Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
# 2002 Oxford University Press Human Molecular Genetics, 2002, Vol. 11, No. 26 3361–3369 Identification of the gene responsible for the cblB complementation group of vitamin B12-dependent methylmalonic aciduria C. Melissa Dobson1, Timothy Wai2, Daniel Leclerc4, Hakan Kadir1, Monica Narang1, Jordan P. Lerner-Ellis2, Thomas J. Hudson2,3,5, David S. Rosenblatt2,3,4 and Roy A. Gravel1,* 1 Department of Biochemistry and Molecular Biology, University of Calgary, Canada, 2Department of Human Genetics and 3Department of Medicine, McGill University, Montreal, Canada, 4McGill University Health Center, McGill University, Montreal, Canada and 5Montreal Genome Center, Montreal, Canada Received September 27, 2002; Revised and Accepted October 25, 2002 DDBJ/EMBL/GenBank accession no. AF550396–AF550404 The methylmalonic acidurias are metabolic disorders resulting from deficient methylmalonyl-CoA mutase activity, a vitamin B12-dependent enzyme. We have cloned the gene for the cblB complementation group caused by deficient activity of a cob(I)alamin adenosyltransferase. This was accomplished by searching bacterial genomes for genes in close proximity to the methylmalonyl-CoA mutase gene that might encode a protein with the properties of an adenosyltransferase. A candidate was identified in the Archaeoglobus fulgidus genome and was used to probe the human genome database. It yielded a gene on chromosome 12q24 that encodes a predicted protein of 250 amino acids with 45% similarity to PduO in Salmonella enterica, a characterized cob(I)alamin adenosyltransferase. A northern blot revealed an RNA species of 1.1 kb predominating in liver and skeletal muscle. The gene was evaluated for deleterious mutations in cblB patient cell lines. Several mutations were identified including a 5 bp deletion (5del572gggcc576), two splice site mutations (IVS2–1G>T, IVS3–1G>A), andt several point mutations (A135T, R186W, R191W and E193K). Two additional amino acid substitutions (R19Q and M239K) were found in several patient cell lines but were found to be common polymorphisms (36% and 46%) in control alleles. The R186W mutation, which we suggest is disease-linked, is present in four of the six patient cell lines examined (homoallelic in two) and in 4 of 240 alleles in control samples. These data confirm that the identified gene, MMAB, corresponds to the cblB complementation group and has the appearance of a cob(I)alamin adenosyltransferase, as predicted from biochemical data. INTRODUCTION In mammals, vitamin B12 (cobalamin, Cbl) functions as a coenzyme for the enzymes methionine synthase and methylmalonyl CoA mutase (MCM). Methylcobalamin (MeCbl) is the coenzyme for methionine synthase and functions in methyl transfer to homocysteine to form methionine. Adenosylcobalamin (AdoCbl) is the coenzyme for MCM and participates in the rearrangement reaction that converts methylmalonyl CoA to succinyl CoA. In microbes there are numerous AdoCbl-dependent adenosyltransferases and they participate in a variety of intramolecular rearrangements. In all organisms, whether they synthesize vitamin B12de novo or acquire it externally, the Cbl must be converted from a Co(III) oxidation state to Co(I) prior to addition of the functional ligand. In humans, these reactions are becoming known through genetic disorders in patients who have blocks in subcellular processing or enzymatic reactions leading to the active coenzymes. Eight complementation groups have been identified among patients with blocks in the cellular metabolism of vitamin B12 (1). Three of these, cblA, cblB and cblH, are blocked uniquely in the synthesis of AdoCbl. All are expected to involve mitochondrial functions, since this is the subcellular location of methylmalonyl-CoA mutase. To date, genes for three of the eight complementation groups have been identified. The gene responsible for the cblA complementation group, MMAA, encodes a protein of unidentified function that may be involved in vitamin B12 transport into mitochondria (2). Two other identified genes are involved in the functional expression of methionine synthase (3). These are the MTR gene, encoding methionine synthase and defined by the cblG complementation group (4–6), and MTRR, encoding the *To whom correspondence should be addressed at: Department of Biochemistry and Molecular Biology, Room 250, Heritage Medical Research Building, 3330 Hospital Drive NW, Calgary, AB, Canada T2N 4N1. Tel: 403 2202268; Fax: 403 2108115; Email: [email protected] 3362 Human Molecular Genetics, 2002, Vol. 11, No. 26 reactivating enzyme, methionine synthase reductase, and defined by the cblE complementation group (7). The remaining groups, cblC, cblD and cblF, affect both MeCbl and AdoCbl synthesis, likely involving lysosomal efflux of Cbl into the cytosol (cblF) and initial steps in the reduction of Cbl (cblC and cblD) (1) (Fig. 1). This study concerns the identification of the gene responsible for the cblB complementation group (8). Patients comprising this group have methylmalonic aciduria and metabolic ketoacidosis, despite a functional methylmalonyl-CoA mutase (1). In severe cases, they may present in the newborn period with