Biosynthesis of nucleic acid purines in Mycobacterium tuberculosis H37Rv

1. The biosynthesis of nucleic acid purine in Mycobacterium tuberculosis H(37)R(v) has been studied by using (14)C-labelled precursors. 2. The results indicate that C-2 and C-8 of the purine ring are derived most efficiently from serine and glycine and not from formate. 3. [(14)C]Methionine is not incorporated into the ureide carbon atoms of the purine ring.     

Leishmania mexicana: Purine metabolism in promastigotes,axenic amastigotes,and amastigotes derived from vero cells

Leishmania mexicana mexicana promastigotes, axenic amastigotes, and amastigotes derived from Vero cells were examined for de novo purine synthesis and mechanisms of purine salvage. Both promastigotes and axenic amastigotes were incapable of de novo purine synthesis, as shown by the lack of [¹⁴C]formate and [¹⁴C]glycine incorporation into purine nucleotide pools. However, the ready incorporation of [¹⁴C]hypoxanthine, [¹⁴C]adenine, and [¹⁴C]guanine suggested that purine salvage pathways were operating. In addition, a significant percentage (?60%) of the total label from these purine precursors was associated with adenylate nucleotides. Nucleotide pool levels of axenic amastigotes were consistently greater but the specific activities were less than those of promastigotes, suggesting a slower rate of purine metabolism in the axenic amastigote form. Similar results were obtained from amastigotes isolated from infected Vero cells.     

The effects on 4-aminoantifolates on 5-formyltetrahydrofolate metabolism in L1210 cells. A biochemical basis of the selectivity of leucovorin rescue

This report describes studies designed to evaluate possible inhibitory effects of diaminoantifolates on folate-dependent biosynthetic enzymes in intact L1210 leukemia cells. A novel approach is described which involves an assessment of the metabolism of and biosynthetic flux of the one-carbon moiety from (6S)5-formyltetrahydrofolate in folate-depleted cells. Pretreatment with methotrexate (10 microM), resulting in the formation of methotrexate polyglutamates, or continuous incubation with trimetrexate (1 microM) inhibited growth of folate-depleted L1210 cells in the presence of folic acid or 5-formyltetrahydrolate. In both control and drug-treated cells, double-labeled (6S)-5-[14C]formyl[3H]tetrahydrofolate was rapidly metabolized with the loss of the [14C]formyl group. Under all conditions, the predominant metabolite was 10-formyl[3H]tetrahydrofolate, detectable both intracellularly and extracellularly. In drug-treated cells, there was a remarkably small decrease in the level of 10-formyl[3H]tetrahydrofolate (approximately 30%) and a 10-fold rise in the level of [3H]dihydrofolate to less than 20% of the total folate pool. The incorporation of [14C]formyl group from 5-[14C]formyltetrahydrofolate into thymidylate, serine, and methionine was unaffected by the presence of 1 microM trimetrexate, consistent with the generation of sufficient 5,10-[14C]methylenetetrahydrofolate to drive these reactions. Similarly, the presence of methotrexate polyglutamates had no effect at the level of amino acid synthesis; however, carbon transfer into thymidylate was markedly inhibited. Even though 10-formyltetrahydrofolate was readily formed from 5-formyltetrahydrofolate in this model, the net incorporation of 14C from 5-[14C]formyltetrahydrofolate into purine nucleotides was inhibited by both methotrexate and trimetrexate treatments. Similar findings were obtained when [14C]glycine incorporation into purine nucleotides was monitored in cells incubated with unlabeled 5-formyltetrahydrofolate. Finally, in antifolate-treated cells incubated with unlabeled 5-formyl-tetrahydrofolate, transfer of 14C from [14C]formate or [14C]serine into biosynthetic products or incorporation of [3H]deoxyuridine into nucleic acids was potently inhibited. These results suggest that insufficient levels of tetrahydrofolate and 5, 10-methylenetetrahydrofolate were formed to drive these reactions despite the presence of high levels of 10-formyltetrahydrofolate.(ABSTRACT TRUNCATED AT 400 WORDS)     

