals after transfer of cells treated with [1,4-14C]Put to label-free medium Data presented are mean ± SE of six replicates from two experiments. An asterisk indicates that the values for 2E cells were significantly different from the NT cells at a given time (P  0.05).

Put Turnover in Transgenic and NT Cells

The turnover of cellular Put in the two cell lines was investigated in two different ways: (a) labeling endogenous Put in the cells by feeding them with [¹⁴C]Orn and analyzing the loss of label from the Put fraction in the cells with time after removal of the label, and (b) allowing the cells to accumulate exogenously supplied [¹⁴C]Put and following its loss with time after being transferred to label-free medium. The cells were incubated with either [U-¹⁴C]Orn or with [1,4-¹⁴C]Put for 2 h, washed with label-free medium, and then transferred to label-free fresh medium. After transfer to the label-free medium, samples were collected at different times for analysis of their [¹⁴C]polyamine content by thin-layer chromatography (TLC) and the total polyamine content by HPLC.

Recovery of Label during Dansylation and Processing of Polyamines

The efficiency of extraction of radioactivity from the cells into perchloric acid (PCA) by freeze thawing, and the recovery of label during various steps of the dansylation procedure and TLC separation, were followed carefully to assess the extent of loss of labeled metabolites during the processing of samples. The supernatant fraction from the freezing and thawing process (three times) contained more than 90% of the total radioactivity present in the cells incubated with [¹⁴C]Put for 2 h, i.e. at the time of transfer to label-free medium (Table II). The second and third PCA extractions contained about 8% and 1% of the remaining radioactivity, respectively. Less than 1% of the total radioactivity was found in the PCA-insoluble fraction. Data presented in Table II further show that during the various steps of dansylation, no losses of radioactivity were incurred, all of the [¹⁴C] being recovered in the toluene or the aqueous fractions. The toluene fraction contained more than 70% of the radioactivity from the PCA extract. The steps of vacuum drying of the toluene fraction and the reconstitution of dansyl-polyamines in methanol resulted in about 25% loss of radioactivity. Of the total amount of radioactivity loaded on to the TLC plates in the methanol fraction, 67% was present in the three polyamine fractions. Assuming that losses for the three polyamines in both cell lines were proportionate to their amounts present at each step, we are confident that the data on specific activities of Put and Spd, and the calculations of amounts of Put loss and its conversion into Spd, represent reliable and realistic measures of actual cellular contents of these metabolites.

Labeling with [U-¹⁴C]Orn

1

1

View larger version (18K): [in this window] [in a new window]   Figure 2.   Changes in the amount of [14C]Put (A) and [14C]Spd (B) derived from [U-14C]Orn in the NT and 2E cells at different times of analysis after treatment with [U-14C]Orn for 2 h and transfer of cells to label-free medium. Inset, Regression curve for the loss of [14C]Put over the 8-h period after transfer of labeled cells to label-free medium. C, Specific radioactivity of [14C]Spd derived from [U-14C]Orn in NT and 2E cells at different times after transfer of cells to label-free medium. D, Amount of label present in the aqueous fraction of dansylated PCA extract of NT and 2E cells at different times after transfer to label-free medium. This fraction is counted after the removal of dansyl-polyamines by partitioning with toluene. Data presented for A through D are mean ± SE of nine replicates representing three separate experiments. Asterisks in B and D indicate that the values for 2E cells are significantly different (P  0.05) from NT cells at a given time; all values in A were significantly different, and in C differences were insignificant. Values have been adjusted for losses during dansylation procedure.

View this table: [in this window] [in a new window]   Table IV.   The amount of total Put lost during the 2- and 8-h period and the rate of conversion of Put to Spd during the 8-h period following transfer of cells from labeled substrates to label-free medium Values for total polyamines used for calculations (nmol g1 fresh wt) for [1,4-14C]Put are from Table I. The radioactive polyamine data (dpm g1 fresh wt) for [14C]Orn-feeding experiment are given in Figure 2 and for [14C]Put-feeding experiment are given in Figure 4. The calculations were done as shown in "Materials and Methods." Data presented are mean of nine replicates for [14C]Orn-feeding experiments and six replicates for [14C]Put-feeding experiments. All values for 2E were significantly different from the NT cells at a given time (P  0.05).

