- Email: [email protected]
Characterization of the human gene encoding α-aminoadipate aminotransferase (AADAT)
Molecular Genetics and Metabolism 76 (2002) 172–180 www.academicpress.com
Characterization of the human gene encoding a-aminoadipate aminotransferase (AADAT)q Denise L.M. Goh,a,b Ankita Patel,c George H. Thomas,a,c Gajja S. Salomons,d Danielle S.M. Schor,d Cornelis Jakobs,d and Michael T. Geraghtya,* a
Department of Pediatrics, McKusick-Nathans Institute of Genetic Medicine, Johns Hopkins University School of Medicine, Blalock 10-08, 600 North Wolfe Street, Baltimore, MD 21287, USA b Department of Paediatrics, National University of Singapore, Singapore c Kennedy Krieger Institute, Baltimore, MD, USA d Department of Clinical Chemistry, Metabolic Unit, VU University Medical Center, Amsterdam, The Netherlands Received 19 February 2002; received in revised form 25 April 2002; accepted 25 April 2002
Abstract In mammals, the conversion of a-aminoadipate to a-ketoadipate by a-aminoadipate aminotransferase (AADAT) is an intermediate step in lysine degradation. A gene encoding for a-aminoadipate aminotransferase and kynurenine aminotransferase activites had been previously identified in the rat (KAT/AadAT). We identified the human gene (AADAT) encoding for AADAT. It has a 2329 bp cDNA, a 1278 bp open-reading frame, and is predicted to encode 425 amino acids with a mitochondrial cleavage signal and a pyridoxal-phosphate binding site. AADAT is 73% and 72% identical to the mouse and rat orthologs, respectively. The genomic structure spans 30 kb and consists of 13 exons. FISH studies localized the gene to 4q32.2. Two transcripts ( 2:9 and 4:7 kb) were identified, with expression highest in liver. Bacterial expression studies confirm that the gene encodes for AADAT activity. The availability of the DNA sequence and enzyme assay will allow further evaluation of individuals suspected to have defects in this enzyme. Ó 2002 Elsevier Science (USA). All rights reserved. Keywords: a-Aminoadipate aminotransferase; Human; Chromosome 4; 2-Aminoadipate-2-oxoglutarate aminotransferase; Pyridoxine; AADAT; KAT; Lysine
1. Introduction a-Aminoadipate is a six-carbon homolog of glutamate. The L -isomer is an intermediate in the metabolism of D L -lysine and L -hydroxylysine as well as a component of the precursor to penicillin and cephalosporin [1–3] The D -isomer is a constituent of some natural b-lactam antibiotics [3]. In humans, lysine is the main source of a-aminoadipate. Lysine, an essential amino acid in humans, is degraded when there is an excess beyond that required for protein synthesis (Fig. 1). The saccharopine pathway is the major route for the catabq Note. Gene symbol AADAT has been approved by HUGO. Sequence data from this article have been deposited with the GenBank Data Libraries under Accession No. AF481738. * Corresponding author. Fax: +410-614-9246. E-mail address: [email protected] (M.T. Geraghty).
olism of L -lysine and occurs in the liver. The pipecolate pathway is used primarily for D -lysine catabolism, though in the brain, it is said to be the only pathway for L -lysine degradation [4,5]. The two pathways converge at a-aminoadipate semialdehyde, which is subsequently converted to a-aminoadipate by a-aminoadipate semialdehyde dehydrogenase. Degradation of a-aminoadipate to a-ketoadipate involves a transamination reaction mediated by a-aminoadipate aminotransferase (AADAT) (EC.2.6.1.39). We had previously identified and characterized two other enzymes involved in lysine degradation viz. a-aminoadipate semialdehyde synthase (AASS) and a-aminoadipate semialdehyde dehydrogenase-phosphopantetheinyl transferase (AASD-PPT) [6,7]. In mammals, a-aminoadipate is converted to a-ketoadipate through a transamination reaction mediated by a-aminoadipate aminotransferase (AADAT). Human
1096-7192/02/$ - see front matter Ó 2002 Elsevier Science (USA). All rights reserved. PII: S 1 0 9 6 - 7 1 9 2 ( 0 2 ) 0 0 0 3 7 - 9
D.L.M. Goh et al. / Molecular Genetics and Metabolism 76 (2002) 172–180
173
Fig. 1. Lysine metabolic pathways.