lethargy, vomiting and failure to thrive. They have normal methionine synthesis and normal levels of MeCbl. Studies on cell extracts from fibroblast cultures from cblB patients suggested that they were blocked in the mitochondrial synthesis of AdoCbl, either in the reduction of Cbl or in a cob(I)alamin adenosyltransferase (9,10). Subsequently, Fenton and Rosenberg (11) confirmed a deficiency of cob(I)alamin adenosyltransferase activity. While the evidence has pointed to cob(I)alamin adenosyltransferase as the enzyme responsible for the cblB defect, there has been little success in characterizing the protein further and the gene has remained unidentified. Microbes, in contrast, utilize vitamin B12 in numerous reactions and distinct corrinoid adenosyltransferases have been identified and shown to participate in several metabolic pathways. Some bacteria and archea are able to synthesize vitamin B12 (12–16), which can proceed by anaerobic or aerobic pathways requiring over 25 enzymes (for review see 17). Others, including E. coli, do not synthesize vitamin B12 but do utilize it metabolically. Cbl is a cofactor in a wide variety of reactions, including acetyl-CoA synthesis, methyl transfer for methane production, ribonucleotide reductase and in the fermentation-dependent utilization of a variety of molecules, including ethanolamine, glutamate, glycerol and 1,2-propanediol (17). Methane production and acetyl-CoA synthesis require MeCbl, while ribonucleotide reductase and fermentation enzymes require the AdoCbl form. To produce AdoCbl for a range of metabolic pathways under different regulatory regimens, bacterial genomes have evolved several genes encoding distinct corrinoid adenosyltransferases specific to different enzyme systems (17). Interestingly, the different groups of adenosyltransferases can be structurally distinct. Based on amino acid sequence similarity, the adenosyltransferases required for the synthesis of AdoCbl can be divided into three main families: CobA-type, EutT-type and PduO-type. CobA, an ATP:corrinoid adenosyltransferase, participates in Cbl biosynthesis by the adenosylation of an intermediate prior to Cbl, but it also acts as a cob(I)alamin adenosyltransferase with the ability to form AdoCbl directly. It can generate AdoCbl required by AdoCbl-dependent enzymes, such as 1,2-propanediol dehydratase, as effectively as authentic AdoCbl (18). CobA is encoded by the cobA gene in Salmonella enterica, cobO in Paracoccus denitrificans and btuR in Salmonella typhimurium. The three-dimensional structure of CobA has been solved and revealed a unique nucleotide binding orientation to the P-loop of CobA by comparison with all other nucleotide hydrolases (19). The unusual orientation of the nucleotide arises because this enzyme transfers the adenosyl group rather than the gamma-phosphate to its acceptor. However, the two other cob(I)alamin adenosyltransferase families do not contain a recognizable P loop-nucleotide binding motif. Their mechanism of ATP binding and adenosyl release is unknown. One of the adenosyltransferases, EutT, belongs to an operon of Salmonella typhimurium encoding proteins involved in the Cbl-dependent degradation of ethanolamine (20). The other is PduO, an ATP:cob(I)alamin adenosyltransferase in Salmonella enterica that is integral to growth on 1,2-propanediol (21). Interestingly, PduO was proposed to be a bifunctional protein, acting as an adenosyltransferase as well as a reductase that might convert cob(II)alamin to cob(I)alamin (21). We set out to identify the cob(I)alamin adenosyltransferase defective in the cblB complementation group by conducting homology searches using bacterial adenosyltransferases as query sequences. The identification of the cblB gene came from making use of bacterial gene clusters to identify the relevant bacterial template. It was detected as one of the potential genes in operons containing bacterial orthologs to the human MCM gene and to the recently identified human MMAA gene. Several mutations confirming the authenticity of the gene identification were found. We propose the gene name, MMAB, to denote ‘methylmalonic aciduria (cobalamin deficiency) type B.’ RESULTS Our initial efforts to identify genes that might represent a human cobalamin adenosyltransferase focused on known bacterial ATP:cob(I)alamin adenosyltransferases. Previous reports on different Cbl-dependent metabolic pathways allowed the search to be narrowed to specific families of proteins, including CobA, EutT and PduO. To further narrow the search for candidates, we made use of the Clusters of Orthologous Groups of proteins (COG) database at NCBI (htttp:// www.ncbi.nlm.nih.gov/COG) to search for genes in microbial organisms that were arranged in clusters with genes encoding Cbl-dependent enzymes. A search for methylmalonyl-CoA mutase yielded COG2185, corresponding to the Cbl-binding, carboxyl-terminal domain/subunit of MCM. Adjacent to it is COG1703, which proved to be the ortholog of the gene responsible for the cblA complementation group (MMAA) (2). Flanking COG1703 is COG2096 in Archaeoglobus fulgidus. This COG, an open reading frame, is described as an ‘uncharacterized acr’ (ancient conserved region), but it has