View larger version (11K): [in this window] [in a new window]   Figure 3.   Calculated half-life (L50) of [14C]Put in NT and 2E cells treated with [U-14C]Orn or [14C]Put. Each bar represents the mean L50 value of combined data from three (for [14C]Orn) or two (for [14C]Put) experiments, each with three replicates. The L50 was calculated by using data on the loss of [14C]Put at various times during the first 8-h period after transfer of cells to label-free medium.

Labeling with [1,4-¹⁴C]Put

View larger version (16K): [in this window] [in a new window]   Figure 4.   Changes in the amount of [14C]Put (A), specific radioactivity of [U-14C]Put (B), and amount of [14C]Spd (C) in the NT and 2E cells at different times after transfer of cells treated with [1,4-14C]Put for 2 h to label-free medium. Data presented are mean ± SE of six replicates from two separate experiments. An asterisk indicates that the values for 2E cells are significantly different (P  0.05) from NT cells at a given time. Values have been adjusted for losses during dansylation procedure.

1

1

Activity of DAO

As described earlier, Put is catabolized primarily by DAO, the product of the reaction being ¹-pyrroline, which eventually enters the TCA cycle via GABA and succinic acid (Fig. 1). Because the 2E cells show severalfold higher rate of Put catabolism, one can ask the question: Is the increased catabolism of Put in 2E cells accompanied by increased DAO activity? To test this, we compared the activities of DAO in the two cell lines over the 7 d culture period. Before measurement of DAO activity, the protocol for enzyme assay in poplar cells was optimized by modification of the procedure of Santanen and Simola (1994) and Santanen (2000). A major change was that the assays were conducted using frozen-thawed cells without homogenization. We first compared the enzyme activity in the frozen-thawed cells with the supernatant and the pellet fractions of homogenized cells. The data presented in Figure 5A show that: (a) about two-thirds of the DAO activity in the cells was present in the pellet and about one-third in the supernatant fraction of the homogenate, and (b) the total DAO activity in the frozen-thawed cells was comparable with the combined activity in the pellet and the supernatant fraction of the homogenate. After this determination, the DAO activity in the frozen-thawed cells was analyzed at three different pH values, at various incubation periods, and six different quantities of cells per tube. For frozen-thawed cells, the rate of [¹⁴C]¹-pyrroline formation was proportionate to the amount of cells for up to 200 mg fresh weight per tube (Fig. 5B), the optimum pH was 8.0 (Fig. 5C), and the reaction was linear for at least up to 120 min (Fig. 5D).

View larger version (35K): [in this window] [in a new window]   Figure 5.   Characterization of DAO activity in poplar cells with respect to its extractability by freeze thawing versus homogenization of cells (A), linearity with the amount of cells used (B), the effect of pH (5), and the effect of time of incubation (D). For A, C, and D, 200 mg fresh weight of frozen-thawed cells were used per tube; for A, B, and D, the reaction was run at pH 8.0. The data are mean ± SE of three replicate assays from a single representative experiment.

Based upon the above characterization, DAO activity was measured on different days of culture in the two cell lines using 200 mg fresh weight of cells that were frozen and thawed three times. The reaction was run at pH 8.0 for a period of 60 min. Data presented in Figure 6 show that the enzyme activity per gram fresh weight of cells varied considerably on different days of culture in both the cell lines. It was the lowest at the time of subculture (0 d in Fig. 6) and increased severalfold during the next 2 d in both cell lines. The enzyme activity declined significantly after 4 d of culture, reaching the lowest level by d 7 (the same as d 0 in Fig. 6). The trend of changes in DAO activity during the 7-d culture period was similar in the two cell lines. Except for d 6 and 7, when the 2E cells had somewhat higher DAO activity, the enzyme activity in the two cell lines was either comparable, or it was lower in the 2E cells than the NT cells.

View larger version (40K): [in this window] [in a new window]   Figure 6.   The activity of DAO (amount of 1-[14C]pyrroline formed) in NT and 2E cells over a 7-d time course. The cells (200 mg) were collected in 0.1 M potassium phosphate buffer, frozen thawed three times, and analyzed for the formation of [14C]pyrroline from [1,4-14C]Put over a period of 60 min at 37°C. Each bar represents mean ± SE of three replicate assays from a representative experiment. An asterisk indicates that the values for 2E cells are significantly different (P  0.05) from NT cells at a given time.