Regulation of Purine Metabolism in Intact Leaves of Coffea arabica

The capacity of Coffea arabica leaves (5- x 5-mm pieces) to synthesize de novo and catabolize purine nucleotides to provide precursors for caffeine (1,3,7-trimethylxanthine) was investigated. Consistent with de novo synthesis, glycine, bicarbonate, and formate were incorporated into the purine ring of inosine 5[prime]-monophosphate (IMP) and adenine nucleotides ([sigma]Ade); azaserine, a known inhibitor of purine de novo synthesis, inhibited incorporation. Activity of the de novo pathway in C. arabica per g fresh weight of leaf tissue during a 3-h incubation period was 8 [plus or minus] 4 nmol of formate incorporated into IMP, 61 [plus or minus] 7 nmol into [sigma]Ade, and 150 nmol into caffeine (the latter during a 7-h incubation). Coffee leaves exhibited classical purine catabolism. Radiolabeled formate, inosine, adenosine, and adenine were incorporated into hypoxanthine and xanthine, which were catabolized to allantoin and urea. Urease activity was demonstrated. Per g fresh weight, coffee leaf squares incorporated 90 [plus or minus] 22 nmol of xanthine into caffeine in 7 h but degraded 102 [plus or minus] 1 nmol of xanthine to allantoin in 3 h. Feedback control of de novo purine biosynthesis was contrasted in C. arabica and Cucurbita pepo, a species that does not synthesize purine alkaloids. End-product inhibition was demonstrated to occur in both species but at different enzyme reactions.     

Biosynthesis and scavenging of purines by pathogenic mycobacteria including Mycobacterium leprae

Purine biosynthesis de novo could not be detected in suspensions of Mycobacterium leprae isolated from armadillo tissue. In contrast, non-growing suspensions of other pathogenic mycobacteria, also isolated from infected host tissue did synthesize purines. Rates of synthesis, judged by incorporation of [2-14C]glycine or [3-14C]serine into nucleic acid purines were 600 times higher in M. microti and 110 times higher in M. avium--both isolated from infected mouse tissue--than the lowest possible rate detectable and therefore the highest possible rate in M. leprae. The rate of purine synthesis relative to purine scavenging (judged by comparing incorporation of [3-14C]serine and [8-14C]hypoxanthine into nucleic acid purines in suspensions of mycobacteria) varied only slightly--4-fold in M. microti and 6-fold in M. avium--whether organisms were harvested from media with or without purines, from media with a low nitrogen content but containing a purine, from mice or even with starved organisms. Thus, the failure of M. leprae to synthesize purines could not be explained as either a result of using non-growing mycobacteria in the incubations with 14C-labelled precursors or as repression or inhibition of synthesis de novo. It appears that M. leprae requires a supply of the purine ring from its environment. Nucleotides, which may be the major source of the purine ring in the intracellular environment, were not taken up directly by M. leprae but could be hydrolysed first to nucleosides and then taken up.     

Purine Nucleotide Synthesis in Adrenal Chromaffin Cells

Abstract: The synthesis of purine nucleotides from the salvage precursors adenine and adenosine, and from the de novo precursors formate and glycine, was studied in isolated adrenal chromaffin cells. Both [8-¹⁴C]adenine and [8-¹⁴C]adenosine from extracellular medium are effectively incorporated into intracellular nucleotides. [¹⁴C]Formate and [U-¹⁴C]glycine are also incorporated, but de novo synthesis is clearly lower than synthesis from salvage precursors, although similar to de novo synthesis in liver. The enzymes responsible for adenine and adenosine salvage, adenine phosphoribosyltransferase and adenosine kinase, were purified about 1,500-fold. Both enzymes are mainly cytosolic and exhibit a similar molecular weight of around 42,000. The results suggest that chromaffin cells can replenish their intracellular nucleotides lost during the secretory event by an active synthesis from salvage and de novo precursors.     