sequence similarity to PduO, an ATP:cob(I)alamin adenosyltransferase from Salmonella enterica (21). Several other genomes, not identified in the COG database, were also found to contain these three genes in cluster form, including Sulfolobus solfataricus, Pyrococcus horikoshii, Pyrococcus abyssi, Thermoplasma volcanium and Thermoplasma acidophilum. Identification of the MMAB cDNA The Archaeoglobus fulgidus sequence (gi:11498888) was used as the template for a BLASTP search of the NCBI nonredundant database (www.ncbi.nlm.nih.gov/blast). It resulted in a 37% similarity (22% identity) with the N-terminal half of Salmonella enterica PduO (gi:5069458) and 43% similarity (28% identity) to the full-length of an unknown Human Molecular Genetics, 2002, Vol. 11, No. 26 3363 Figure 1. Intracellular processing of Cbl. Steps are identified according to complementation group names. Their gene (upper case italics) and predicted functions are: cblA (MMAA), unknown function, proposed to be an accessory to mitochondrial import of B12; cblB, cob(I)alamin adenosyltransferase (this study, MMAB); cblC/D, possibly cob(III)alamin reductase; cblE (MTRR), methionine synthase reductase; cblF, lysosomal Cbl efflux; cblG (MTR), methionine synthase; cblH, possibly a component of cobalamin transport or cob(II)alamin reductase. protein in humans (gi:15080110). This protein was constructed from a predicted mRNA and supported by the assembly of expressed sequence tags (ESTs) found in the dbEST (www.ncbi.nlm.nih.gov) (22). A Unigene cluster (Hs.12106) containing 119 ESTs constitutes the candidate MMAB cDNA. Based on the assembly of these ESTs, the mRNA size was found to vary at the 30 end, with two sizes of 1128 (BC005054) and 2361 nucleotides (BC011831) in length, due to the evident use of alternate polyadenylation sites. Both transcripts would result in an open reading frame of 250 codons with the same start and stop codons and differing only in the location of the polyadenylate track. Based on the ESTs examined in the database, there was no evidence of variant sequence that would be generated from alternate splicing (Fig. 2). The coding region from the cDNA sequence was verified experimentally by sequencing a 1075 bp RT–PCR product (description of primers in Table 1) from the transformed human liver cell line Huh-7. It contains an open reading frame 750 bp in length and predicts a polypeptide of 250 amino acids with a calculated molecular mass of 27.3 kDa (Fig. 3). Orthologous proteins Sequences corresponding to the candidate MMAB cDNA were also identified in the mouse and C. elegans database as complete predicted cDNAs. This allowed us to verify the reading frame throughout and to enssure that the sequence is full-length. Further work was done to verify the ESTs including reconstructing the appropriate length mRNA. The mouse Unigene cluster Mm.19757 contains 8 ESTs. These were assembled into a composite cDNA sequence, which verified the corresponding GenBank entry (gi:18381148). The mouse cDNA is 86% identical to the human cDNA. The corresponding proteins are 88% identical, with similarity based on conserved amino acid properties of 91%. The human, mouse, C. elegans, Archaeoglobusfulgidus, and Salmonella enterica sequences were aligned with the proposed human coding sequence (Fig. 3). Structure of the MMAB gene and chromosomal location The MMAB gene was localized to chromosome 12q24 (Fig. 2) based on the linkage of contig NT_009770 to markers DS12S172, DS12S234, RH45414 and the mapping of sequence tagged sites (STS), including WI-14208, A005L23 and STSG2679. To define the structure of the human gene, the predicted cDNA sequences (longer and shorter versions) were compared to the whole human genome using BLASTN. Nine hits were obtained with 100% continuous alignment in each segment, indicating that the gene contains nine exons (Fig. 2). Exon 9 ends at two alternative polyadenylation sites, one at nucleotide 1080 in the mRNA sequence and the second at nucleotide 2307. The first one shows a consensus polyadenylation signal, AATAAA, 20 bp upstream of the polyadenylation site. Interestingly, there is a TG dinucleotide repeat (23 repeats) beginning 9 bp beyond the first polyadenylation site. In total, the human MMAB gene was found to encompass 18.87 kb and is oriented towards the telomere. The coding sequence is flanked by a mevalonate kinase gene on the 50 side (55 kb away) and Loc160753 encoding a protein similar to fatty aldehyde dehydrogenase (35 kb away) followed by an ubiquitin protein ligase (57 kb away) on the 30 side. Comparison of the human and mouse genes show preservation of all nine exon–intron splice junctions. The mouse and human chromosomes show a large amount of synteny around the MMAB gene including the genes for mevalonate kinase, fatty aldehyde dehydrogenase and ubiquitin protein ligase. This is in agreement with the human–mouse homology map (http:// www.ncbi.nlm.nih.gov/) that links human 12q24 with chromosome 5 of mouse. Expression in tissues A 338 bp probe, amplified from positions 276–613 bp of the full-length cDNA, hybridized to an RNA species of 1.1 kb on the human tissue northern blot, with highest levels of expression in liver and skeletal