	DISCUSSION

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS LITERATURE CITED

The amount of free Put present in a plant tissue is the net result of its biosynthesis, conjugation with phenolic acids, transport to other tissues, and its degradation during a given period of time. Poplar cell suspension offers an excellent experimental system to analyze the contributions made by the rates of biosynthesis and degradation to the regulation of Put homeostasis, with relatively little interference from intercellular transport and conjugation into insoluble fraction. These cells contain very little conjugated polyamines and the amount of polyamines excreted from the cells into the medium can also be monitored. The culture is composed of a population of single cells and small filaments, without nodular structures, providing a system that is amenable to feeding of radioactive precursors and following their metabolism over short time intervals. This is in contrast to whole plants and plant organs in which uptake of the precursors by different tissues and their intercellular transport complicate the calculation of flux rates of labeled metabolites. The lack of fully developed chloroplasts also eliminates the possibility of compartmentation of polyamines in this organelle. In addition, these cells possess only one biosynthetic pathway for Put production, i.e. via ADC; the alternate pathway of ODC is not functional.

Regulation of Cellular Put in Poplar Cells

In a previous study, we reported that wild-type poplar cells grown in culture normally contain relatively low levels of Put but can tolerate this diamine at considerably higher levels, as seen in the transgenic cells overexpressing a mouse odc cDNA (Bhatnagar et al., 2001). The transgenic cells maintain a significantly higher threshold of Put throughout the 7-d culture period. Two important questions that can be asked of these cells are: (a) Why does Put in the NT cells plateau at a lower level than in the transgenic cells? and (b) Why do the transgenic cells even have a plateau of Put when the biosynthetic enzyme (ODC in this case) is being constitutively produced? Three possible answers for the first question are: (a) because the NT cells use only the ADC pathway for Put biosynthesis, the amount of Arg in the cells becomes limiting; (b) there is a feedback inhibition of ADC by Put; and (c) an equilibrium is established between the rates of Put biosynthesis and its degradation. The earlier results from our laboratory showed that the addition of either Orn (a precursor of Arg biosynthesis) or Arg to the medium does cause increased accumulation of Put in the NT cells, showing that these precursors may be limiting in the cells (Bhatnagar et al., 2001). The data presented here provide additional support for this explanation. It is apparent that the exogenous [¹⁴C]Orn is metabolized quite rapidly in these cells; as much as two-thirds of the label in the 2E cells appeared as [¹⁴C]Put within 2 h of [¹⁴C]Orn feeding. Earlier studies also showed that there was no significant feedback inhibition of ADC by Put in these cells. Thus, it can be argued that a steady-state equilibrium between biosynthesis and degradation of Put is established, the former being largely responsible for regulating the amounts of Put present at a given time.

If the same arguments were applied to the transgenic cells possessing a heterologous ODC pathway for Put production in addition to the native ADC, it would be expected that the cellular Put content is again the net product of its biosynthesis and degradation with only small amounts being converted into Spd or secreted into the medium. It is clear that Put biosynthesis in the transgenic cells is increased severalfold because of the constitutive presence of the mouse ODC without adverse effects on the ADC pathway. As a consequence, in the transgenic cells, the ODC pathway is able to continuously produce Put, in turn causing a steady rise in its cellular concentration. However, there apparently is a mechanism to achieve an increased degradation of this diamine in these cells so that a plateau is established, albeit at higher levels, for Put.

Regardless of whether Put is synthesized in the cells from exogenously supplied Orn, or it is directly taken up as Put from outside, its loss is proportionate to its amount present in the cells, i.e. 2E cells with higher amount of Put lose it at a higher rate than NT cells with low levels of Put. The relative importance of the three components of Put loss (i.e. conversion into Spd, excretion/secretion into the medium, and catabolic breakdown) can be estimated from the data presented here. First of all, the amount of Put converted into Spd constitutes only a small fraction of the total amount of Put in the cells. For radioactive Put produced either from [¹⁴C]Orn or taken up from the medium as [¹⁴C]Put, only 6% to 10% is converted into Spd in the first 8 h of feeding the label in either cell line. For [¹⁴C]Put-feeding experiments, this amount is similar in the two cell lines in terms of dpm g¹ fresh weight but almost 3-fold higher in the 2E cells compared with the NT cells in terms of nmol g¹ fresh weight. The second component of Put loss, i.e. by secretion from the cells is also relatively small (less than 10% of the Put pool in the cells; data not shown), but again proportionate to the amount of Put in the two cell lines. The observation that a higher proportion of radioactivity is lost to the medium from cells fed with [¹⁴C]Put than those fed with [¹⁴C]Orn indicates that some of the [¹⁴C]Put may be present in the apoplast at the time of transfer, and thus washed out easily. Therefore, it can be argued that most of the Put loss from the cellular pool is by catabolic breakdown into other products that appear mostly in the aqueous fraction after dansylation, and/or as ¹⁴CO₂. That is probably why the radioactivity in the aqueous fraction is higher in the 2E cells than the NT cells at all times after the transfer of cells to the label-free medium.