liver has been shown to have two such enzymes (AADAT-I and AADAT-II) [8]. Bovine kidney has appreciable amount of AADAT activity but no kynurenine aminotransferase (KAT) activity [9]. In the rat, AADAT activity has been found in liver and kidney tissues [10– 16]. A gene (KAT/AadAT also known as KAT2) coding for a soluble rat kidney aminotransferase with both KAT and AADAT activity has been identified [17]. The cDNA contained a single 1275 bp open-reading frame (ORF) encoding for a soluble protein of 425 amino acid residues that appeared to be structurally homologous to aspartate aminotransferase in its pyridoxal 50 -phosphate binding domain. RNA blot analysis of rat tissues revealed a single transcript of approximately 2.1 kb in kidney, liver, and brain. Expression studies confirm that the gene encodes for both KAT and AADAT enzymatic activities as well as activity with tryptophan and aspartate. The rat KAT/AadAT is highly homologous to mouse Kat2 [18]. The full-length cDNA of mouse Kat2 also encodes for 425 amino acids, spans 30 kb, and is composed of 13 exons. It is expressed mainly in kidney and to a lesser amount in liver and brain. In Saccharomyces cerevisiae, a protein with similar function has also been described. ARO8 encodes for a protein known as aromatic aminotransferase I [19,20]. It is involved in phenylalanine and tyrosine biosynthesis and is able to use glutamate, phenylalanine, tyrosine, and tryptophan as amino do-
nors and phenylpyruvate, hydroxyphenylpyruvate, 2oxoglutarate, and pyruvate as amino acceptors. It is also able to use 2-aminoadipate, methionine, and leucine as amino donors and with the corresponding oxoacids as amino acceptors. We report the identification and characterization of the human gene encoding a-aminoadipate aminotransferase (AADAT).
2. Materials and methods 2.1. Cloning of the human AAAT The BLAST algorithm was used to scan human databases of The National Center for Biotechnology Information (NCBI) (http://www.ncbi.nlm.nih.gov/) and the Institute of Genomic Research (TIGR) (http://www. tigr.org/docs/tigr-scripts/nhgi_scripts/tgi_blast.pl?organism ¼ Human) for sequences homologous to the rat KAT/AadAT cDNA (GenBank Acession No. Z50144). A human EST contig containing the putative full-length human AADAT cDNA was assembled. Alignment of the products of human AADAT, mouse Kat2, rat KAT/ AadAT (also known as Kat2), and yeast ARO8 was performed using MacVector Version 6.23. Further analysis for molecular weight, pI, and functional
174
D.L.M. Goh et al. / Molecular Genetics and Metabolism 76 (2002) 172–180
domains in the predicted human protein product was done using the Compute pI/Mw tool, PROSITE and PSORTII programs, respectively (http://www.expasy.ch/ tools/pi_tool.html, http://www.expasy.ch/prosite/ and http://psort.nibb.ac.jp/). The sequence of the ORF of the human AADAT cDNA was confirmed by gene specific amplification, followed by direct sequencing of the amplified DNA. The ORF was amplified from human liver cDNA (Clontech, Palo Alto, CA, USA) by PCR using oligonucleotides containing gene specific sequences and the restriction endonuclease sites SalI and NotI, respectively. (50 -GCCGTCGACCATGAATTACGCACGGT TCATCA CGGC and 50 -CAGGCGGCCGCTCATAA AGATTC T TTTATAAGTTGTGC). The BLAST algorithm was used to search the high throughput genomic sequence (HTGS) database at NCBI for sequences identical to the human AADAT cDNA. A human genomic DNA BAC clone containing the full-length cDNA was identified (Homo sapiens clone RP11-6E9, GenBank Accession No. AC084866). Using the BLAST 2 sequences algorithm, the intron– exon boundaries, 50 UTR, and 30 UTR were identified. 2.2. Chromosomal localization Genomic DNA of clone RP11-25G11 (GenBank Accession No. AC022631) was purified (Qiagen, Valencia, CA) and used in standard fluorescence in situ hybridization to further localize AADAT [21]. 2.3. Transcript analysis and functional characterization of human AAAT Tissue expression analysis was performed using multi-tissue Northern blots (Clontech, Palo Alto, CA, USA). The blots were hybridized with an ½a-32 P labeled probe containing the full-length AADAT ORF, washed according to manufacturer’s instructions, and subjected to autoradiography at )70 °C. 2.4. Functional characterization of human AAAT The ORF PCR product was subcloned into a bacterial expression vector (pMBPT: gift of B.V. Geisbrecht, Johns Hopkins University, Baltimore, MD). The construct was sequenced to verify the nucleotide sequence. The vector pMBPT was derived from the pMAL2 vector (New England BioLabs, MA, USA) by inserting the SalI and NotI restriction sites into the multiple cloning site [22]. Further, the tobacco etch virus protease recognition site (ENLYFQ/G) was inserted 50 and in-frame with the SalI site. The protein was over-expressed in bacteria and AADAT activity was measured using a modification of the procedure used by Schor et al. [23] to study GABA-T activity. Briefly, in a total volume of
200 ll containing complete buffer (50 mmol/L potassium phosphate buffer (pH 8.0), supplemented with 0.25 mmol/L dithiothreitol, 0.05 mmol/L disodium EDTA, and 0.1 mmol/L pyridoxal-5-phosphate), 5 lg of bacterial lysate was incubated for 45 min at 37 °C with 70 nmol a-ketoglutaric acid and 1240 nmol substrate 15N-a aminoadipate. The formation of 15N-glutamic acid was quantified by GC-MS using 2 nmol [2; 3; 3; 4; 4-2 H5 ]glutamic acid as internal standard.