The interconversion of glycine and serine by plant tissue extracts

1. Extracts prepared from a variety of higher-plant tissues by ammonium sulphate fractionation were shown to catalyse the interconversion of glycine and serine. This interconversion had an absolute requirement for tetrahydrofolate and appeared to favour serine formation. 2. The biosynthesis of serine from glycine was studied in more detail with protein fractionated from 15-day-old wheat leaves. Synthesis of [¹⁴C]serine from [¹⁴C]glycine was not accompanied by labelling of glyoxylate, glycollate or formate. 3. The synthesis of serine from glycine was stimulated by additions of formaldehyde, and [¹⁴C]formaldehyde was readily incorporated into C-3 of serine in the presence of tetrahydrofolate. 4. The results are interpreted as indicating that serine biosynthesis involves a direct cleavage of glycine whereby the α-carbon is transferred via N⁵N¹⁰-methylenetetrahydrofolate to become the β-carbon of serine.     

Methylthioadenosine serves as a single carbon source to the folate co-enzyme pool in rat bone marrow cells

[ribose-U-14C]Methylthioadenosine (MTA) was prepared by incubating methionine with [14C-U]ATP in the presence of methionine adenosyltransferase and the resulting S-adenosylmethionine was heated to release MTA. Labelled [14C]MTA, when incubated with rat bone marrow cells, yielded [14C]formate which was used in the synthesis of adenine and guanine. Unlike 14C from sodium, formate, serine and glycine, there was no decline in 14C utilization from MTA with bone marrow cells from rats in which cobalamin had been inactivated by exposure to nitrous oxide. It was concluded that methionine via MTA is a significant contributor of single-carbon units at the formate level of oxidation and that this pathway is maintained in cobalamin 'deficiency'.     

Metabolism of xanthine and hypoxanthine in the tea plant (Thea sinensis L.).

1. The metabolism of xanthine and hypoxanthine in excised shoot tips of tea was studied with micromolar amounts of [2(-14)C]xanthine or [8(-14)C]hypoxanthine. Almost all of the radioactive compounds supplied were utilized by tea shoot tips by 30 h after their uptake. 2. The main products of [2(-14)C]xanthine and [8(-14)C]hypoxanthine metabolism in tea shoots were urea, allantoin and allantoic acid. There was also incorporation of the label into theobromine, caffeine and RNA purine nucleotides. 3. The results indicate that tea plants can catabolize purine bases by the same pathways as animals. It is also suggested that tea plants have the ability to snythesize purine nucleotides from glycine by the pathways of purine biosynthesis de novo and from hypoxanthine and xanthine by the pathway of purine salvage. 4. The results of incorporation of more radioactivity from [8(-14)C]hypoxanthine than from [2(-14)C]xanthine into RNA purine nucleotides and caffeine suggest that hypoxanthine is a more effective precursor of caffeine biosynthesis than xanthine. The formation of caffeine from hypoxanthine is a result of nucleotide synthesis via the pathway of purine salvage.     

The specific radioactivity of the tissue free amino acid pool as a basis for measuring the rate of protein synthesis in the rat in vivo

1. Rats were infused in vivo with [U-(14)C]glycine for periods of 2-6h, during which time the specific radioactivity of the free glycine in plasma and tissue approached a constant value. 2. Free serine also became labelled. The ratio of specific radioactivity of serine to that of glycine in the protein of liver, kidney, brain, jejunum, heart, diaphragm and gastrocnemius muscle was closer to the ratio in the free amino acid pool of the tissue than that of the plasma. 3. The kinetics of incorporation of [(14)C]glycine and [(14)C]serine into the protein of gastrocnemius muscle further suggested that the plasma free amino acids were not the immediate precursors of protein. 4. Infusion of rats with [U-(14)C]serine resulted in labelling of free glycine. The ratio of specific radioactivity of glycine to serine in the protein of liver, kidney, brain, jejunum and heart again suggested incorporation from a pool similar to the free amino acid pool of the tissue. 5. Rates of tissue protein synthesis calculated from the incorporation into protein of both radioactive glycine and serine, either infused or derived, were very similar when the precursor specific radioactivity was taken to be that in the total free amino acids of the tissue. Except for gastrocnemius muscle and diaphragm during the infusion of radioactive serine, the rates of tissue protein synthesis calculated from the specific radioactivity of the free glycine and serine in plasma differed markedly.     