The first step in Put breakdown is catalyzed by DAO. The question then is: Is the catabolic mechanism (i.e. the enzyme DAO) that handles extra Put already present in the cells or is it induced by a higher threshold of this diamine in the transgenic cells? The observation that increased catabolism of Put in the 2E cells occurs without an increase in the DAO activity leads us to conclude that the former situation is probably responsible for the increased degradation of Put in the transgenic cells. In other words, when Put degradation in poplar cells is increased concomitant with the rates of its biosynthesis, this increase happens without induction of the catabolic enzyme DAO. The data presented in Figure 6 show that the measurable amounts of DAO activity in both the NT and the transgenic cells were quite comparable (or smaller in the 2E than the NT cells) at most of the times during the 7-d culture period, except when the cells were in a stationary phase of growth, i.e. on the 6th and 7th d of culture. A possible explanation for the same amount of DAO to catalyze higher rates of Put degradation is that in the wild-type cells, the enzyme is either functioning at a non-saturating level (i.e. high K_m of the enzyme but low amounts of available substrate) or the enzyme is present in non-saturating amounts (i.e. excess amount of the enzyme). Thus, any additional Put in the cells will be degraded at a rate proportionate to its availability up to the level of saturation.

The above explanation is consistent with the earlier reports showing that DAO may be constitutively present in quantities that are sufficient to catabolize a severalfold higher amount of Put than normally found in the cells (Burtin and Michael, 1997). In a situation reminiscent of what we have observed for the relationship of Put with its degradative enzyme DAO, Forlani et al. (2000) detected no relationship between the cellular contents of free Pro and 1-pyrroline-5-carboxylate (P5C) dehydrogenase, a key degradative enzyme for Pro catabolism in potato tubers. Likewise, Hua et al. (1997) observed no positive correlation between cellular Pro and P5C reductase, another key enzyme involved in Pro catabolism. The above examples illustrate that the synthesis of these enzymes is not regulated by the amounts of the respective metabolites that serve as their substrates in the cells. In other words, the catabolism of these compounds is a function of their presence at non-saturating concentrations. Based upon the above arguments, we believe that a similar situation exists for Put catabolism in poplar cells. The higher threshold of Put in the transgenic cells is thus the net result of increased biosynthesis (via ADC and ODC) and a constitutively high rate of degradation, the former being the regulatory step responsible for the overall metabolic flux of this diamine.

Existence of One or Two Pools of Put

Because plant cells often possess and utilize both ADC and ODC pathways for Put biosynthesis, the relative contribution of the two pathways to cellular Put pools has been variously debated (Cohen, 1998). It has been suggested that Put produced by the two pathways may exist in separate pools and/or may be involved in different physiological functions. For example, ADC may play a greater role in stress-induced Put production, whereas ODC may be important for cell division and development. Although no direct evidence for the existence of multiple pools of Put has been forthcoming, the idea has persisted for many years. A transgenic approach using cells in which a single endogenous pathway for Put biosynthesis is supplemented by an alternate pathway, and in which the metabolism of [¹⁴C]Put produced endogenously (from [¹⁴C]Orn) versus that given exogenously (as [¹⁴C]Put) can be studied, provides an excellent opportunity to test these ideas.

The observation that the loss of labeled Put in the NT cells using two different sources of label (i.e. [¹⁴C]Orn or [¹⁴C]Put) follows similar kinetics (i.e. similar L₅₀) indicates that there probably is only a single pool of Put in these cells. The reasoning for this argument is as follows: If exogenously supplied Put was degraded differently from that produced endogenously, its degradation rate would be either higher (if it was preferentially degraded) or lower (if it was sequestered into a separate compartment). This, however, will not be the case if either there was only one pool of Put in the cells, or both pools were equally accessible to the catabolic enzyme(s). Because a significant proportion of DAO is believed to be wall bound (Angelini et al., 1993; Møller and McPherson, 1998; Wisniewski et al., 2000; Fig. 5A), one would expect that the exogenous Put would be catabolized faster than the intracellular Put. The results presented here, however, show that the profiles of [¹⁴C]Put loss are similar in the two cell lines and the half-life (L₅₀) of Put remains unchanged, independent of the source of Put, indicating an equal access of the endogenous as well as exogenous Put to the catabolic pathway.