muscle (Fig. 4). The length of RNA is consistent with the predicted length of the shortest composite EST assembly of 1.1 kb. The longer version was not observed, but this may be in keeping with the presence of a 3364 Human Molecular Genetics, 2002, Vol. 11, No. 26 Figure 2. Structural organization of the human MMAB gene. ‘Genes’ show the flanking genes and map positions surrounding MMAB according to mapviewer at NCBI (http://www.ncbi.nlm.nih.gov) on chromosome 12q24.1 to 12q24.2. The ‘Exons’ line identifies the location of exons (numbered boxes) and introns (lines) along the MMAB gene located in Contig NT_009770.12. Our independently confirmed sequences from genomic DNA were deposited in GenBank under the accession numbers AF550396–AF550404. Introns are drawn to scale. The cDNA (‘cDNA’) is illustrated, showing the splice sites and start/stop codons. The hatched section is the longer predicted sequence based on EST assembly. The cDNA is illustrated, with splice and start/stop codons indicated, showing that it was constructed from numerous ESTs belonging to Unigene cluster Hs.12106 (119 ESTs) and verified by RT–PCR sequencing. Only some ESTs are illustrated here. From top to bottom are BG754143, BI192839, BE250480, AA703387, BM462123, AI691079, A702361, AA340851, BE391487 and AV701242 . The dotted ESTs are the longer cDNA predicted based on assembly of the ESTs contained within Unigene cluster Hs.12106. Table 1. Primers used to amplify the exons and flanking introns of the cblB gene and the splice junctions Exon Primer Sequencea Primer Sequence 1 2 3 4 5 6 7 8 9 1F 2F 3F 4F 5F 6F 7F 8F 9F accaataaaaattgaaacctgagcg agatgaccatctaagcagtaagg gcctggatgacagagtgactgtt gactggaaataatgagcatccga cacctccaactctggggatcaca cacgggtgggtcagttagccttgct atccctcttttcacagtgacaga ctcaccagccatccttgatccct cattaagtgcaagatggctgggc 1R 2R 3R 4R 5R 6R 7R 8R 9R aacagcgtggagacgtcacctgac cgcgcctggtatattatacttta ataagggctgacaacctccgagg taactggtggctggatgctgagt tgggatcccagatggtgacccta ttggggaactggaggacagcgtct gtctgttgatgtgtggctggatg acactaagggaatgaaggagggg tccttctcattgcagcaatcact RNA RNAF gcacgagggtcaagcagcctg RNAR gcacagctccttctcattgcag PCR size 365 304 338 357 339 386 483 430 561 exon size >153 62 94 58 73 98 65 60 436 to polyA 50 spliceb tagGCCA . . . cagGGTT . . . tagGTTT . . . cagATCC . . . cagAGTA . . . cagGCAA . . . tagTGTG . . . cagACTC . . . 30 splice Exon boundaries . . .ACAGgtg . . .AAAGgta . . .TTGGgta . . .GAAAgtg . . .TTAAgta . . .GTAGgta . . .GACGgta . . .ACAGgta >–18–134 135–196 197–290 291–348 349–421 422–519 520–584 585–644 645–1080 1080 a The primer sequences are listed 50 to 30 . The upper case letters designate exon sequence and the lower case letters designate the intron sequences. b very small proportion of 30 ESTs containing the longer sequence. The presence of an efficient polyadenylation signal at the first polyadenylation site (not shown), but not at the second, may be the explanation for it not being detected. Mutations in the cblB complementation group Fibroblast lines from patients from the cblB complementation group were analysed for mutations by PCR-heteroduplex analysis to detect multibase changes and by DNA sequencing of PCR-amplified exons and flanking sequences to detect point mutations. In each case, a PCR-dependent diagnostic test was developed using restriction enzyme digestion to distinguish normal and mutant sequences. The primers used for sequencing are listed in Table 1. The diagnostics used to verify the mutation and test the normal population are listed in Table 2. If the mutation did not result in the loss or addition of a restriction site, one was created with modified primers. Otherwise the same primers were used as with sequencing. Mutant cell lines and their mutations are listed in Table 3. Human Molecular Genetics, 2002, Vol. 11, No. 26 3365 Figure 3. Alignment of amino acid sequences similar to that of human (Hsa) MMAB, including mouse (Mmu) (gi:12860826), Caenorhabditis elegans (Cel) (gi:532812), Archaeoglobus fulgidus (Aful) (gi:2649297) and Salmonella enterica (Sent) (gi: 5069458). The arrow denotes the predicted mitochondrial leader sequence cleavage site. Amino acids that are similar in all species are shaded with progressively increasing grey background. Of note is the highly conserved region (underlined). The letters (amino acid designations) above the human (Has) sequence correpond to amino acid mutations in patient cell lines; R19Q and M239K are polymorphisms. Table 2. Mutation detection including diagnostic tests and expected banding patterns Mutation Diagnostic enzyme Forward primer useda Reverse primer Normal fragments (bp)b Mutant fragments (bp)b 556C>T (R186W) MspIa 98, 22 HindIII 130 107, 23 56–57GC>AA (R19Q) 716 T>A (M238K) 403 G>A (A135T) 571C>T (R191W) IVS2–1G>T 575G>A (E193K) HhaI NlaIII TaqI MspI NlaIII HaeI 7Rshort: ctccagccctcttaccgtct IVS3R: tctgtgactaattccagagcaag 1R 9R 5R 7R 3R 7E193KR: gtctgttgatgtgtggctggatg 83, 15, 22 IVS3–1 G>A 7Fshort: cacacctttcccaccctgct IVS3F: gctaaattaactaaatggcctgcc 1F 9F 5F 7F 3F 7E193KF: tttctgccgggccgtgtgccgcctgg 186, 202, 339 190, 304, 213 134, 31, 14 122, 73, 29 132, 63, 44, 39, 15 34 186, 202, 217, 205, 103, 186, 148, 122, 122 132, 201, 27 31 102 63, 44, 39, 15 