Skip to main content
NCBI home page
Plant Physiology logoLink to Plant Physiology
. 1998 Jan;116(1):137–144. doi: 10.1104/pp.116.1.137

Effects of Sulfanilamide and Methotrexate on 13C Fluxes through the Glycine Decarboxylase/Serine Hydroxymethyltransferase Enzyme System in Arabidopsis1

Vikram Prabhu 1,*, K Brock Chatson 2, Helen Lui 1, Garth D Abrams 1, John King 1
PMCID: PMC35151  PMID: 9449840

Abstract

In C3 plants large amounts of photorespiratory glycine (Gly) are converted to serine by the tetrahydrofolate (THF)-dependent activities of the Gly decarboxylase complex (GDC) and serine hydroxymethyltransferase (SHMT). Using 13C nuclear magnetic resonance, we monitored the flux of carbon through the GDC/SHMT enzyme system in Arabidopsis thaliana (L.) Heynh. Columbia exposed to inhibitors of THF-synthesizing enzymes. Plants exposed for 96 h to sulfanilamide, a dihydropteroate synthase inhibitor, showed little reduction in flux through GDC/SHMT. Two other sulfonamide analogs were tested with similar results, although all three analogs competitively inhibited the partially purified enzyme. However, methotrexate or aminopterin, which are confirmed inhibitors of Arabidopsis dihydrofolate reductase, decreased the flux through the GDC/SHMT system by 60% after 48 h and by 100% in 96 h. The uptake of [α-13C]Gly was not inhibited by either drug class. The specificity of methotrexate action was shown by the ability of 5-formyl-THF to restore flux through the GDC/SHMT pathway in methotrexate-inhibited plants. The experiments with sulfonamides strongly suggest that the mitochondrial THF pool has a long half-life. The studies with methotrexate support the additional, critical role of dihydrofolate reductase in recycling THF oxidized in thymidylate synthesis.

THF coenzymes are required in the synthesis of thymidylate, purine nucleotides, amino acids, and in organellar protein synthesis (Cossins, 1987; Appling, 1991). In many cellular systems the single carbons involved in THF-dependent processes are derived from the β-carbon of Ser (Schirch, 1984; Narkewicz et al., 1996). SHMT catalyzes the transfer of a methylene group from Ser to THF for direct use in thymidylate synthesis; alternatively, CH2-THF is reduced to CH3-THF by methylene-THF reductase or oxidized to HCO-THF by C1-THF-synthase (Appling, 1991; Nour and Rabinowitz, 1991) and then used in other THF-dependent cellular processes. In C3 plants the GDC/SHMT enzyme system is considered to be the major pathway for the generation of single-carbon units via Ser. This is a consequence of photorespiration, which requires large amounts of Gly to be metabolized by this enzyme system (Oliver, 1994). In Arabidopsis thaliana the flux of single carbons into Ser via the GDC/SHMT pathway is 4-fold greater than that through the alternative C1-THF synthase/SHMT pathway (Prabhu et al., 1996a). Below we describe some biochemical relationships between enzymes that regenerate the cofactor THF (Fig. 1) and carbon flux through the THF-dependent GDC/SHMT enzyme system in A. thaliana (L.) Heynh. Columbia wild type.

Figure 1.

Figure 1

Open in a new tab

The enzymatic reactions and intermediates in the biosynthesis of THF. HPPK, 6-Hydroxymethyl-7,8-dihydropterin pyrophosphokinase; DHPS, dihydropteroate synthase (EC 2.5.1.15); DHFS, dihydrofolate synthase (EC 6.3.2.12); DHFR (EC 1.5.1.3); and FPGS, folylpolyglutamate synthetase (EC 6.3.2.17).

Antagonists of THF biosynthesis have been developed for clinical use as antiproliferative and antimicrobial agents, which exploit the critical role that THF coenzymes play in cellular metabolism (Pratt and Taylor, 1990). Sulfanilamide and its analogs (Fig. 2) inhibit DHPS (Fig. 1), whereas methotrexate (Fig. 2) and its analogs inhibit DHFR (Fig. 2) isolated from a number of organisms (Cossins, 1987; Pratt and Taylor, 1990; Schweitzer et al., 1990). DHPS is involved in the de novo pathway of THF biosynthesis, whereas DHFR has a dual role (Fig. 1); it is involved in the reduction of dihydrofolate to THF from both the de novo pathway and that arising from the oxidation of THF during TS activity. These drugs may be useful in manipulating the availability of THF in higher plants, hence permitting a better understanding of its metabolism.

Figure 2.

Figure 2

Open in a new tab

Structures of folate analogs. Note that the commercial leucovorin used in our experiments contained both the physiological stereoisomer depicted here and the nonphysiological stereoisomer.