The approach of using two different means of labeling cellular Put provides further insight into the metabolism of this diamine with regard to its turnover rates. Using a combination of these two approaches, we estimate the half-life (L₅₀) of Put loss in both the NT and the 2E poplar cells to be about 5 to 6 h, although the calculated L₅₀ of Put in the 2E cells using [¹⁴C]Orn as the precursor was found to be somewhat longer than that calculated from experiments using [¹⁴C]Put (Fig. 3). We believe that the apparent discrepancy is because of an underestimate of the real loss of this diamine in the 2E cells. The rationale is that during the first few hours after transfer of cells from [¹⁴C]Orn medium to label-free medium, the cells still contain a relatively large amount of [¹⁴C]Orn (Fig. 2D) that can be converted into [¹⁴C]Put by the mouse ODC in the 2E cells, thus replenishing some of the radioactive Put lost by the cells. In [¹⁴C]Put-feeding experiments, no such replenishment of the label in Put fraction can occur. Because our calculations of L₅₀ are based only upon the loss of radioactivity from the Put fraction at different times during the first 8-h period, the 2E cells add more radioactivity to the Put pool from [¹⁴C]Orn than the NT cells. To the best of our knowledge, this is the first report on direct estimate of the rates of turnover and half-life of Put in plants.

Importance of Put Degradation

From the published literature (Smith 1985; Cohen, 1998; Bagni and Tassoni, 2001) it is clear that the catabolic degradation of Put by DAO is not simply a means to regulate the cellular content of this diamine. Put catabolism provides intermediates that play important roles in the growth and development of plants as well as in the response of plants to various forms of abiotic stress. For example, GABA has been implicated to play a key role in signal transduction pathways during stress response of many plants (Ramputh and Bown, 1996; Bown and Shelp, 1997; Penel, 1997; Shelp et al., 1999) and also in the development of roots (Hausman et al., 1997a, 1997b). Likewise, Put catabolites serve as precursors of important alkaloids in several plants (Cohen, 1998; Hartmann, 1999). The degradation of Put also provides a metabolically crucial link between the polyamines and the TCA cycle, resulting in recycling of nitrogen as well as the carbon skeleton of Put. In addition, Put catabolism results in hydrogen peroxide production that is a substrate for peroxidation of lignin precursors in the cell wall (Federico and Angelini, 1988; Angelini et al., 1993). A specific role for DAO in cross-linking of amines to proteins as an alternative pathway to transglutaminases has also been suggested (Chiarello et al., 1996).

Whether or not increased catabolism of Put in the transgenic poplar cells has a physiological function beyond the removal of excess Put has not been investigated yet. Recycling of the carbon skeleton of polyamines would certainly allow the continued production of Glu, whose consumption is increased for Orn production to keep pace with its excessive utilization by the transgenic ODC. This would then create a futile cycle of Put biosynthesis and degradation in response to the presence of transgenic ODC and prevent the deleterious effects of excessive utilization of Orn and Glu, both of which are key metabolites serving as precursors of Arg, Pro, and several other amino acids. If a similar enhanced flux cycle was to function under conditions of abiotic stresses that cause overproduction of Put, one of its roles would be to continuously recycle the NH₃ often produced under stress-induced physiological reactions, thus minimizing its toxicity (Kronzucker et al., 2001).

Nanjo et al. (1999) have discussed the existence of such futile cycles that may interfere with the accumulation of a product in genetically engineered cells overexpressing a biosynthetic enzyme gene. For example, the biosynthesis of Pro from Glu by the enzyme P5C synthetase and its reconversion into Glu by a two-step process involving Pro dehydrogenase and P5C dehydrogenase (Verbruggen et al., 1996) constitutes a similar futile cycle that prevents overaccumulation of Pro in transgenic cells overexpressing a P5C synthetase gene. Under normal conditions, the cycle of Pro degradation is prevented by subcellular compartmentation of the catabolic enzymes in the mitochondria (Brandriss and Magasanik, 1981; Hare and Cress, 1997). Under hyperosmotic stress conditions, the production of Pro dehydrogenase is inhibited, thus allowing the maintenance of higher homeostatic levels of Pro in the cells. The importance of similar futile cycles in the regulation of Suc metabolism and Suc loading and unloading in plants has been discussed by Nguyen-Quoc and Foyer (2001).