Purine and phosphoribosylpyrophosphate synthesis in differentiating murine virus-induced erythroleukemic cells in vitro.

Phosphoribosylpyrophosphate synthetase activity was determined in Friend virus-inducted erythroleukemic cells in culture, stimulated to differentiate in the presence of dimethylsulfoxide. The activity of phosphoribosylpyrophosphate synthetase did not decrease in cells which had acquired the specialized function of hemoglobin synthesis, nor was the phosphoribosylpyrophosphate content of untreated erythroleukemic cells significantly different from that of cultures exposed to dimethylsulfoxide for 96 hours. However, the rate of the early steps of de novo purine biosynthesis as measured by the incorporation of [1-14C] glycine and [1-14C] formate into formyglycinamide ribonucleotide, was significantly lower in differentiating cell cultures. The addition of glutamine or ammonia increased glycine incorporation of control cultures, but failed to do so in treated cultures. In the course of the normal development of erythrocytes in vivo, phosphoribosylpyrophosphate synthetase activity is preserved, while the capacity to synthesize purines de novo is lost, as is the activity of the phosphoribosyl-l-amine synthesizing enzymes. Our present study suggests that the rate of de novo purine biosynthesis in this erythroleukemic cell line is not limited by the availability of phosphoribosylphrophosphate, but rather by a decrease in the phosphoribosyl-l-amine synthesizing enzymes. These findings provide further evidence that during dimethylsulfoxide-stimulated erythroid maturation, the same regulatory mechanisms are operative as in normal cellular development, and that ammonia-dependent purine biosynthesis is subject to the same regulatory mechanisms as is glutamine-dependent biosynthesis.     

ACTH stimulation of purine nucleotide biosynthesis in the adrenal cortex. ACTH stimulation of purine biosynthesis

In the presence of azaserine an inhibitor of phosphoribosylformylglycineamidine synthetase (EC 6.3.5.3) the incorporation of [2-¹⁴C]glycine into 5′-phosphoribosylglycineamide and its formyl derivative was measured in 105,000g supernatant fraction prepared from a homogenate of adrenal cortex. Corticotropin at a level of 1-0.001 nm markedly stimulated in 10 min these early steps of purine biosynthesis. The stimulus was in addition to that achieved with added glucose-6-phosphate and NADP. Increased synthesis of precursors of purine nucleotides is due to ACTH activation of adrenal glucose-6-phosphate dehydrogenase (EC 1.1.1.49) and thus the pentose cycle with an increase in 5′-phosphoribosylpyrophosphate. The generation of this latter compound is presumed to be a rate-limiting factor to 5′-phosphoribosylpyrophosphate amidotransferase (EC 2.4.2.14) the first enzyme of de novo purine biosynthesis.     

Biosynthesis of thiamine. Incorporation experiments with 14C-labelled substrates and with [15N]glycine in Saccharomyces cerevisiae

1. Yeast was grown in a minimal synthetic medium together with a range of (14)C-labelled substrates under standardized conditions. After isolation, the purified thiamine was cleaved by sulphite and the pyrimidine and thiazole moieties were purified and assayed for radioactivity. 2. In order of decreasing incorporation, [(14)C]formate, [3-(14)C]serine, [2-(14)C]glycine and [2-(14)C]acetate supplied label for the pyrimidine, and [2-(14)C]glycine, [3-(14)C]serine, [1-(14)C]glycine, [(14)C]formate and [2-(14)C]acetate for the thiazole. Incorporation of label into the fragments from several other (14)C-labelled substrates, including [Me-(14)C]- and [3,4-(14)C(2)]-methionine, was insignificant. 3. [3-(14)C]Serine was shown not to contribute label to C-2 of the thiazole ring. 4. Significant incorporation of nitrogen from [(15)N]glycine into the thiazole moiety, but not into the pyrimidine moiety, was established. 5. It appears that C-2 and N-3 of the thiazole ring are formed from C-2 and the nitrogen atom of glycine, but the entire methionine molecule does not appear to be implicated.     