We previously used a combined dosage of sulfanilamide and methotrexate with qualitative NMR observations to demonstrate the requirement of THF for Ser synthesis via the GDC/SHMT and C1-THF synthase/SHMT pathways in Arabidopsis (Prabhu et al., 1996a). Combined dosages have also been used to demonstrate the THF-dependent metabolism of Ser through the C1-THF synthase/SHMT pathway in yeast (Pasternack et al., 1994). The Arabidopsis study suggested that THF metabolism in this organism is organized differently from that in other eukaryotes. Recent studies of pea leaves suggested that THF metabolism is compartmented largely in the mitochondria (Neuburger et al., 1996), unlike that in other organisms (Appling, 1991). Here, using NMR, we quantify the fluxes of carbon through the GDC/SHMT enzyme system in Arabidopsis, as influenced by exposure to drugs that inhibit DHPS and DHFR, which are key enzymes that produce and maintain cellular THF levels. The results shed new light on interactions between the THF pathway and THF-dependent Gly metabolism in plant mitochondria.

MATERIALS AND METHODS

[α-13C]Gly and sodium salt 3-(trimethylsilyl)propionic-2,2,3,3-d4 acid (TSP), were purchased from Cambridge Isotopes Laboratories Inc. (Andover, MA). All other chemicals were from Sigma.

Arabidopsis thaliana (L.) Heynh. Columbia Culture Conditions and 13C-Labeling Procedures

The procedures for the growth and 13C labeling of plants and the preparation of leaf extracts were essentially as described previously (Prabhu et al., 1996a). All experiments were repeated three or more times and the data are presented as the means with respective se values.

NMR Parameters

NMR spectra were collected on an AMX 500-MHz instrument (Bruker, Billerica, MA). 13C chemical shifts from natural abundance spectra of authentic samples were first obtained in 100 mm phosphate buffered to pH 7.3 and 25°C, internally referenced to 2,2-dimethyl-2-silapentane-5-sulfonic acid, sodium salt (Prabhu et al., 1996b). Assignments of resonances in the extracts were confirmed by spiking the extracts with authentic samples (Prabhu et al., 1996a).

Quantitative analyses of 13C-enriched compounds in the plants were accomplished using extracts. Two sets of acquisition parameters were used: (a) inverse-gated decoupled experiments with a 20-s delay time and (b) standard broad-band experiments, details of which have been described previously (Prabhu et al., 1996a). All peaks were calibrated relative to a solution of 2 mm sodium [13C]formate in 100 mm potassium phosphate (pH 7.3), sealed in a capillary and inserted concentrically into the 5-mm tube containing the sample. The 13C-13C multiplets versus 13C-12C center-line singlets of the protonated carbons of interest were quantitated using integration and isotopomer analyses; their intensities are independent of potential differences in T1 and nuclear Overhauser effects (London et al., 1975; Suzuki et al., 1975), especially under conditions of gated decoupling with long delay times (Gadian, 1982), as used here. However, the differential intensities of the β-13C versus the α-13C in Ser under broad-band decoupling were corrected for using data generated from standard curves of authentic compounds obtained under similar experimental conditions.

1H spectra were acquired using a 9.7-μs (90°) pulse, a spectral width of 7812 Hz, an acquisition time of 2 s, and a delay time of 1 s. The sample temperature was maintained at 25°C and 32,000 data points were acquired for each sample. Attenuation of the water resonance was achieved by presaturation (2 s at 40 dB). One hundred twenty scans were acquired for each sample and a line broadening of 0.3 Hz was used in the processing of the free induction decay. Chemical shifts from natural abundance spectra of authentic samples were first obtained in 100 mm phosphate buffered to pH 7.3 and 25°C, referenced internally to 2,2-dimethyl-2-silapentane-5-sulfonic acid and externally to a 10 mm solution of TSP in 100 mm potassium phosphate (pH 7.3), sealed in a capillary and inserted concentrically into the 5-mm tube containing the sample. The spectra of the plant extracts were acquired with the external standard, TSP only; the singlet of the trimethyl resonance of TSP was set to zero. Assignments of peaks in the extracts were confirmed by spiking the sample with pure compounds.

Extraction and Assay of DHFR

Root material from plants grown in liquid culture was harvested, blotted dry with paper towels, and frozen at −80°C. The tissue was then ground in a mortar on ice using a small amount of acid-washed sand. A buffer containing 100 mm potassium phosphate (pH 7.5), 1 mm EDTA, 1 mm PMSF, 10% (v/v) glycerol, and 20 mm 2-mercaptoethanol was added while grinding until a fine slurry was achieved. The slurry was filtered through three layers of cheesecloth and centrifuged at 10,000g for 10 min in a JA-25.5 rotor using an Avanti J-25 instrument (Beckman). The supernatant was treated with solid ammonium sulfate, and protein precipitating between 30 and 60% ammonium sulfate saturation was removed by centrifugation as described above and stored overnight at −20°C. Storage overnight resulted in better activity than if the enzyme was assayed directly after salt precipitation. The protein was desalted over a Sephadex G-25 column (Pharmacia) and its concentration was estimated by the method of Bradford (1976). DHFR activity was assayed spectrophotometrically (Misra et al., 1961). The assay was done in a final volume of 500 μL containing 100 mm potassium phosphate (pH 7.5), 100 μm NADPH, 100 μm dihydrofolic acid, 500 μg of protein, and various concentrations of methotrexate or aminopterin. The enzyme activity was monitored by recording the change in A340 once every 30 s for 10 min, using a photodiode array spectrophotometer (DU 7400, Beckman). Controls lacking dihydrofolate were used to subtract nonspecific oxidation of NADPH. Specific activity was expressed as nanomoles of dihydrofolate reduced to THF per milligram of protein per hour.