34 a The primer sequences are listed 50 to 30 when given or refer to the primers used in Table 1. The bold segments are the ones observed on Figure 7. b Heteroduplex analysis revealed a single multibase deletion in cell line WG2633 in a PCR of exon 7 and flanking sequences. DNA sequencing revealed a 5 bp deletion, 572– 576delGGGCC, which disrupts the reading frame after R190 (Fig. 5A). The heteroduplex test was used as the diagnostic. It was not observed in any other mutant line. The second mutation in WG2633 is a point mutation, 556C>T, which generates an amino acid substitution, R186W. This mutation 3366 Human Molecular Genetics, 2002, Vol. 11, No. 26 IVS2 –1G > T. It was validated in a PCR test using restriction enzyme NlaIII (Fig. 5H). The second mutation is 575G>A (E193K). This mutation was detected by PCR using an artificial restriction site, MscI (Fig. 5I). The mutation was not observed in any other mutant line or in 48 control alleles. DISCUSSION Figure 4. Northern blot analysis of MMAB mRNA. A northern blot of poly(A)þ RNA from the indicated human tissues. The membrane was hybridized with a [32P]-labeled 338 bp probe from MMAB mRNA (A) or a b-actin probe (B). A molecular size marker positions is indicated on the right. was identified in four of six cell lines (six of 12 alleles) during the DNA sequencing survey (Fig. 5B). WG1185 and WG2186 are both homoallelic for this mutation, while WG2027 and WG2633 are heteroallelic. One hundred and twenty non-cblB cell lines were tested for 556C>T and four were found to contain this genetic variation in heterozygous fashion. This gave an allele frequency of 1/60. It remains uncertain as a disease-causing mutation. The second mutation in WG2027 is at the acceptor splice site of intron 3, IVS3–1G>A, which would be expected to disrupt splicing. The mutation was confirmed by testing for the generation of an artificially created HindIII site in a 130 bp PCR product (Fig. 5C). The mutation was not observed in any other mutant line or in 194 control alleles. WG2492 and WG2523 were found to contain a 2 bp substitution, 56–57GC>AA (R19Q). However, this change is also observed in 6 of 27 ESTs included in Hs.12106. A diagnostic test was generated which tests for the mutation by digestion with HhaI (Fig. 5D). Based on the examination of 97 normal cell lines, 36% of the alleles contained the mutation. Figure 5D illustrates two normal cell lines, one homozygous for GC (lane 1) and one homozygous for AA (lane 4). As a result, we surmise that R19Q is a frequent polymorphism that is unlikely to be contributing to disease. Another amino acid substitution, 716T>A (M239K) was also observed in WG2523 (Fig. 5E). This is also recorded as a polymorphism in the dbSNP as rs9593 (www.ncbi.nlm.nih.gov/SNP) with heterozygosity reported at 48% by Ensembl (www.ensembl.org). In keeping with this, we found a frequency of 46% in 72 normal alleles tested. As expected, two additional mutations were identified in WG2492. Both are point mutations, 403 G>A (A135T) and 571C>T (R191W). The A135T mutation results in the creation of a TaqI site (Fig. 5F). The R191W mutation can be observed in conjunction with digestion with MspI (Fig. 5G). These mutations were not observed in any other patient cell line or in 72 alleles tested in control samples. Along with R19Q and M239K, WG2523 also contains two additional point mutations. One is a splice mutation, We have combined the analysis of prokaryotic gene arrangements and specific protein function reported in bacteria to identify the disease-causing gene in patients from the cblB complementation group. The key finding confirming the identity of the MMAB gene was the detection of null-type mutations, including a 5 bp deletion frameshift and two splice junction mutations (Table 3). In addition, the deduced MMAB protein aligns with PduO from Salmonella enterica (Fig. 4), as enzyme that is biochemically confirmed as a cob(I)alamin adenosyltransferase (21). Early studies pointed to a mitochondrial defect in the human disorder since the block in AdoCbl synthesis was localized to mitochondria (11). As anticipated, MMAB has a predicted leader sequence and signal cleavage site (P ¼ 0.9755, http:// www.mips.biochem.mpg.de/cgi-bin/proj/medgen/mitofilter) at residue 33 (Fig. 4). Examination of amino acid sequence patterns in the MMAB protein did not reveal any known functional domains. Instead, the sequence was found to contain the DUF80 domain (domain of unknown function) defined by pfam01923 (http://pfam.wustl.edu). Significantly, this pfam domain can also be found in the EutT adenosyltransferase in addition to PduO; yet, none of these proteins contains a P loop as described for CobA. We anticipate that this conserved domain will be important for Cbl binding or the adenosylation function. This sequence is conserved in a variety of MMAB orthologs (including those listed in Fig. 3) and the DUF80 consensus sequence in this region is R-[AT]-[LIVQ]-x-R-R[ALS]-E-R. Interestingly, a large number of mutations (R186W, R191W and E193K) focus on this sequence in the human protein, which reads 186RAVCRRAERR195. In PduO from Salmonella, the protein was proposed to have two functions, the adenosyltransferase