Isolation and sequencing of the cDNA coding for spinach 10-formyltetrahydrofolate synthetase. Comparisons with the yeast, mammalian, and bacterial proteins

ONE-CARBON METABOLISM IN HIGHER PLANTS

The metabolism of one-carbon (C1) units is essential to plants, and plant C1 metabolism has novel features not found in other organisms-plus some enigmas. Despite its centrality, uniqueness, and mystery, plant C1 biochemistry has historically been quite poorly explored, in part because its enzymes and intermediates tend to be labile and low in abundance. Fortunately, the integration of molecular and genetic approaches with biochemical ones is now driving rapid advances in knowledge of plant C1 enzymes and genes. An overview of these advances is presented. There has also been progress in measuring C1 metabolite fluxes and pool sizes, although this remains challenging and there are relatively few data. In the future, combining reverse genetics with flux and pool size determinations should lead to quantitative understanding of how plant C1 pathways function. This is a prerequisite for their rational engineering.

Folates and one-carbon metabolism in plants and fungi

Folate-dependent pathways of one-carbon metabolism are essential for the synthesis of purines, formylmethionyl-tRNA, thymidylate, serine and methionine. These syntheses use a cellular source of one-carbon substituted, tetrahydrofolate polyglutamate derivatives which are the preferred substrates of most folate-dependent enzymes. In the last decade, there have been major advances in the folate biochemistry of animal, bacterial, fungal and plant systems. These have included the refinement of methods for folate isolation and characterization, basic work on key enzymes of folate biosynthesis and the detailed characterization of proteins that catalyze the generation and utilization of one-carbon substituted folates.

Genetic and physiologic analysis of a formyl-tetrahydrofolate synthetase mutant of Streptococcus mutans

Previously we reported that transposon Tn917 mutagenesis of Streptococcus mutans JH1005 yielded an isolate detective in its normal ability to produce a mutacin (P. J. Crowley, J. D. Hillman, and A. S. Bleiweis, abstr. D55, p. 258 in Abstracts of the 95th General Meeting of the American Society for Microbiology 1995, 1995). In this report we describe the recovery of the mutated gene by shotgun cloning. Sequence analysis of insert DNA adjacent to Tn917 revealed homology to the gene encoding formyl-tetrahydrofolate synthetase (Fhs) from both prokaryotic and eukaryotic sources. In many bacteria, Fhs catalyzes the formation of 10-formyl-tetrahydrofolate, which is used directly in purine biosynthesis and formylation of Met-tRNA and indirectly in the biosynthesis of methionine, serine, glycine, and thymine. Analysis of the fhs mutant grown anaerobically in a minimal medium demonstrated that the mutant had an absolute dependency only for adenine, although addition of methionine was necessary for normal growth. Coincidently it was discovered that the mutant was sensitive to acidic pH; it grew more slowly than the parent strain on complex medium at pH 5. Complementation of the mutant with an integration vector harboring a copy of fhs restored its ability to grow in minimal medium and at acidic pH as well as to produce mutacin. This represents the first characterization of Fhs in Streptococcus.

Site-directed mutagenesis of putative catalytic and nucleotide binding sites in N10-formyltetrahydrofolate synthetase

To determine the importance of specific amino-acid residues in catalysis and substrate binding by N10-formylH4 folate synthetase, one lysine and three histidine residues in the enzyme from Clostridium cylindrosporum were mutated to glutamine and serine residues, respectively. These residues, Lys-71, His-125, His-131, and His-268, are conserved in four bacterial and five eukaryotic proteins for which the amino-acid sequences are known. Previous evidence indicated that a histidine residue may play a role in catalysis and it has been proposed that Lys-71 could be a member of a putative nucleotide binding consenus sequence. The histidine mutations, H125S, H131S, and H268S, produced proteins that were unstable and were proteolytically degraded to different extents. No activity of purified H268S could be detected and the 240 kDa native tetramer was also absent. Activities of the H125S and H131S mutants could be measured and the Km values of the substrates were similar to those for the wild-type enzyme. It is concluded that the mutations resulted in monomers that do not fold properly and/or do not associate to the active tetramer and, as a consequence, are susceptible to intracellular proteolytic digestion. On the other hand, the K71Q mutation did not produce proteolyzed material. The resulting protein had a kcat value which was reduced by a factor of 3.3 x 10(-4). Km values of the substrates were not affected, nor were the affinty constants for MgATP and H4PteG3. CD and fluorescence spectra demonstrated that little change in the tertiary structure of the protein had occurred as a result of the mutation. The monomer form of K71Q was less stable than the monomer of the wild-type enzyme and reassociated less efficiently than the wild-type. From these results it is suggested that Lys-71 plays a critical role in catalysis by N10-formylH4 folate synthetase and that this residue may reside at an intersubunit interface.

The two monofunctional domains of octameric formiminotransferase-cyclodeaminase exist as dimers

Formiminotransferase-cyclodeaminase is a bifunctional enzyme arranged as a circular tetramer of dimers that exhibits the ability to efficiently channel polyglutamylated folate between catalytic sites. Through deletion mutagenesis we demonstrate that each subunit consists of an N-terminal transferase active domain and a C-terminal deaminase active domain separated by a linker sequence of minimally eight residues. The full-length enzyme and both isolated domains have been expressed as C-terminally histidine-tagged proteins. Both domains self-dimerize, providing direct evidence for the existence of two types of subunit interfaces. The results suggest that both the transferase and the deaminase activities are dependent on the formation of specific subunit interfaces. Because channeling is not observed between isolated domains, only the octamer appears able to directly transfer pentaglutamylated intermediate between active sites.

Sequence and expression of the gene for N10-formyltetrahydrofolate synthetase from Clostridium cylindrosporum

Sau3 A and Hind III restriction fragments of Clostridium cylindrosporum genomic DNA were used to isolate clones containing 80% of the N10-H4folate synthetase gene in a 5' fragment and the remaining 20% of the gene in the 3' fragment. These fragments were joined at a common SnaB I restriction site and expressed in Escherichia coli at a level equivalent to what is normally found in C. cylindrosporum. Sequence comparisons show a large degree of homology with genes from two other clostridial species, including a thermophile. Certain conserved sequences found in the three clostridial proteins and in the N10-H4folate synthetase portion of eukaryotic C1-H4folate synthases may represent consensus sequences for nucleotide and H4folate binding.