RESULTS

Metabolism of [α-13C]Gly to Ser as Detected by 13C NMR

The rationale for the 13C-enrichment patterns of Ser in Arabidopsis plants supplied with [α-13C]Gly has been described previously (Prabhu et al., 1996a). For this study only [α,β-13C]Ser and [β-13C]Ser, the two species unequivocally ascribable to synthesis via GDC/SHMT activities, were taken into account; [α-13C]Ser could be synthesized using the alternative C1-THF synthase pathway (Prabhu et al., 1996a). For a measure of flux through the GDC/SHMT enzyme system the 13C-enriched isotopomers of Ser described above were quantitated after 6 h of supply of [α-13C]Gly. This was based on the relative patterns and concentrations of isotopomers described from time-course experiments in which we did not detect signals from potential products of further metabolism of Ser such as hydroxypyruvate or other sugars (Prabhu et al., 1996a). Other NMR studies of Gly metabolism in plant cells also noted little metabolism of the Ser produced via the GDC/SHMT pathway in comparable periods (Ashworth and Mettler, 1984; Neeman et al., 1985). In some animal cellular systems the de novo incorporation of amino acids into proteins has been shown to be sufficient to warrant correction in measurements of flux for specific pathways (Flogel et al., 1997). However, even in hydrolyzed extracts of our samples we did not see NMR signals that would indicate incorporation of 13C-enriched compounds into macromolecules (V. Prabhu, B. Chatson, G. Abrams, and J. King, unpublished data). Thus, our quantitative measurements of Ser accumulation are a good representation of flux through the GDC/SHMT enzyme system, which we express as micromoles of 13C-enriched Ser/g fresh weight/6 h.

Individual Effects of Methotrexate and Sulfanilamide

Experiments were first performed with sulfanilamide (Fig. 3a) or methotrexate (Fig. 3b) at a range of concentrations to evaluate their use in subsequent experiments. Nonlimiting concentrations were then used to examine potential differences in the effects of the two classes of drugs. Methotrexate-treated plants displayed a strong reduction in the flux of 13C through the GDC/SHMT system, whereas with sulfanilamide only a small effect was observed (Table I). After 96 h of exposure to the drugs, this flux could no longer be detected in methotrexate-treated plants, whereas with sulfanilamide only a slight reduction with respect to the control was detected.

Figure 3.

Figure 3

Open in a new tab

Effects of varying concentrations of inhibitors on the flux of [α-13C]Gly through the GDC/SHMT enzyme system in Arabidopsis leaves. a, Sulfanilamide; b, methotrexate. Plants were exposed to inhibitors for 48 h and then supplied with [α-13C]Gly for 6 h.

Table I.

Relative effectiveness of methotrexate (100 μm) versus sulfanilamide (2 mm) in reducing flux through the GDC/SHMT enzyme system

Time Methotrexate Percent Sulfanilamide Percent
h μmol Ser g−1 fresh wt 6 h−1 μmol Ser g−1 fresh wt 6 h−1
0 1.1  ± 0.1 100 1.1  ± 0.1 100
48 0.5  ± 0.04 45 0.89  ± 0.05 81
96 0 0 0.88  ± 0.04 80
Open in a new tab

The data are expressed as the means and respective ses of flux.

Time-Dependent Effects of Simultaneous Exposure to Methotrexate and Sulfanilamide

Plants exposed simultaneously to methotrexate (100 μm) and sulfanilamide (2 mm) showed a time-dependent reduction in the flux of 13C through the GDC/SHMT system, although the cellular concentrations of [α-13C]Gly did not decrease (Fig. 4). The reduction was approximately 50% in the first 24 h, but a 96-h period was required to completely reduce the Ser signal to that of the background. No additive effect of sulfanilamide to methotrexate was obvious, because the results were similar to those obtained with methotrexate alone (Table I). At 0 h of drug exposure the concentration of [α-13C]Gly was slightly lower than at other time points, coinciding with the maximum flux. Quantitative isotopomer analyses showed that the exposure to antifolates also resulted in an alteration in the pools of single carbon (Table II). The proportion of the dually enriched [α,β-13C]Ser was reduced, whereas those of the singly enriched species, [α-13C]Ser and [β-13C]Ser, increased. Based on the experiments described in Table I, it appears that the changes in isotopomer proportions were caused by methotrexate alone (not shown).