and a Co(II) to Co(I) reductase (21). As well, these functions were proposed to be localized to the N- and C-terminal halves of the protein, respectively. The human ortholog aligns only with the Nterminal half of PduO, suggesting that it does not contain a function corresponding to its C-terminal half. In this regard, previous studies showed that cblB cells have a functional cob(II)alamin reductase (23). This suggests that another protein may have evolved in humans to account for the cob(II)alamin reductase function. The presence of mutations in the patients belonging to the cblB complementation group verifies this hypothesis. Several confirmatory mutations were identified, including one deletion of 5 bp (del572–576gggcc) and two splice mutations, IVS2–1 G>T and IVS3–1G>A, at obligatory 1 positions. In addition, four missense mutations, A135T, R186W, R191W and E193K, were found among the six cblB cell lines examined. Based on our analysis, we conclude that the sequence variant R186W is a rare polymorphism (4 out of 240 alleles) in the general population. However, the significantly increased frequency of Human Molecular Genetics, 2002, Vol. 11, No. 26 3367 Figure 5. PCR-based diagnostic assay. (A) Heteroduplex test for 572–576delgggcc: lane 1, normal cells; lane 2, WG2633 (deletion on one allele). (B) Test for R186W: lane 1, normal; lane 2, WG2523; lane 3, WG2186, all following digestion with MspI using 7Fshort and 7Rshort primers. (C) Test for IVS3–1 G>A: lane 1, normal undigested; lane 2: normal; lane 3: WG2186, all digestions with HindIII using IVS3F and IVS3R primers. (D) Test for R19Q: lane 1, normal; lane 2, WG2492; lane 3, WG2523; lane 4, normal #2; lane 5, normal #3, all following digestion with HlaI. (E) Test for M239K: lane 1, normal undigested; lane 2: normal; lane 3, WG2523; lane 4, normal, all digestions completed with NlaIII. (F) Test for A135T: lane 1, normal undigested; lane 2, normal; lane 3, WG2186; lane 4, WG2492, all digestions performed using Taq1. (G) Test for R191W: lane 1, normal undigested; lane 2, normal; lane 3, WG2492 all digestion performed with MspI. (H) Test for IVS2–1G>T: lane 1, normal undigested; lane 2, normal; lane 3, WG2523, all digestions performed with NlaIII. (I) Test for E193K: lane 1, normal undigested; lane 2, normal; lane 3, WG2523, all digestion performed with MscI using E193KF and E193KR specific primers. Asterisks denote diagnostic bands. Note throughout, small fragments not included in gel patterns (ran off the gel). Table 3. Summary of mutations listed Cell lines Mutation 1 Mutation 2 Propionate uptakea Basal/ þ B12 AdoCbl levelb Ethnic origin Age at onset WG1185 WG2027 WG2186 WG2492 WG2523 WG2633 R186W IVS3–1G>A R186W A135T IVS2–1G>T 572-576delgggcc R186W R186W R186W R191W E193K R186W — 0.89/0.91 0.99/1.1 2.0/3.6 0.22/0.45 0.47/0.50 4.29% 13.1% 2.5% 6.5% 1.8% 0.8% — Caucasian — Black Saudi Caucasian — 3 days <7 days 3 years <1 years 7days a C-propionate uptake into fibroblasts: Controls: Basal (no added vitamin B12 to culture medium) ¼ 10.8 3.2 nmol/mg protein/18 h þ B12 (vitamin B12 added to medium) ¼ 10.9 3.5 nmol/mg protein/18 h (24). b As percent of total Cbl (control values, 15.29 4.2%) (25). this variant in cblB patient cell lines (6 out of 12 alleles) remains intriguing since there is no evidence of consanguinity in the two individuals who are homozygous for the mutation. It will be important to study the prevalence of the R186W mutation on a larger scale to determine if unaffected homozygotes are identified and to determine if harmful catabolic products are elevated in such individuals. Analysis of the protein with this site mutated will also be required to shed light on the importance of this amino acid residue. Significantly, the other nearby mutations (R191W, E193K), which are clustered in this highly conserved region, adds weight to the importance of this area to enzyme function. This heterogeneity of mutations and their clustering in a very tight region may address the relationship between genotype and the response to high levels of OH-Cbl. This relationship is not clear from our initial studies and it is evident that expression experiments will be required to assess the impact of the different mutations. For example, WG2633, with the 5 bp deletion in exon 7 and R186W, had a decreased AdoCbl level (0.8%) compared with a normal value of 15% of total Cbl. In contrast, WG2027, which has a splice mutation and R186W, had 13% AdoCbl (comparable to normal limits) but the patient was sick within three days of birth. Interestingly, these data bracket the values for WG1185 and WG2186, which are 3368 Human Molecular Genetics, 2002, Vol. 11, No. 26 homozygous for R186W. They had levels of 4.28% and 2.5% AdoCbl. In summary, we conclude that the MMAB gene encodes the ATP:cob(I)alamin adenosyltransferase as predicted by mutation studies and sequence similarity to PduO (gi: 5069458). Sequences in prokaryotes such as Archaeoglobus fulgidus, which contain orthologs of MCM, MMAA and MMAB in contiguous arrangement, add compelling weight to this enzyme playing a valuable role in the MCM pathway and demonstrate the enormous power of prokaryotic gene arrangements for the identification of human