3. Results 3.1. Cloning of human AADAT The full-length AADAT cDNA was derived using the rat cDNA (GenBank Accession No. Z50144) and a query in a blast search (NCBI). It is 2329 bp long and predicted to have an ORF of 1278 bp (GenBank Accession No. AF481738) (Fig. 2). The sequence of the ORF was confirmed by gene specific amplification using PCR with a liver cDNA library (Clontech) as template, followed by direct sequencing of the amplified DNA. The predicted initiation codon (ATG at nt +1) corresponds to the start site determined for the rat and mouse homologues. There was no further ATG found upstream of this predicted initiation codon and an inframe termination codon (TGA) was found upstream at nt )93. The translation termination codon (TGA) was found at nt +1276, in the same position as several human ESTs as well as in the rat and mouse homologs (GenBank Accession Nos. Z50144, AF072376). A polyadenylation signal (AATAAA) was found at nt +1960. The 50 UTR and 30 UTR of the gene were derived using two methods: (i) Assembly of a contig consisting of ESTs showed that multiple clones had similar 30 UTR and 50 UTRs. (GenBank Accession Nos. AI798080, AA404282, and BF221663, AA488185); (ii) In addition, the human cDNA is homologous to the rat and mice orthologs. These findings suggest that we have identified the 50 UTR and 30 UTR of the human gene. The cDNA is predicted to encode 425 amino acids with a calculated molecular weight of 50.8 kDa and an isoelectric point of 5.03. The protein is also predicted to have a mitochondrial leader cleavage signal (SRG — PK) at amino acid position 27 and the expected pyridoxalphosphate binding site at amino acid position 260. The human protein is 73% identical to the mouse KAT-II protein, 72% identical to the rat KAT-II protein, and 29% identical to the yeast counterpart (Fig. 3). Alignment analysis also showed that the pyridoxal phosphatebinding site was conserved across all four species. The human AADAT spans 30 kb and consists of 13 exons divided by 12 introns (Fig. 4). The exons range from 62 to 750 bp and all intron–exon boundaries conformed to the 50 AG to 30 GT rule [24].
D.L.M. Goh et al. / Molecular Genetics and Metabolism 76 (2002) 172–180
175
Fig. 2. Nucleotide and predicted amino acid sequences of human AADAT. Nucleotides are numbered starting with +1 at the predicted initiation codon (ATG). The deduced amino acid sequence is shown in single-letter code below the nucleotide sequence. The predicted mitochondrial cleavage signal is boxed. The residues involved in cofactor binding are shaded and the lysine reisdue information in the formation of an aldimine bond with pyridoxal 50 -phosphate (PLP) is indicated in bold [25].
3.2. Chromosomal localization FISH studies done confirmed the localization of the gene to chromosome 4q32.2 (Fig. 5). This is consistent
with our BLAST results, which also mapped the gene to chromosome 4. In addition, this region is syntenic to the mouse chromosome 8, which contains the locus of the mouse homolog.