13C NMR detection of folate-mediated serine and glycine synthesis in vivo in Saccharomyces cerevisiae.

Saccharomyces cerevisiae has both cytoplasmic and mitochondrial C1-tetrahydrofolate (THF) synthases. These trifunctional isozymes are central to single-carbon metabolism and are responsible for interconversion of the THF derivatives in the respective compartments. In the present work, we have used 13C NMR to study folate-mediated single-carbon metabolism in these two compartments, using glycine and serine synthesis as metabolic endpoints. The availability of yeast strains carrying deletions of cytoplasmic and/or mitochondrial C1-THF synthase allows a dissection of the role each compartment plays in this metabolism. When yeast are incubated with [13C]formate, 13C NMR spectra establish that production of [3-13C]serine is dependent on C1-THF synthase and occurs primarily in the cytosol. However, in a strain lacking cytoplasmic C1-THF synthase but possessing the mitochondrial isozyme, [13C]formate can be metabolized to [2-13C]glycine and [3-13C]serine. This provides in vivo evidence for the mitochondrial assimilation of formate, activation and conversion to [13C]CH2-THF via mitochondrial C1-THF synthase, and subsequent glycine synthesis via reversal of the glycine cleavage system. Additional supporting evidence of reversibility of GCV in vivo is the production of [2-13C]glycine and [2,3-13C]serine in yeast strains grown with [3-13C]serine. This metabolism is independent of C1-THF synthase since these products were observed in strains lacking both the cytoplasmic and mitochondrial isozymes. These results suggest that when formate is the one-carbon donor, assimilation is primarily cytoplasmic, whereas when serine serves as one-carbon donor, considerable metabolism occurs via mitochondrial pathways.     

Characterization of purine nucleotide metabolism in primary rat cardiomyocyte cultures

Primary rat cardiomyocyte cultures were utilized as a model for the study of purine nucleotide metabolism in the heart muscle, especially in connection with the mechanisms operating for the conservation of adenine nucleotides. The cultures exhibited capacity to produce purine nucleotides from nonpurine molecules (de novo synthesis), as well as from preformed purines (salvage synthesis). The conversion of adenosine to AMP, catalyzed by adenosine kinase, appears to be the most important physiological salvage pathway of adenine nucleotide synthesis in the cardiomyocytes. The study of the metabolic fate of IMP formed from [14C]formate or [14C]hypoxanthine and that of AMP formed from [14C]adenine or [14C]adenosine revealed that in the cardiomyocyte the main flow in the nucleotide interconversion pathways is from IMP to AMP, whereas the flux from AMP to IMP appeared to be markedly slower. Following synthesis from labeled precursors by either de novo or salvage pathways, most of the radioactivity in purine nucleotides accumulated in adenine nucleotides, and only a small proportion of it resided in IMP. The results suggest that the main pathway of AMP degradation in the cardiomyocyte proceeds through adenosine rather than through IMP. About 90% of the total radioactivity in purines effluxed from the cells during de novo synthesis from [14C]formate or following prelabeling of adenine nucleotides with [14C]adenine were found to reside in hypoxanthine. The activities in cell extracts of AMP 5'-nucleotidase and IMP 5'-nucleotidase, which catalyze nucleotide degradation, and of AMP deaminase, a key enzyme in the purine nucleotide cycle, were low. The nucleotidase activity resembles, and that of the AMP deaminase contrasts the respective enzyme activities in extracts of cultured skeletal-muscle myotubes. The results indicate that in the cardiomyocyte, in contrast to the myotube, the main mechanism operating for conservation of nucleotides is prompt phosphorylation of AMP, rather than operation of the purine nucleotide cycle. The primary cardiomyocyte cultures are a plausible model for the study of purine nucleotide metabolism in the heart muscle.     