genes. MATERIALS AND METHODS Patient cell lines Fibroblast lines from six patients classified as members of the cblB complementation group were used in this study: WG1185, WG2027, WG2186, WG2492, WG2523 and WG2633. Some of the cell lines were characterized by 14C-propionate uptake (24) and AdoCbl content as percent of total Cbl following incubation of cells for 96 h in medium containing [57Co] OH-Cbl (25). Values for cell lines and controls are given in Table 3. The table also shows ethnic origins and age of onset of disease where known. WG1185 has been reported previously (11). WG2186 presented with respiratory distress and metabolic acidosis, hypocalcemia, thrombocytopenia, neuropenia and moderate hyperammonenia. WG2523 is of Saudi descent whose parents are said to be second cousins (note, distinct alleles were identified in this patient). Materials Oligonucleotide primers were synthesized by Qiagen (Mississauga, Ontario). Taq polymerase and Trizol were purchased from Invitrogen (Burlington, Ontario). Restriction enzymes were purchased from New England Biolabs (Mississauga, Ontario) and Invitrogen. Genomic-tip columns, QIAquick DNA purification kit and Ominiscript amplification kit were purchased from Qiagen (Mississauga, Ontario). PGEM cloning vector was purchased from Promega (Madison, WI). Big Dye terminator version 2 and 3 were purchased from Applied Biosystems and were cycle-sequencing products were characterized on ABI 3700 and 3100 Genetic Analyzers (Foster City, CA). The multiple tissue RNA blot was purchased from Clontech (Palo Alto, CA). QIAquick PCR purification kit and cloned into pGEM according to the supplier’s instructions. Heteroduplex analysis, DNA sequencing and restriction analysis Initial examination of PCR products was performed by heteroduplex analysis, as described previously (7). Heteroduplexes were obtained by mixing, melting and reannealing equivalent concentrations of PCR products of mutant and control DNA prior to electrophoresis. Positive cell lines also underwent the same procedure without mixing with control to determine if the mutation was on one or both alleles. The observation of a heteroduplex, with implication of a multibase mutation, was taken as indication that the correct gene was identified. DNA sequencing was accomplished in 96-well plates in 10 ml reactions made up of 2 ml of purified PCR product, 1 ml of BigDye Terminator Cycle Sequencing version 2.0 and 3.0 and analysed with an ABI 3700 and 3100 system. Confirmation of mutations Restriction analysis was used to verify the mutations in genomic DNA by PCR-dependent amplification and restriction analysis. The primers used were the same as for the PCR amplification (Table 1) where the mutations resulted in the creation or loss of a particular restriction site. If the mutation did not result in a change, one was created through the use of an oligonucleotide that generated a restriction site by altering one of the 30 nucleotides (Table 2). This altered oligonucleotide along with a primer binding to the reverse strand was used to PCR amplify a fragment of genomic DNA. Digestions were carried out using 10–15 mL of the PCR product with the appropriate restriction enzyme and buffers for 2 h at the approved temperature before undergoing electrophoresis using 8% acrylamide. Northern blot A multiple human tissue northern blot was used for determining the level and size of mRNA species. A probe was generated from a 338 bp cDNA segment amplifying regions 276–613 bp of the full-length cDNA, using the Huh-1 cell line as the source of RNA. Detection of b-actin mRNA was used as control. All procedures followed the protocol supplied by Clontech. DNA or RNA extraction and PCR amplification ACKNOWLEDGEMENTS Genomic DNA was isolated as described previously (26) or using Genomic-tip as described in the manufacturer’s protocol. PCR reactions were performed on genomic DNA with primers designed to amplify exons of the coding sequence and portions of the flanking introns (Table 1). Total cellular RNA was purified from a human liver hybridoma cell line (Huh-7) using Trizol. The total cDNA was amplified using oligo-dT as described by the Omniscript protocol followed by PCR amplification with RNA specific primers (Table 1). The PCR products were purified with the We acknowledge the contributions of Nora Matiaszuk (Royal Victoria Hospital) and Gail Dunbar (Royal Victoria Hospital) for growing cell cultures and David Watkins (Royal Victoria Hospital) for helpful discussions. These studies were supported by grants from the Canadian Institute for Health Research (CIHR), NIH-National Heart and Lung Institute (NHLBI) and March of Dimes. Scholarship support to C.M.D. was provided by the CIHR and Alberta Heritage Foundation for Medical Research. A thank you is extended to Genome Quebec for its support of the sequencing core facilities. Human Molecular Genetics, 2002, Vol. 11, No. 26 REFERENCES 1. Rosenblatt, D.S. and Fenton, W.A. (2001) In Scriver, C.R., Beaudet, A.L., Sly, W.S. and Valle, D. (eds), The Metabolic and Molecular Bases of Inherited Disease. McGraw-Hill, New York, pp. 3897–3933. 