176
D.L.M. Goh et al. / Molecular Genetics and Metabolism 76 (2002) 172–180
Fig. 3. Sequence alignment of human AADAT (Hs AADAT), mouse KAT2 (Mm Kat2), rat KAT2 (Rn Kat2), and S. pombe aromatic aminotransferase I (ARO8) proteins. Residues that are identical (dark gray) or similar (light gray) are shown. The lysine residue that forms the aldimine bond with PLP is indicated by a star. The residues involved in cofactor binding are indicated by arrowheads [22].
D.L.M. Goh et al. / Molecular Genetics and Metabolism 76 (2002) 172–180
177
Fig. 4. (A) Genomic structure of AADAT. Vertical lines and numbers indicate exons. Horizontal lines between exons represent introns. The relative sizes of each exon and intron are drawn to scale. (B) Intron–exon boundaries of AADAT genomic DNA. Exon and intron sequences are represented by upper and lower cases, respectively. The sizes of the exons were determined by sequencing.
Fig. 5. FISH. Metaphase cell showing FISH of BAC clone RP11-25G11 (red signals).
178
D.L.M. Goh et al. / Molecular Genetics and Metabolism 76 (2002) 172–180
Fig. 6. Northern blot analysis. Mouse multiple tissue Northern blots were hybridized with a ½a-32 P labeled probe (full-length ORF). Numbers on the left represent standard molecular weight markers.
Fig. 7. AADAT enzyme assay. pMBP ¼ vector alone; pMPB + AADAT ¼ vector with a full length ORF insert.
3.3. Transcript analysis and functional characterization of human AAAT Northern blot analysis performed using a probe containing the ORF revealed the existence of two species of mRNA (Fig. 6). There was a major signal at 2:9 kb and a minor signal at 4:7 kb. Expression was highest in liver but was also seen at much lower levels in heart, brain, kidney, pancreas, prostate, testis, and ovary. Bacterial expression studies confirm that the gene encodes aminoadipate aminotransferase activity (Fig. 7).
4. Discussion The degradation of a-aminoadipate to a-ketoadipate involves a transamination reaction mediated by AADAT.
We report here the identification and characterization of a gene that encodes for the human AADAT enzyme. AADAT is localized to chromosome 4q32.2, a region that is syntenic to the mouse chromosomal locus of the mouse homolog, Kat2 [18]. Both genes span 30 kb and contain 13 exons of similar sizes. The genomic organization of the rat Kat2 is currently not known. The full length AADAT cDNA is very similar to the rat and mouse homologs [17,18]. All three have similar translational initiation and termination sites and are predicted to encode for 425 amino acids. The predicted human AADAT protein is 73% identical to the mouse KAT II enzyme, 72% identical to the rat KAT II enzyme, and 29% identical to the yeast aromatic aminotransferase I enzyme. All four proteins contain a pyridoxal phosphate-binding site, the sequence of which is highly conserved across all four species [25]. This is consistent with the knowledge that most transamination reactions are pyridoxine (B6) dependent. The human AADAT enzyme is also predicted to contain a mitochondrial cleavage signal, suggesting that AADAT is a mitochondrial protein. AADAT activity has been found in the mitochondrial fraction of human, cow, and rat extracts [8–10,13,16]. In addition, the downstream conversion of a-ketoadipate to glutaryl-CoA occurs in the mitochondria. Thus, it is possible that AADAT encodes for a protein found in the mitochondria. Further subcellular localization studies are needed to prove that AADAT is a mitochondrial protein. We suspect that AADAT codes for the protein previously purified and identified as AADAT-II [8]. AADAT-II is found in human liver and has similar molecular weight and pI as
D.L.M. Goh et al. / Molecular Genetics and Metabolism 76 (2002) 172–180
the predicted protein product of AADAT. It has a low Km value for aminoadipate (0.25 mmol/L), while also showing activity towards tryptophan and kynurenine. The rat KAT II enzyme, in addition to its activity with aminoadipate, could also use kynurenine (Km 0.95 mmol/L), tryptophan, and aspartate as substrates [17]. This suggests that human AADAT may also be capable of using these substrates and experiments are underway to test this hypothesis. Northern blot analysis of human AADAT revealed the presence of two transcripts ( 2:9 and 4:7 kb). Northern blot analysis of mouse Kat2 showed the presence of two transcripts (2.0 and 1.8 kb) as well as an additional 5.0 kb transcript [18]. The 5.0 kb transcript is thought to be an alternatively spliced transcript or part of an overlapping gene. The rat, on the other hand, has only a single transcript of approximately 2.1 kb [17]. The 2:9 kb human AADAT transcript is consistent with the size of the predicted cDNA. The origin of the larger 4:7 