Figure 4.

Figure 4

Open in a new tab

Time-dependent effects of combined exposure to methotrexate plus sulfanilamide on flux through the GDC/SHMT enzyme system in Arabidopsis leaves.

Table II.

Relative proportions of the three species of 13C-enriched Ser after treatment with methotrexate (100 μm) plus sulfanilamide (2 mm)

Time [α,β-13C]Ser [β-13C]Ser [α-13C]Ser
h %
0 64  ± 10 12  ± 4 24  ± 6
24 57  ± 8 13  ± 5 30  ± 6
48 46  ± 6 16  ± 3 38  ± 4
Open in a new tab

The data are expressed as the percentage of total 13C-enriched Ser and the experimental details are as described for Figure 5. There are no data for 96 h because the [13C]Ser signal was not distinguishable from the baseline noise.

Comparison of the Effects of Alternative Analogs

We examined the effect of aminopterin (Fig. 2), which like methotrexate is a competitive inhibitor for DHFR, on reduction of THF availability for the operation of the GDC/SHMT enzyme system in Arabidopsis. A 48-h exposure to aminopterin, as with methotrexate, reduced the flux by about 60% (Table III), and after exposures of 96 h, the resonance of 13C-enriched Ser could not be distinguished above the background in the NMR spectra from either treatment (not shown). The uptake of Gly by plants was unaffected by either aminopterin or methotrexate treatment (not shown).

Table III.

Comparison of the effectiveness of alternative analogs of sulfanilamide and methotrexate in reducing flux through the GDC/SHMT enzyme system after 48 h of drug exposure

Treatment Ser Percent
μmol g−1 6 h−1
Control 1.2  ± 0.1 100
Sulfanilamide 0.97  ± 0.04 81
Sulfadiazine 1.1  ± 0.05 91
Methotrexate 0.46  ± 0.07 38
Aminopterin 0.41  ± 0.04 34
Open in a new tab

The concentrations of sulfonamides were 4 mm, whereas those of methotrexate and aminopterin were 200 μm.

A partially purified preparation of DHFR enzyme activity from Arabidopsis leaves was completely inhibited by 1 μm methotrexate or aminopterin (data not shown). However, high levels of a nonspecific NADPH-oxidizing activity were present in this leaf extract, which did not allow nonlimiting concentrations of NADPH to be maintained for the determination of I50 values. These experiments were then attempted using root preparations that had reduced levels of the nonspecific NADPH-oxidizing activity. Controls were included to determine the nonspecific oxidation of NADPH. Methotrexate and aminopterin were strong inhibitors of the root DHFR preparation, with I50 values of less than 10 nm for both analogs (Fig. 5). Germination of seeds was completely inhibited, and growth (change in fresh mass) of 3-week-old plants was severely reduced in flask culture in the presence of both analogs (data not shown).

Figure 5.

Figure 5

Open in a new tab

Inhibition of DHFR enzyme activity in in vitro assays by methotrexate and aminopterin. The 100% activity was 75 ± 5 nmol (mean ± se, n = 3) of THF mg−1 protein h−1.

Experiments in which the effect of sulfadiazine (Fig. 2) was compared with that of sulfanilamide revealed that this analog also did not reduce flux through the GDC/SHMT enzyme system in vivo (Table III). A third analog, sulfacetamide, was also tested and yielded similar results (not shown). We used 1H NMR spectroscopy to determine whether individual sulfonamides were present in the shoots. When plants were supplied with sulfanilamide for 48 h, the accumulation of this compound in the shoots could be detected from the resonances of the aromatic hydrogens (Fig. 6). The endogenous metabolite fumarate was present in all of the samples that were examined. The spectra are plotted using the intensity of the fumarate peak as a convenient internal standard; the intensity of the fumarate peak and that of the external reference in replicates of control and sulfonamide-treated plants were not significantly different from each other. When sulfacetamide was supplied to Arabidopsis, this compound could be detected in the shoots from the aromatic resonances, as well as the methyl resonance of the acetyl group (not shown). These experiments confirmed that the supplied sulfonamides were transported to and accumulated in quantity in the shoot tissues.

Figure 6.

Figure 6

Open in a new tab

1H NMR detection of the accumulation of sulfanilamide in Arabidopsis shoots. a, Spectrum of control plants showing the area of interest that contains the endogenous metabolite fumarate. b, Plants supplied with 5 mm sulfanilamide for 48 h; the aromatic proton resonances of this compound are indicated in the spectrum. c, Spectrum of the sample shown in b spiked with sulfanilamide.

In vitro experiments using a partially purified preparation of the DHPS from Arabidopsis showed that all three analogs were highly potent inhibitors, with I50 values of less than 20 μm (Prabhu et al., 1997). Germination of seeds was completely inhibited and growth of 3-week-old plants was reduced in flask culture in the presence of all three analogs added singly (data not shown).

