Human mitochondrial C1-tetrahydrofolate synthase: gene structure, tissue distribution of the mRNA, and immunolocalization in Chinese hamster ovary calls

Abstract

C1-tetrahydrofolate (THF) synthase is a trifunctional enzyme found in eukaryotes that contains the activities 10-formyl-THF synthetase, 5,10-methenyl-THF cyclohydrolase, and 5,10-methylene-THF dehydrogenase. The cytoplasmic isozyme of C1-THF synthase is well characterized in a number of mammals, including humans; but a mitochondrial isozyme has been previously identified only in the yeast Saccharomyces. Here, we report the identification and characterization of the human gene encoding a functional mitochondrial C1-THF synthase. The gene spans 236 kilobase pairs on chromosome 6 and consists of 28 exons plus one alternative exon. The gene encodes a protein of 978 amino acids, including an N-terminal mitochondrial targeting sequence. The mitochondrial isozyme is 61% identical to the human cytoplasmic isozyme. Expression of the gene was detected in most human tissues, but transcripts were highest in placenta, thymus, and brain. Two mRNAs were detected, a 3.6-kb transcript and a 1.1-kb transcript, and both transcripts were observed in varying ratios in each tissue. The shorter transcript results from an alternative splicing event, where exon 7 is spliced to exon 8a instead of exon 8. Exon 8a is derived from an exonized Alu sequence, sharing no homology with exon 8 of the long transcript, and encodes just 15 amino acids followed by a stop codon and a polyadenylation signal. This short transcript potentially encodes a bifunctional enzyme lacking 10-formyl-THF synthetase activity. Both transcripts initiate at the same 5'-site, 107 nucleotides up-stream of the ATG start codon. The full-length (2934 bp) cDNA fused to a C-terminal V5 epitope tag was expressed in Chinese hamster ovary cells. Immunoblots of subfractionated cells revealed a 107-kDa protein only in the mitochondrial fractions of these cells, confirming the mitochondrial localization of the protein. Yeast cells expressing the full-length human cDNA exhibited elevated 10-formyl-THF synthetase activity, confirming its identification as the human mitochondrial C1-THF synthase.

Figure 1

Compartmentation of folate-mediated one-carbon metabolism in eukaryotes. Reactions 1, 2, and 3, 10-formyl-THF synthetase (EC 6.3.4.3), 5,10-methenyl-THF cyclohydrolase (EC 3.5.4.9) and 5,10-methylene-THF dehydrogenase (EC 1.5.1.5), respectively, are catalyzed by C₁-tetrahydrofolate synthase. The other reactions are catalyzed by 4, serine hydroxymethyltransferase (EC 2.1.2.1); 5, glycine cleavage system (EC 2.1.2.10); 6, methylenetetrahydrofolate reductase (EC 1.5.1.20); 7, methionine synthase (EC 2.1.1.13); 8, methionyl-tRNA formyltransferase (EC 2.1.2.9); 9, methenyltetrahydrofolate synthetase (EC 6.3.3.2). Not all coenzymes are shown for all reactions.

Figure 2

Alignment of human and mouse mitochondrial C₁-THF synthase with human cytoplasmic C₁-THF synthase. A black box denotes identity, and a white box denotes conservative substitutions or identities in two out of three proteins. The alignment was produced by the INRA server at the Laboratoire de Génétique cellulaire (http://prodes.toulouse.inra.fr/multalin/multalin.html) using the MultAlin algorithm (56) and the output was generated by the ESPript program at the same site. hmito; human mitochondrial C₁-THF synthase; mmito, mouse mitochondrial C₁-THF synthase; hcyto, human cytoplasmic C₁-THF synthase.

Figure 3

Subcellular localization of epitope-tagged human mitochondrial C₁-THF synthase expressed in CHO cells. Coomassie blue-stained SDS PAGE (left) and immunoblot (right) of cytoplasmic (c) and mitochondrial (m) fractions from CHO cells transfected (T) with either pcDNA3/humito (lanes 1,2) or untransfected control (C) (lanes 3,4). Each lane contains 80 μg total protein.

Figure 4

Intron/exon structure of human mitochondrial C₁-THF synthase gene. Exons are shown as numbered black bars; introns as thin horizontal lines. Exon and intron sizes and positions are drawn roughly to scale, with the exception of intron 26, which is 55,350 bp. The entire gene spans 236 kbp. The actual sizes of each exon and intron are listed in Table II.

Figure 5

Northern blot analysis of mitochondrial C₁-THF synthase transcripts in adult human tissues. A multiple human tissue RNA blot was hybridized with ³²P-labeled probes to the 5′ end (panel A) or 3′ end (panel B) of human mitochondrial C₁-THF synthase cDNA. In panel C, the membrane was hybridized with a probe from the 3′ end of the human cytoplasmic C₁-THF synthase cDNA. The lanes on each membrane contain RNA from (left to right) brain, placenta, skeletal muscle, heart, kidney, pancreas, liver, lung, spleen, and thymus. The schematic diagram below shows the relative locations of the probes used for panels A and B on the 3.6 and 1.1 kb transcripts.

Figure 6

5′ flanking sequence and exon 1 of human mitochondrial C₁-THF synthase gene. Nucleotides are numbered with the A of the ATG start codon as +1. Coding nucleotides are in upper case; lower case nucleotides represent 5′ non-coding or intron sequences. The cDNA-specific outer and inner primers used in the 5′RACE are indicated by the arrows labeled GSO2 and GSI2, respectively. The arrow beginning at –107 indicates the transcriptional start site based on 5′RACE. The asterisks indicate the 5′ ends of ESTs found in the Human EST Database.

Figure 7

Alternative splicing of human mitochondrial C₁-THF synthase transcript. A. Gene structure and splicing pattern. Alternative exon 8a is in the intron between exons 7 and 8. In the 3.6 kb transcript, exon 7 is spliced to exon 8. In the 1.1 kb transcript, exon 7 is spliced to exon 8a. Exon 8a contains a stop codon (black dot) after 15 sense codons, and contains a polyadenylation signal (AATAAA) near its 3′ end. There is no homology between exons 8a and 8. B. Potential protein products. The long transcript is translated into a 978 aa protein, whereas the short transcript is translated into a 275 aa protein. The amino acid sequence of the two proteins is identical through the first 7 exons, encoding 260 aa. The short protein has 15 aa from exon 8a in place of the aa encoded by exon 8. The junction between the dehydrogenase/cyclohydrolase (D/C) and synthetase (SYN) domains is predicted to lie within aa 330-350 (Fig. 2). The asterisk (*) indicates variable 3′ splice site selection at the exon 6/7 junction (see Fig. 8). C. Nucleotide and amino acid sequence of coding sequence of exon 8a. Sequences 3′ to the stop codon not shown.

Figure 8

Variable 3′ splice site selection at exon 6/7 junction. Top, intron/exon junctions for 3′ end of exon 6 and 5′ end of exon 7. The first two nt of the intron, and the alternative AG splice acceptors sites, are underlined. Bottom, alternative splicing products if the first (left) or second (right) AG acceptor site is used. The amino acid sequence encoded by each product is shown below the nt sequence, and the extra codon and amino acid are in bold.