kb transcript is unknown. It may thus represent a partially unspliced transcript, an alternatively polyadenylated transcript, an alternatively spliced transcript or part of an overlapping gene. The relative abundance of the two human transcripts was also different in different tissues. The 2:9 kb transcript was predominant in liver, kidney, and testis while both transcripts appear to have similar abundance in the heart, brain, pancreas, and prostate. Expression in general was highest in the liver for humans and in the kidney for rat and mouse [17,18]. AADAT was demonstrated to encode for an aminotransferase capable of converting a-aminoadipic acid to a-ketoadipate, a function also demonstrated in rat KAT II and bovine AADAT [9,17,26]. Yeast deletion strains (aro8 and aro9) have been previously described [19]. We, however, could not reproduce the yeast phenotype identified by these authors in strains obtained from ResGen (Huntsville, AL, USA). The lack of observed phenotype in these knock-outs is perhaps the result of differences in strain background and the extra amino acid auxotrophic requirements which provide an alternative nitrogen source other than lysine. Hence, complementation studies using the human ortholog were not done. Deficiency of AADAT is expected to produce a-aminoadipic aciduria, with possible accumulation of other lysine and tryptophan catabolites. a-Aminoadipic acid is a specific gliotoxic agent both in vivo and in vitro [27– 29]. This suggests that individuals with AAAT deficiency may present with neurological impairment, though the precise phenotype is hard to predict. The majority of reported cases of a-aminoadipic aciduria are associated with a-ketoadipic aciduria. These cases are likely to represent defects in a-ketoadipic dehydrogenase, the subsequent enzyme to AADAT in lysine and tryptophan catabolism. Vigabatrin may also cause a-aminoadipic aciduria [30]. To our knowledge, there are only two reported cases of isolated a-aminoadipic aciduria. One
179
case involved a mentally retarded girl with dysmorphism and persistent fetal hemoglobinemia [31]. A normal degradation of D L -aminoadipate in fibroblasts of this patient makes a primary defect in AADAT unlikely. Candito et al. [32] described the second case, a male child with a-aminoadipic aciduria, defect in platelet aggregation and antenatal cerebral hemorrhage. No enzymic studies were done, hence, a defect in AADAT is not excluded. We hypothesized that a deficiency of a-aminoadipic semialdehyde dehydrogenase (AASD), a-aminoadipic semialdehyde dehydrogenase-phosphopantetheinyl transferase (AASS-PPT) or AADAT might lead to elevation of pipecolic acid. Plecko et al. [33] found significant elevation of pipecolic acid in the plasma and cerebrospinal fluid of two patients with B6-responsive seizures. We sequenced the cDNA of four patients with isolated pipecolic acidemia (2 with B6 responsive seizures and two with other neurological phenotypes) but were unable to find any mutations in either AADAT or AAsD-PPT. In conclusion, we report the identification and characterization of the gene that encodes for the human AADAT enzyme. To date, we have not identified any individuals with a defect in this enzyme. However, the availability of the AADAT DNA sequence and a specific enzyme assay will allow us to further evaluate individuals suspected to have defects in this enzyme.
Acknowledgments We thank Silvy J.M. van Dooren and Erwin E.W. Jansenfor their expert laboratory support.
References [1] V.W. Rodwell, in: D.M. Greenberg (Ed.), Metabolic Pathways, Academic Press, New York, 1967, pp. 223–223. [2] R.A. Hiles, K. Triebwasser, C.K. Triebwasser, L.M. Henderson, The degradation of hydroxy-L -lysine in liver via its phosphate ester, Biochem. Biophys. Res. Commun. 41 (1970) 662–668. [3] S.W. Queener, N. Neuss, in: R.B Morin, M. Gorman (Eds.), The Chemistry and Biology of b-Lactam Antibiotics, Academic Press, New York, 1982, p. 1. [4] Y.F. Change, Lysine metabolism in rat brain: the pipecolic acidforming pathway, J. Neurochem. 30 (1978) 347. [5] E. Giacobini, Y. Nomura, T. Schmidt-Glenewinkel, Pipecolic acid: origin, biosynthesis and metabolism in brain, Cell. Mol. Biol. 26 (1980) 135. [6] K.A. Sacksteder, B.J. Biery, J.C. Morrell, B.K. Goodman, B.V. Geisbrecht, R.P. Cox, S.J. Gould, M.T. Geraghty, Identification of the a-aminoadipic semialdehyde synthase gene, which is defective in familial hyperlysinemia, Am. J. Hum. Genet. 66 (2000) 1736–1743. [7] V. Praphanphoj, K.A. Sacksteder, S.J. Gould, G.H. Thomas, M.T. Geraghty, Identification of the a-aminoadipic semialdehyde dehydrogenase-phosphopantetheinyl transferase gene, the human ortholog of the yeast LYS5 gene, Mol. Genet. Metab. 72 (2001) 336–342.