2. Dobson, D.M., Wai, T., Leclerc, D., Wilson, A., Wu, X., Dore, C., Hudson, T.J., Rosenblatt, D.S. and Gravel, R.A. (2002) Identification of the gene responsible for the cblA complementation group of vitamin B12responsive methylmalonic acidemia based on analysis of prokaryotic gene arrangements. Proc. Natl. Acad. Sci. USA, in press. 3. Watkins, D. and Rosenblatt, D.S. (1988) Genetic heterogeneity among patients with methylcobalamin deficiency. Definition of two complementation groups, cblE and cblG. J. Clin. Invest., 81, 1690–1694. 4. Li, Y.N., Gulati, S., Baker, P.J., Brody, L.C., Banerjee, R. and Kruger, W.D. (1996) Cloning, mapping and RNA analysis of the human methionine synthase gene. Hum. Mol. Genet., 5, 1851–1858. 5. Leclerc, D., Campeau, E., Goyette, P., Adjalla, C.E., Christensen, B., Ross, M., Eydoux, P., Rosenblatt, D.S., Rozen, R. and Gravel, R.A. (1996) Human methionine synthase: cDNA cloning and identification of mutations in patients of the cblG complementation group of folate/cobalamin disorders. Hum. Mol. Genet., 5, 1867–1874. 6. Chen, L.H., Liu, M.L., Hwang, H.Y., Chen, L.S., Korenberg, J. and Shane, B. (1997) Human methionine synthase. cDNA cloning, gene localization, and expression. J. Biol. Chem., 272, 3628–3634. 7. Leclerc, D., Wilson, A., Dumas, R., Gafuik, C., Song, D., Watkins, D., Heng, H.H., Rommens, J.M., Scherer, S.W., Rosenblatt, D.S. and Gravel, R.A. (1998) Cloning and mapping of a cDNA for methionine synthase reductase, a flavoprotein defective in patients with homocystinuria. Proc. Natl. Acad. Sci. USA, 95, 3059–3064. 8. Gravel, R.A., Mahoney, M.J., Ruddle, F.H. and Rosenberg, L.E. (1975) Genetic complementation in heterokaryons of human fibroblasts defective in cobalamin metabolism. Proc. Natl. Acad. Sci. USA, 72, 3181–3185. 9. Mahoney, M.J., Hart, A.C., Steen, V.D. and Rosenberg L.E. (1975) Methylmalonicacidemia: biochemical heterogeneity in defects of 50 -deoxyadenosylcobalamin synthesis. Proc. Natl. Acad. Sci. USA, 72, 2799–2803. 10. Kerwar, S.S., Spears, C., McAuslan, B. and Weissbach, H. (1971) Studies on vitamin B12 metabolism in HeLa cells. Arch. Biochem. Biophys., 142, 231–237. 11. Fenton, W.A. and Rosenberg, L.E. (1981) The defect in the cbl B class of human methylmalonic acidemia: deficiency of cob(I)alamin adenosyltransferase activity in extracts of cultured fibroblasts. Biochem. Biophys. Res. Commun., 98, 283–289. 12. Battersby, A.R. (1994) New intermediates in the B12 pathway. Ciba Found. Symp., 180, 267–279. 3369 13. Roessner, C.A., Santander, P.J. and Scott, A.I. (2001) Multiple biosynthetic pathways for vitamin B12: variations on a central theme. Vitam. Horm., 61, 267–297. 14. Scott, A.I. (1994) Recent studies of enzymically controlled steps in B12 biosynthesis. Ciba Found. Symp., 180, 285–303. 15. Raux, E., Schubert, H.L. and Warren, M.J. (2000) Biosynthesis of cobalamin (vitamin B12): a bacterial conundrum. Cell. Mol. Life Sci., 57, 1880–1893. 16. Santander, P.J., Roessner, C.A., Stolowich, N.J., Holderman, M.T. and Scott, A.I. (1997) How corrinoids are synthesized without oxygen: nature’s first pathway to vitamin B12. Chem. Biol., 4, 659–666. 17. Roth, J.R., Lawrence, J.G. and Bobik, T.A. (1996) Cobalamin (coenzyme B12): synthesis and biological significance. Annu. Rev. Microbiol., 50, 137–181. 18. Fonseca, M.V. and Escalante-Semerena, J.C. (2001) An in vitro reducing system for the enzymic conversion of cobalamin to adenosylcobalamin. J. Biol. Chem., 276, 32101–32108. 19. Bauer, C.B., Fonseca, M.V., Holden, H.M., Thoden, J.B., Thompson, T.B., Escalante-Semerena, J.C. and Rayment, I. (2001) Three-dimensional structure of ATP:corrinoid adenosyltransferase from Salmonella typhimurium in its free state, complexed with MgATP, or complexed with hydroxycobalamin and MgATP. Biochemistry, 40, 361–374. 20. Kofoid, E., Rappleye, C., Stojiljkovic, I. and Roth, J. (1999) The 17-gene ethanolamine (eut) operon of Salmonella typhimurium encodes five homologues of carboxysome shell proteins. J. Bacteriol., 181, 5317–5329. 21. Johnson, C.L., Pechonick, E., Park, S.D., Havemann, G.D., Leal, N.A. and Bobik, T.A. (2001) Functional genomic, biochemical, and genetic characterization of the Salmonella pduO gene, an ATP:cob(I)alamin adenosyltransferase gene. J. Bacteriol., 183, 1577–1584. 22. Schuler, G.D. (1997) Pieces of the puzzle: expressed sequence tags and the catalog of human genes. J. Mol. Med., 75, 694–698. 23. Fenton, W.A. and Rosenberg, L.E. (1978) Genetic and biochemical analysis of human cobalamin mutants in cell culture. Annu. Rev. Genet., 12, 223–248. 24. Mellman, I., Willard, H.F. and Rosenberg, L.E. (1978) Cobalamin binding and cobalamin-dependent enzyme activity in normal and mutant human fibroblasts. J. Clin. Invest., 62, 952–960. 25. Rosenblatt, D.S., Cooper, B.A., Pottier, A., Lue-Shing, H., Matiaszuk, N. and Grauer, K. (1984) Altered vitamin B12 metabolism in fibroblasts from a patient with megaloblastic anemia and homocystinuria due to a new defect in methionine biosynthesis. J. Clin. Invest., 74, 2149–2156. 26. Wilson, A., Leclerc, D., Rosenblatt, D.S. and Gravel, R.A. (1999) Molecular basis for methionine synthase reductase deficiency in patients belonging to the cblE complementation group of disorders in folate/cobalamin metabolism. Hum. Mol. Genet., 8, 2009–2016.