Effects of gibberellic acid and L-methionine on C1 metabolism in aerated carrot disk

Aeration of carrot storage tissue disks in water was accompanied by net folate synthesis and by changes in the specific activities of key folate-dependent enzymes. Disks aerated in 0.1 mM gibberellic acid (GA₃) for 48 hr contained higher concentrations of methyltetrahydrofolates but aeration in 5 mM L-methionine reduced net folate synthesis. Gibberellic acid also increased the specific activities of 5,10-methylenetetrahydrofolate reductase (E.C. 1.1.1.68), serine hydroxymethyltransferase (E.C. 2.1.2.1) and 5-methyltetrahydrofolate: homocysteine transmethylase. The levels of these enzymes in disks aerated in L-methionine (5 mM) were comparable or slightly higher than those of disks aerated in water. Activity of the reductase and 10-formyltetrahydrofolate synthetase (E.C. 6.3.4.3) was inhibited by L-methionine in vitro. Aeration increased ability to incorporate formate [¹⁴C] into serine, glycine and methionine. Disks aerated for 36 hr in 0.1 mM GA₃ incorporated greater amounts of ¹⁴C into free methionine but those aerated in L-methionine (5 mM) had less ability to metabolize formate and the specific radioactivities of free glycine, serine and methionine were low.     

Serine utilization as a precursor of phosphatidylserine and alkenyl-(plasmenyl)-, alkyl-, and acylethanolamine phosphoglycerides in cultured glioma cells

In several tissues and cell lines, serine utilized for phosphatidylserine (PS) synthesis is an eventual precursor of the base moiety of ethanolamine phosphoglycerides (PE). We investigated the biosynthesis and decarboxylation of PS in cultured C6 glioma cells, with particular attention to 1-O-alk-1'-enyl-2-acyl-sn-glycero-3-phosphoethanolamine (plasmenylethanolamine) biosynthesis. Incorporation of [3H]serine into PS reached a maximum within 4-8 h, and label in nonplasmenylethanolamine phosphoglyceride (NP-PE) and plasmenylethanolamine was maximal by 12-24 h and 48 h, respectively. After 8 h, label in PS decreased even though 40-60% of initial label remained in the culture medium. Serial additions of fresh [3H]serine restored PS synthesis to higher levels of labeled PS accumulation followed by a subsequent decrease in 4-8 h. High performance liquid chromatographic analyses confirmed that medium serine was depleted by 8 h, and thereafter metabolites, including acetate and formate, accounted for radioactivity in the medium. The rapid but transient appearance of labeled glycine and ATP inside the cells indicated conversion of serine by hydroxymethyltransferase. 78-85% of label from serine was in headgroup of PS or of PE formed by decarboxylation. A precursor-product relationship was suggested for label from [3H]serine appearing in the headgroup of diacyl, alkylacyl, and alkenylacyl subclasses of PE. By 48 h, a constant specific activity, ratio of approximately 1:1 was reached between plasmenylethanolamine and NP-PE, similar to the molar distribution of these lipids. In contrast, equilibrium was not achieved in cells incubated with [1,2-14C]ethanolamine; plasmenylethanolamine had 2-fold greater specific activity than labeled NP-PE by 72-96 h. These observations indicate that in cultured glioma cells 1) serine serves as a precursor of the head group of PS and of both plasmenyl and non-plasmenyl species of PE; 2) exchange of headgroup between NP-PE and plasmenylethanolamine may involve different donor pools of PE depending on whether the headgroup originates with exogenous serine or ethanolamine; 3) serine is rapidly converted to other metabolites, which limits exogenous serine as a direct phospholipid precursor.     

Growth of Paracoccus denitrificans on [2,3-13C]succinate and [1,4-13C]succinate. III. Biosynthetic pathways

The biosynthesis in vivo of a number of amino acids, sugars, and purines in Paracoccus denitrificans grown on either [2,3-13C]succinate or [1,4-13C]succinate was investigated by using gas chromatography-mass spectrometry. The distribution of label in the TCA-cycle-related amino acids indicated that carbon intermediates of energy metabolism were utilized as precursors for the biosynthesis of these amino acids in vivo. The biosynthesis of glycine, serine, phenylalanine and glycerol from labelled succinate in vivo were consistent with phosphoenol pyruvate as an intermediate. A mechanism for the formation of C4, C5 and C6 sugars without the use of fructose-1,6-bisphosphate aldolase (which has not been detected in P. denitrificans) is proposed. The 13C-enrichments of ribose in the bacterium indicate that there are at least three routes of ribose biosynthesis operating during growth on labelled succinate. The probability distribution of labelled purine molecules was successfully predicted for adenine, guanine and adenosine, thus confirming their generally accepted route of biosynthesis in vivo.     