Relevance of Cell Culture Studies to Whole Organisms

As discussed earlier, the poplar cell culture system used in the present study, although serving as an excellent experimental model for metabolic studies, lacks the natural features of transport in the whole plant. Thus, the cells themselves act as the sites of all biosynthetic as well as catabolic reactions. Although these cells can (and do) excrete Put into the medium, they do so only to a limited extent (less than 10% of the pool; P. Bhatnagar, unpublished data) and lack the features of intercellular transport. How much of a role the intercellular transport of polyamines plays in the homeostatic regulation of their cellular levels in plants has not yet been studied. Based upon the limited knowledge about cellular content of polyamines in transportation fluids (xylem and phloem saps; Friedman et al., 1986) in plants, we envision only a minor role for polyamine transport in this process. Because leaves and roots of plants do possess polyamine biosynthetic as well as catabolic enzymes, it can be argued that cellular levels of polyamines in these organs are a function of local reactions based upon the availability of precursors and the enzymes. The cell line being used in the present study is not capable of regeneration into whole plants; therefore, the 2E cells cannot be used to address this question. Parallel work on transformation of poplar cells capable of regeneration into whole plants is currently under way in our laboratory.

The results obtained so far with poplar cells are quite consistent with the published work on transgenic manipulation of polyamines in animals, and advance our knowledge about the metabolism of polyamines in plants. In the testis of transgenic mice overexpressing a human odc gene, Halmekytö et al. (1991) observed a significant increase in not only the activity of ODC but also the activities of SAMDC, Spd synthase, and Spm synthase. The catabolic enzymes Spd/Spm acetyltransferase and PAO were not affected. Other tissues (spleen, heart, kidney, and liver) only showed an increase in ODC and did not produce extra Spd and Spm (Halmekytö et al., 1993). Only testis and brain showed an increase in Put and Spd levels, the former increasing severalfold and the latter only slightly. A noteworthy observation in animal studies was that excess Put was readily secreted into the blood stream by the transgenic tissues except the brain and the testis. In this regard, the poplar cells differ in that the amount of Put secreted into the medium is relatively small.

Genetic Manipulation of a Metabolic Pathway. Implications of Modulating One Step

The present study clearly illustrates the need for a critical examination of the metabolic effects of genetically manipulating a metabolite that: (a) accumulates in large quantities, (b) is not an end product but is a substrate for other enzymes that produce physiologically important metabolites, and (c) is actively secreted as well as degraded in the cells (see also Kinney, 1998; DellaPenna, 2001). This is in contrast to the transgenic expression of genes whose products are novel proteins that do not have a metabolic function, e.g. Bt protein or virus coat proteins. The Put biosynthetic pathway represents an excellent example of a model pathway for analysis of its regulation through genetic manipulation. The results obtained thus far using this approach show that genetic manipulation of a single step in the polyamine biosynthetic pathway has pleiotropic effects on both the downstream as well as the upstream reactions in which the substrate of the transgene is used. Elevated levels of Put in the cells are not only accompanied by increased biosynthesis of Orn, the substrate (Bhatnagar et al., 2001), but also by an increase in subsequent biosynthetic (Spd biosynthesis) and degradative reactions. It is notable that the introduction of the transgenic ODC pathway does not adversely affect the native pathway for Put biosynthesis, i.e. via ADC. Although it is not known as to how an increase in Spd biosynthesis is brought about (i.e. is there a corresponding induction of SAMDC and Spd synthase?), the degradation of Put occurs by a constitutive mechanism without involving increased production of the DAO. It is envisioned that there most likely are further downstream effects on Spd and Spm metabolism that contribute to the maintenance of their homeostasis in the cells. It is also expected that there will be additional effects on pathways that interact with the polyamine biosynthetic pathway, e.g. excessive utilization of Glu could affect the biosynthesis of other amino acids as well as the assimilation of NH₃ by the cells. These aspects of the metabolic effect of Put overproduction are currently being investigated.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS LITERATURE CITED