Specificity of the Inhibition by Methotrexate

To determine if the decrease in flux after methotrexate exposure was attributable to depletion of THF levels by inhibition of DHFR, or rather to direct inhibition of the GDC/SHMT enzymes by polyglutamylated forms of methotrexate, we performed rescue experiments using leucovorin (Fig. 2). Figure 7 shows that the supply of leucovorin after complete methotrexate inhibition restored flux through the GDC/SHMT enzyme system. During a 24-h period the plants supplied with leucovorin showed a measurable flux, indicating that there was little direct inhibition of these enzymes by methotrexate (Table IV). The proportions of the individual species of 13C-enriched Ser in the leucovorin-rescued plants were considerably different from those in the control (Table V). (Note that the quantities of the isotopomers of [α-13C]Ser versus [β-13C]Ser were corrected for differential intensities and do not correspond directly with the intensities of the individual peaks in Fig. 7.) In methotrexate-plus-leucovorin-treated plants, the proportions of both the singly enriched species, [α-13C]Ser and [β-13C]Ser, increased relative to that of [α,β-13C]Ser. This was similar to the observation recorded in Table II.

Figure 7.

Figure 7

Open in a new tab

Leucovorin rescue of methotrexate inhibition of flux through the GDC/SHMT enzyme system in Arabidopsis leaves. a, Broad-band decoupled spectra of control plants supplied with [α-13C]Gly for 24 h. b, Plants exposed to 200 μm methotrexate for 96 h, transferred to methotrexate-free media for 12 h, and then supplied with [α-13C]Gly for 24 h. c, Plants exposed to 200 μm methotrexate for 96 h, transferred to methotrexate-free media containing 1 mm leucovorin for 12 h, and then supplied with [α-13C]Gly for 24 h.

Table IV.

Leucovorin rescue after methotrexate inhibition of Ser synthesis

Treatment Ser Percent
μmol g−1 24 h−1
Control 3.2  ± 0.15 100
Methotrexate (96 h) 0 0
Methotrexate (96 h); leucovorin (12 h) 1.1  ± 0.05 35
Open in a new tab

The methotrexate-treated plants were transferred to fresh media lacking methotrexate, with or without leucovorin for 12 h. Subsequently, [α-13C]Gly (1 mm) was supplied for 24 h and the flux through the GDC/SHMT enzymes was determined.

Table V.

Relative proportions of the three species of 13C-enriched Ser in leucovorin rescue experiments

Treatment [α,β-13C]Ser [β-13C]Ser [α-13C]Ser
%
Control 68.5  ± 6 10.5  ± 2 21  ± 3
Methotrexate + leucovorin 42  ± 4 18  ± 3 40  ± 5
Open in a new tab

The treatments are the same as those described in Table IV. The data are expressed as the percentage of total 13C-enriched Ser. There are no data for the methotrexate-only treatment (Table IV).

DISCUSSION

The sulfonamide class of inhibitors caused only a slight reduction in THF-dependent flux through the GDC/SHMT enzyme system in Arabidopsis. Studies of the target enzyme showed that the sulfonamide analogs were strong inhibitors of the Arabidopsis DHPS (Prabhu et al., 1997). 1H NMR studies confirmed that under the conditions of our experiments, the supplied sulfonamides were present in the shoot tissue. The continued metabolism of Gly in plants exposed to sulfonamides is unlikely to have occurred using de novo-synthesized THF. Therefore, our results suggest that the mitochondrial THF pool is either large and/or has a long half-life.

High levels of folate were recorded in mitochondria isolated from pea leaves (Neuburger et al., 1996); the large mitochondrial folate pools likely reflect the requirement of C3 plants to metabolize large amounts of Gly to Ser (Oliver, 1994). Studies of the partially purified DHPS from Arabidopsis (Prabhu et al., 1997) lend some indirect support to the idea that THF levels in Arabidopsis may reach high or nonlimiting levels. The product of the biosynthetic reaction, dihydropteroic acid, competitively inhibits the enzyme activity. The DHPS from pea leaves was also found to be inhibited in this manner (Rebeille et al., 1997). The existence of such a mechanism indicates that product feedback inhibition could regulate the flow of metabolites when THF levels are sufficient to meet cellular requirements.

Methotrexate may inhibit THF-dependent enzymes in vivo after undergoing polyglutamylation as observed in animal cells (McGuire and Coward, 1984; Kim et al., 1993). However, our leucovorin rescue experiments strongly suggested that the effects of methotrexate were restricted to the inhibition of DHFR and that there was little, if any, direct inhibition of GDC/SHMT enzymes. Leucovorin is widely used in animal systems as a rescue agent from methotrexate toxicity (Stover and Schirch, 1993). Because it is a fully reduced form of folic acid, its supply can alleviate the inability of methotrexate-inhibited DHFR to supply THF. Also, in previous studies of the uptake and metabolism of methotrexate by plant cells it was found that methotrexate was a poor substrate for folylpolyglutamate synthetase (see Fig. 1) from Datura innoxia in in vitro assays, and that methotrexate polyglutamylation in the cells themselves was only slight (Wu et al., 1993, 1994). Thus, unlike in animal cells, polyglutamylation in plant cells may not be a significant factor in the action of methotrexate.