Purine metabolism in Leishmania donovani and Leishmania braziliensis

We have studied purine metabolism in the culture forms of Leishmania donovani and Leishmania braziliensis. These organisms are incapable of synthesizing purines de novo from glycine, serine, or formate and require an exogenous purine for growth. This requirement is better satisfied by adenosine or hypoxanthine than by guanosine. Bothe adenine and inosine are converted to a common intermediate, hypoxanthine, before transformation to nucleotides. This is due to the activity of an adenine aminohydrolase (EC 3.5.4.2), a rather unusual finding in a eukaryotic cell. There is a preferential synthesis of adenine nucleotides, even when guanine or xanthine are used as precursors.The pathways of purine nucleotide interconversions in these Leishmania resemble those found in mammalian cells except for the absence of de novo purine biosynthesis and the presence of an adenine-deaminating activity.     

Investigating the regulation of one-carbon metabolism in Arabidopsis thaliana

Serine (Ser) biosynthesis in C(3) plants can occur via several pathways. One major route involves the tetrahydrofolate (THF)-dependent activities of the glycine decarboxylase complex (GDC, EC 2.1.1.10) and serine hydroxymethyltransferase (SHMT, EC 2.1.2.1) with glycine (Gly) as one-carbon (1-C) source. An alternative THF-dependent pathway involves the C1-THF synthase/SHMT activities with formate as 1-C source. Here, we have investigated aspects of the regulation of these two folate-mediated pathways in Arabidopsis thaliana (L.) Heynh. Columbia using two approaches. Firstly, transgenic plants overexpressing formate dehydrogenase (FDH, EC 1.2.1.2) were used to continue our previous studies on the function of FDH in formate metabolism. The formate pool size was approximately 73 nmol (g FW)(-1) in wild type (WT) Arabidopsis plants; three independent transgenic lines had similar-sized pools of formate. Transgenic plants produced more (13)CO(2) from supplied [(13)C]formate than did WT plants but were not significantly different from WT plants in their synthesis of Ser. We concluded that FDH has no direct role in the regulation of the above two pathways of Ser synthesis; the breakdown of formate to CO(2) by the FDH reaction is the primary and preferred fate of the organic acid in Arabidopsis. The ratio between the GDC/SHMT and C1-THF synthase/SHMT pathways of Ser synthesis from [alpha-(13)C]Gly and [(13)C]formate, respectively, in Arabidopsis shoots was 21 : 1; in roots, 9 : 1. In shoots, therefore, the pathway from formate plays only a small role in Ser synthesis; in the case of roots, results indicated that the 9 : 1 ratio was as a result of greater fluxes of (13)C through both pathways together with a relatively higher contribution from the C1-THF synthase/SHMT route than in shoots. We also examined the synthesis of Ser in a GDC-deficient mutant of Arabidopsis (glyD) where the GDC/SHMT pathway was impaired. Compared with WT, glyD plants accumulated 5-fold more Gly than WT after supplying [alpha-(13)C]Gly for 24 h; the accumulation of Ser from [alpha-(13)C]Gly was reduced by 25% in the same time period. On the other hand, the accumulation of Ser through the C1-THF synthase/SHMT pathway in glyD plants was 2.5-fold greater than that in WT plants. Our experiments confirmed that the GDC/SHMT and C1-THF synthase/SHMT pathways normally operate independently in Arabidopsis plants but that when the primary GDC/SHMT pathway is impaired the alternative C1-THF synthase/SHMT pathway can partially compensate for deficiencies in the synthesis of Ser.