Culture Conditions

Liquid and solid cultures of hybrid poplar (Populus nigra × maximowiczii) cells were maintained in Murashige and Skoog medium (Murashige and Skoog, 1962) containing vitamins of B-5 medium (Gamborg et al., 1968), 2% (w/v) Suc, and 0.5 mg L¹ 2,4-dichlorophenoxyacetic acid. The pH of the medium was adjusted to 5.7 before autoclaving. Suspension cultures were maintained by transferring 7 mL of the 7-d-old cell suspensions to 50 mL of fresh medium in a 125-mL Erlenmeyer flask, and kept on a gyratory shaker at 160 rpm. Stock cultures were maintained on solid medium (1.0% [w/v] agar type A, Sigma, St. Louis) and subcultured at 3- to 4-week intervals. All cultures were kept at 25°C ± 1°C under 12-h photoperiod (80 ± 10 µE m² s¹). The medium for maintenance of transgenic cells contained 100 mg L¹ kanamycin; however, the antibiotic was not present during the experimental treatments (Bhatnagar et al., 2001).

Precursor Feeding Experiments

Three-day-old cells of the transgenic cell line 2E, transformed with a mouse (Mus musculus) odc cDNA under the control of a 2× 35S cauliflower mosaic virus promoter (Bhatnagar et al., 2001), and a corresponding NT cell line were used for this study. These cells had earlier been characterized for the presence of the foreign DNA, the ODC activity resulting from its expression, and its effects on polyamine biosynthesis (Bhatnagar et al., 2001). At the time of starting the experiments, twice the normal amount (i.e. 14 versus 7 mL) of 7-d-old cells were added to 50 mL of fresh culture medium to achieve a cell density of about 1.0 g per 10 mL of culture. At the beginning of treatments (i.e. after 3 d of culture), the flasks were placed in the hood for 10 min to let the cells settle at the bottom of the flask, and 24 mL of the medium was decanted from the top. To the remaining cells in the flask, either 50 µL of radioactive Put (1 µCi of [1,4-¹⁴C]Put-diHCl; specific activity 107 mCi mmol¹; Amersham Life Science, Elk Grove, IL) or 20 µL of radioactive Orn (1µCi of L-[U-¹⁴C]Orn-HCl; specific activity 261 mCi mmol¹; Amersham Life Science) were added and the flasks placed back on the shaker. After incubation with the radioactive substrates for 2h, the suspensions were transferred to 50 mL of sterile polypropylene centrifuge tubes and centrifuged at 1,000 rpm in a swinging bucket rotor for 3 min to pellet the cells. The medium was decanted, the cells were washed with 30 mL of fresh Murashige and Skoog medium per tube by centrifugation, the pellet transferred to 50 mL of fresh medium, and placed back on the shaker. Aliquots (5-8 mL) were removed from each flask at different times (0, 2, 4, 8, 24, 72, and 96 h for [U-¹⁴C]Orn-treated cells and 0, 2, 4, 8, 24, 72, and 84 h for [1,4 ¹⁴C]Put-treated cells). The cells were vacuum filtered on Miracloth (Calbiochem, La Jolla, CA), weighed, and mixed with 7.5% (v/v) PCA (1.0 g cells:2 mL of PCA), and frozen at 20°C for polyamine analysis.

Analysis of Free Polyamines

Tissue samples in PCA were frozen (20°C overnight) and thawed (3-4 h at room temperature) three times before dansylation (Minocha et al., 1994). After final thawing, the cells were vortexed and centrifuged at 14,000g for 10 min. For determination of the efficiency of extraction of polyamines and other radioactive derivatives, the pellet from the above centrifugation step was re-extracted twice with 2.0 mL of 5% PCA, and aliquots from each fraction counted for radioactivity. Aliquots (50 µL) of PCA extracts from sample collections at different times after transfer of cells to label-free medium were counted to determine the total loss of radioactivity from the cells into the medium. For analysis of radioactivity in the three major polyamines, 1.0 mL of the first supernatant fraction was dansylated according to Andersen et al. (1998) as modified by Bhatnagar et al. (2001). The dansylated polyamines were partitioned into toluene, the toluene fraction was vacuum dried, and the dansyl-polyamines were reconstituted in 50 µL of 100% methanol. Radioactivity in each fraction was counted to follow the recovery of radioactive polyamines through various steps. Of the 50 µL of methanol, an aliquot of 3.75 µL was mixed with 496.25 µL of 100% methanol and used to determine total polyamines by HPLC (Minocha et al., 1990, 1994). From the remaining methanol extract, 40-µL samples were spotted on TLC plates and developed in a solvent mixture of chloroform:triethylamine (5:1 [v/v]). The bands of the three polyamines were visualized under UV light, marked with a lead pencil, scraped, and counted for radioactivity. An aliquot of 20 µL from the aqueous phase (after extraction with toluene) was also counted for radioactivity. This fraction contains the unused [¹⁴C]Orn, metabolic breakdown products of Put, and other amino acids derived from labeled precursors.