DHFR plays an important role in the cellular functioning of all organisms. Our studies strengthen the idea of a crucial role for DHFR in the reduction of dihydrofolate produced by oxidation of THF by TS activity. In plants, including Arabidopsis, DHFR and TS are encoded by bifunctional genes coding for bifunctional polypeptides (Lazar et al., 1993; Luo et al., 1993, 1997; Wang et al., 1996). The direct channeling of dihydrofolate between the TS and DHFR domains in bifunctional polypeptides has been proposed (Knighton et al., 1994). In plants thymidylate synthesis may occur exclusively in the mitochondria (Neuburger et al., 1996). This could explain the rapid depletion of the THF required for mitochondrial Gly metabolism in Arabidopsis when DHFR activity is inhibited by methotrexate. Thus, methotrexate treatment had a more dramatic effect than the sulfonamides on reducing the availability of THF for mitochondrial Gly metabolism.

Methotrexate is a useful drug to manipulate THF availability in plants. Exposure of Arabidopsis plants to methotrexate resulted in changes in the isotopomers of [13C]Ser produced by the GDC/SHMT system. The change in isotopomers could reflect an alteration in the pools of active THF (HCO-, CH-, CH2-, and CH3-) in the cellular system arising from methotrexate action. In mammalian cells rapid alterations in the proportions of one-carbon THF occur after exposure to methotrexate, along with an increase in the amount of dihydrofolates (Allegra et al., 1986).

In Arabidopsis methotrexate treatment reduced THF availability, which in turn reduced the flux through the GDC/SHMT system. The reduced flux caused endogenous (photorespiratory) Gly concentrations to rise and increased the 12CH2-THF:13CH2-THF ratio; this was recorded by NMR as a decrease in the proportion of [α,β-13C]Ser relative to the singly enriched species. The ratio of [α-13C]Ser to [β-13C]Ser was approximately 2:1 in control and methotrexate-treated plants. The 2:1 ratio in control plants is reflective of the added flux through the C1-THF-synthase pathway, resulting in higher levels of [α-13C]Ser (Prabhu et al., 1996a). The maintenance of the 2:1 ratio in methotrexate-treated plants suggests that the GDC activity was not directly inhibited by methotrexate. Furthermore, the pools of THF used by the two pathways are well equilibrated, because reduction in the [α-13C]Ser signal paralleled that of [β-13C]Ser. These suggestions are also supported by the leucovorin rescue experiments, which showed the ability to restore flux through the GDC/SHMT and C1-THF-synthase pathways (again 2:1 ratio).

Our antimetabolite experiments suggest several additional ideas about THF metabolism in Arabidopsis. The lack of reduction in Gly metabolism by sulfonamides, and the long time period required for complete inhibition by methotrexate, indicate that cellular folate pools have a relatively long half-life. The leucovorin rescue experiments indicate that reduced folates can be transported across mitochondrial membranes because the supplied leucovorin was able to restore mitochondrial metabolism of Gly. In other eukaryotes the parallel paths of THF metabolism in the cytosol and mitochondria are largely interconnected by transport of single carbons in the form of formate, Gly, or Ser between these two cellular compartments (Appling, 1991). However, information on the compartmentation of THF metabolism and thymidylate synthesis in plants is limited, and the few individual studies have suggested quite different pictures of these processes in plants (Cella et al., 1991; Huangpu et al., 1996; Neuburger et al., 1996; Luo et al., 1997).

To our knowledge, this study demonstrated for the first time in a higher plant, and as far as we know in any cellular system, the distinctly different effects of methotrexate and sulfanilamide on THF-dependent metabolism. The NMR approach to examine these problems was particularly useful. The results shed new light on the regulation of THF metabolism in plants and accentuate the dual role played by DHFR in contrast to that of DHPS. The direct monitoring of the effects of these drugs on the major pathway producing carbons for use in other THF-dependent processes is our first step toward understanding the relationships between THF biosynthesis and THF-dependent cellular metabolism in plants. Studies of THF-dependent processes that occur outside the mitochondria should lead to a better understanding of the compartmentation of THF metabolism in higher plants.

ACKNOWLEDGMENTS

We acknowledge the useful suggestions of the anonymous reviewers.

Abbreviations:

CH2-THF

5,10-methylene-THF

CH3-THF

5-methyl-THF

DHFR

dihydrofolate reductase

DHPS

dihydropteroate synthase

GDC

Gly decarboxylase complex

HCO-THF

10-formyl-THF

I50

concentration of inhibitor resulting in 50% reduction in enzyme activity

leucovorin

5-formyl-THF

SHMT

Ser hydroxymethyltransferase

THF

tetrahydrofolate

TS

thymidylate synthase

Footnotes

1

This study was supported in part by grants-in-aid of research from the Natural Sciences and Engineering Research Council of Canada to J.K.

LITERATURE  CITED

  1. Allegra CJ, Fine RL, Drake CJ, Chabner BA. The effect of methotrexate on intracellular folate pools in human MCF-7 breast cancer cells: evidence for direct inhibition of purine synthesis. J Biol Chem. 1986;261:6478–6485. [PubMed] [Google Scholar]
  2. Appling DJ. Compartmentation of folate-mediated one-carbon metabolism in eukaryotes. FASEB J. 1991;5:2645–2651. doi: 10.1096/fasebj.5.12.1916088. [DOI] [PubMed] [Google Scholar]
  3. Ashworth DJ, Mettler IJ. Direct observation of glycine metabolism in tobacco suspension cells by carbon-13 NMR spectroscopy. Biochemistry. 1984;23:2252–2257. [Google Scholar]
  4. Bradford MM. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem. 1976;72:248–254. doi: 10.1016/0003-2697(76)90527-3. [DOI] [PubMed] [Google Scholar]
  5. Cella R, Balestrazzi A, Carbonera D, Nano R, Rezzani R. Cellular localization of dihydrofolate reductase-thymidylate synthase from Daucus carota. Basic Appl Histochem. 1991;35:61–71. [PubMed] [Google Scholar]
  6. Cossins EA (1987) Folate biochemistry and the metabolism of one-carbon units. In DD Davies, ed, The Biochemistry of Plants, Vol 11. Academic Press, New York, pp 317–353
  7. Flogel U, Wilker W, Leibfritz D. Determination of de novo synthesized amino acids in cellular proteins revisited by C-13 NMR spectroscopy. NMR in Biomedicine. 1997;10:50–58. doi: 10.1002/(sici)1099-1492(199704)10:2<50::aid-nbm450>3.0.co;2-1. [DOI] [PubMed] [Google Scholar]
  8. Gadian DG. Nuclear Magnetic Resonance and its Applications to Living Systems. Oxford, UK: Oxford University Press; 1982. [Google Scholar]
  9. Huangpu J, Pak JH, Graham MC, Rickle SA, Graham JS. Purification and molecular analysis of an extracellular γ-glutamyl hydrolase present in young tissues of the soybean plant. Biochem Biophys Res Commun. 1996;228:1–6. doi: 10.1006/bbrc.1996.1608. [DOI] [PubMed] [Google Scholar]
  10. Kim J-S, Lowe KE, Shane B. Regulation of folate and one-carbon metabolism in mammalian cells. IV. Role of folylpoly-glutamate synthetase in methotrexate metabolism and cytotoxicity. J Biol Chem. 1993;268:21680–21685. [PubMed] [Google Scholar]
  11. Knighton DR, Kan CC, Howland E, Janson CA, Hostomska Z, Welsh KM, Mathews DA. Structure of and kinetic channeling in bifunctional dihydrofolate reductase-thymidylate synthase. Nature Struct Biol. 1994;1:186–194. doi: 10.1038/nsb0394-186. [DOI] [PubMed] [Google Scholar]
  12. Lazar G, Zhang H, Goodman H. The origin of the bifunctional dihydrofolate reductase-thymidylate synthase isogenes of Arabidopsis thaliana. Plant J. 1993;3:657–668. doi: 10.1046/j.1365-313x.1993.03050657.x. [DOI] [PubMed] [Google Scholar]
  13. London RE, Matwiyoff NA, Kollman VH, Mueller DD. Differential nuclear Overhauser enhancement for 13C singlet and 13C-13C multiplet resonances of a carboxyl carbon. J Magn Reson. 1975;18:555–557. [Google Scholar]
  14. Luo M, Orsi R, Patrucco E, Pancaldi S, Cella R. Multiple transcription start sites of the carrot dihydrofolate reductase-thymidylate synthase gene, and sub-cellular localization of the bifunctional protein. Plant Mol Biol. 1997;33:709–722. doi: 10.1023/a:1005798207693. [DOI] [PubMed] [Google Scholar]
  15. Luo M, Piffanelli P, Rastelli L, Cella R. Molecular cloning and analysis of a cDNA coding for the bifunctional dihydrofolate reductase-thymidylate synthase of Daucus carota. Plant Mol Biol. 1993;22:427–435. doi: 10.1007/BF00015973. [DOI] [PubMed] [Google Scholar]
  16. McGuire JJ, Coward JK. Pteroylpolyglutamates: biosynthesis, degradation and function. In: Blakely RL, Benkovic SJ, editors. Folates and Pterins, Vol 1. New York: John Wiley & Sons; 1984. pp. 135–189. [Google Scholar]
  17. Misra DK, Humphreys SR, Friedkin M, Goldin A, Crawford EJ. Increased dihydrofolate reductase activity as a possible basis of drug resistance in leukaemia. Nature. 1961;189:39–42. doi: 10.1038/189039a0. [DOI] [PubMed] [Google Scholar]