DAO Enzyme Assay

Three- to 6-d-old cells were used to optimize conditions for assaying the enzyme activity of DAO by modifying the procedure of Santanen and Simola (1994) and Santanen (2000). The conditions were optimized with respect to the amount of cells used (25-300 mg fresh weight per assay), pH of the buffer (6.0, 7.0, and 8.0), time of incubation (15-120 min), and the use of frozen-thawed cells versus cell homogenates. In a typical assay, cells were collected by vacuum filtration, and 200 mg fresh weight of cells were added to 550 µL of 0.1 M potassium phosphate buffer (pH 8.0) containing 2.0 mM dithiothreitol in a 2-mL microfuge tube. The cells were frozen and thawed three times and vortexed before assay. To each tube, 50 µL of labeled substrate (0. 1 µCi of [1,4-¹⁴C]Put in 1.0 mM cold Put) was added, the tubes were vortexed briefly, and they were placed on a shaker (100 rpm) horizontally at 37°C. After incubation for 60 min, the reaction was terminated by adding 150 µL of saturated sodium carbonate. Labeled ¹-pyrroline was extracted immediately by partitioning with 500 µL of Photrex grade toluene (Fisher Scientific, Fairlawn, NJ). The samples were vortexed for 1 min and centrifuged at 10,000 rpm for 5 min. Three hundred microliters of the toluene layer was removed and counted in 10 mL of Scintilene (Fisher Scientific). For comparison of the enzyme activity in frozen-thawed cells versus homogenates, the cells were homogenized using a Polytron, the homogenate was centrifuged at 14,000g for 20 min, and the supernatant and the pellet used for enzyme assays separately.

Calculations and Statistical Analysis

After the determination of radioactivity in different polyamine fractions by TLC and the amounts of total soluble polyamines in the cells by HPLC, the following formulae were used for various calculations presented here:

Loss of Put over 2h = [(dpm in Put at 0 h  dpm in Put at 2 h)/dpm in Put at 0 h] × mean¹ amounts of Put at 0 to 2 h

Loss of Put over 8h = [(dpm in Put at 0 h  dpm in Put at 8 h)/dpm in Put at 0 h] × mean² amounts of Put during 0 to 8 h

% loss of Put (24-72 h) = [(dpm in Put at 24h  dpm in Put at 72 h)/dpm in Put at 24 h] × 100

Specific activity of Spd = dpm in Spd/nmol Spd g¹ fresh weight at a given time

Amount of Put converted into Spd in 8 h = [(dpm in Spd at 8 h  dpm in Spd at 0 h)/mean³ dpm in Put at 0 to 8 h] × mean² amounts of Put during 0 to 8 h

where mean¹ Put = [nmol Put at 0 h + nmol Put at 2 h]/n ( n = total number of samples), mean² Put = [nmol Put at 0 h + nmol Put at 2 h + nmol Put at 4 h + nmol Put at 8 h]/n (n = total number of samples), and mean³ dpm in Put = dpm Put at 0 h + dpm Put at 2 h + dpm Put at 4 h + dpm Put at 8 h]/n (n = total number of samples).

Each treatment included at least three replicates and each experiment was done at least two or three times. The results of the replicate experiments were quite consistent. Except where specified, data from two or three experiments were pooled together. The data were subjected to one-way ANOVA using Systat version 9 (SPSS, Inc., Chicago). Student's t test was used to determine significance at P  0.05. The statistical comparisons were usually made between the NT and the 2E only at a given time, except when stated otherwise. Determination of initial half-life (L₅₀) for turnover of Put was done using data for the first 8 h from two or three different experiments.

Distribution of Materials

Upon request, all novel materials described in this publication will be made available in a timely manner for noncommercial research purposes, subject to the requisite permission from any third party owners of all or parts of the material. Obtaining any permissions will be the responsibility of the requestor.