  18. Narkewicz MR, Sauls SD, Tjoa SS, Teng C, Fennessey PV. Evidence for intracellular partitioning of serine and glycine metabolism in Chinese hamster ovary cells. Biochem J. 1996;313:991–996. doi: 10.1042/bj3130991. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Neeman M, Aviv D, Degani H, Galun E. Glucose and glycine metabolism in regenerating tobacco protoplasts. Plant Physiol. 1985;77:374–378. doi: 10.1104/pp.77.2.374. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Neuburger M, Rebeille F, Jourdain A, Nakamura S, Douce R. Mitochondria are a major site for folate and thymidylate synthesis in plants. J Biol Chem. 1996;271:9466–9472. doi: 10.1074/jbc.271.16.9466. [DOI] [PubMed] [Google Scholar]
  21. Nour JM, Rabinowitz JC. Isolation, characterization, and structural organization of 10-formyltetrahydrofolate synthetase from spinach leaves. J Biol Chem. 1991;266:18363–18369. [PubMed] [Google Scholar]
  22. Oliver DJ. The glycine decarboxylase complex from plant mitochondria. Annu Rev Plant Physiol Plant Mol Biol. 1994;45:323–337. [Google Scholar]
  23. Pasternack LB, Laude DA, Jr, Appling DR. Whole-cell detection by 13C NMR of metabolic flux through the C1-tetrahydrofolate synthase/serine hydroxymethyl transferase enzyme system and effect of antifolate exposure in Saccharomyces cerevisiae. Biochemistry. 1994;33:7166–7173. doi: 10.1021/bi00189a020. [DOI] [PubMed] [Google Scholar]
  24. Prabhu V, Chatson KB, Abrams GD, King J. 13C nuclear magnetic resonance detection of interactions of serine hydroxymethyltransferase with C1-tetrahydrofolate synthase and glycine decarboxylase complex activities in Arabidopsis. Plant Physiol. 1996a;112:207–216. doi: 10.1104/pp.112.1.207. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Prabhu V, Chatson KB, Abrams GD, King J. 13C chemical shifts of 20 free amino acids and their use in detection by NMR of free amino acids in intact plants of Arabidopsis. J Plant Physiol. 1996b;149:246–250. [Google Scholar]
  26. Prabhu V, Lui H, King J. Arabidopsis dihydropteroate synthase: general properties and inhibition by reaction product and sulfonamides. Phytochemistry. 1997;45:23–27. doi: 10.1016/s0031-9422(96)00793-5. [DOI] [PubMed] [Google Scholar]
  27. Pratt WB, Taylor P (1990) Principles of Drug Action. Churchill Livingstone, New York
  28. Rebeille F, Macherel D, Moillon J-M, Garin J, Douce R. Folate biosynthesis in higher plants: purification and molecular cloning of a bifunctional 6-hydroxymethyl-7,8-dihydropterin pyrophosphokinase/7,8-dihydropteroate synthase localized in mitochondria. EMBO J. 1997;16:947–957. doi: 10.1093/emboj/16.5.947. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Schirch L. Folates in glycine and serine metabolism. In: Blakely RL, Benkovic SJ, editors. Folates and Pterins, Vol 1. New York: John Wiley & Sons; 1984. pp. 399–431. [Google Scholar]
  30. Schweitzer BI, Dicker AP, Bertino JR. Dihydrofolate reductase as a therapeutic target. FASEB J. 1990;4:2441–2452. doi: 10.1096/fasebj.4.8.2185970. [DOI] [PubMed] [Google Scholar]
  31. Stover P, Schirch V. The metabolic role of leucovorin. Trends Biochem Sci. 1993;18:102–106. doi: 10.1016/0968-0004(93)90162-g. [DOI] [PubMed] [Google Scholar]
  32. Suzuki KT, Cary LW, Kuhlmann KF. Carbon-13 relaxation in doubly labeled sodium acetate. J Magn Reson. 1975;18:390–392. [Google Scholar]
  33. Wang M, Ratnam S, Freisham JH. Cloning, nucleotide sequencing and expression of the bifunctional dihydrofolate reductase-thymidylate synthase from Glycine max. Biochim Biophys Acta. 1996;1261:325–336. doi: 10.1016/0167-4781(94)00251-w. [DOI] [PubMed] [Google Scholar]
  34. Wu K, Atkinson IJ, Cossins EA, King J. Methotrexate resistance in Datura innoxia. Uptake and metabolism of methotrexate in wild-type and resistant cell lines. Plant Physiol. 1993;101:477–483. doi: 10.1104/pp.101.2.477. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Wu K, Cossins EA, King J. Folate metabolism in Datura innoxia. In vivo and in vitro folylpolyglutamate synthesis in wild-type and methotrexate-resistant cells. Plant Physiol. 1994;104:373–380. doi: 10.1104/pp.104.2.373. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Plant Physiology are provided here courtesy of Oxford University Press

RESOURCES