180
D.L.M. Goh et al. / Molecular Genetics and Metabolism 76 (2002) 172–180
[8] E. Okuno, M. Tsujimoto, M. Nakamura, R. Kido, 2-Aminoadipate-2-oxoglutarate aminotransferase isoenzymes in human liver: a plausible physiological role in lysine and tryptophan metabolism, Enzyme Protein 47 (1993) 136–148. [9] D.R. Deshmukh, S.M. Mungre, Purification and properties of 2aminoadipate: 2-oxoglutarate aminotransferase from bovine kidney, Biochem. J. 261 (1989) 761–768. [10] Y. Nakatani, M. Fujioka, K. Higashino, a-Aminoadipate aminotransferase of rat liver mitochondria, Biochim. Biophys. Acta 198 (1970) 219–228. [11] M.C. Tobes, M. Mason, L -Kynurenine aminotransferase and L -aaminoadipate aminotransferase. I. Evidence for identity, Biochem. Biophys. Res. Commun. 62 (1975) 390–397. [12] M.C. Tobes, M. Mason, a-Aminoadipate aminotransferase and kynurenine aminotransferase. Purification, characterization, and further evidence for identity, J. Biol. Chem. 252 (1977) 4591–4599. [13] F. Takeuchi, H. Otsuka, Y. Shibata, Purification, characterization and identification of rat liver mitochondrial kynurenine aminotransferase with alpha-aminoadipate amniotransferase, Biochim. Biophys. Acta 743 (1983) 323–330. [14] M.R. Mawal, D.R. Deshmukh, Alpha-aminoadipate aminotransferase and kynurenine aminotransferase activities from rat kidney, J. Biol. Chem. 266 (1991) 2573–2575. [15] M.R. Mawal, D.R. Deshmukh, Purification and properties of aaminoadipate aminotransferase from rat kidney, Prep. Biochem. 21 (1991) 63–73. [16] M.R. Mawal, A. Mukhopadhyay, D.R. Deshmukh, Purification and properties of alpha-aminoadipate aminotransferase from rat liver and kidney mitochondria, Prep. Biochem. 21 (1991) 151–162. [17] R. Buchli, D. Alberatu-Giani, P. Malherbe, C. Kohler, C. Broger§, A. Cesura, Cloning and functional expression of a soluble form of kynurenine/alpha-aminoadipate aminotransferase from rat kidney, J. Biol. Chem. 270 (1995) 29330–29335. [18] P. Yu, D.M. Mosbrook, D.A. Tagle, Genomic organization and expression analysis of mouse kynurenine aminotransferase II, a possible factor in the pathophysiology of Huntington’s disease, Mamm. Genome 10 (1999) 845–852. [19] I. Iraqui, S. Vissers, M. Cartiaux, A. Urrestarazu, Characterisation of Saccharimyces cerevisiae ARO8 and ARO9 genes encoding aromatic aminotransferases I and II reveals a new aminotransferase subfamily, Mol. Gen. Genet. 257 (1998) 238–248. [20] A. Urrestarazu, S. Vissers, I. Iraqui, M. Grenson, Phenylalanineand tyrosine-auxotrophic mutants of Saccharomyces cerevisias impaired in transamination, Mol. Gen. Genet. 257 (1998) 230– 237. [21] D. Pinkle, J. Landegent, C. Collins, J. Fuscoe, R. Segraves, J. Lucas, J. Gray, Fluorescence in situ hybridization with human
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33]
chromosome-specific libraries: detection of trisomy 21 and translocations of chromosome 4, Proc. Natl. Acad. Sci. USA 85 (1988) 9138–9142. B.V. Geisbrecht, D. Zhu, K. Schulz, K. Nau, J.C. Morrell, M. Geraghty, H. Schulz, R. Erdmann, S.J. Gould, Molecular characterization of Saccharomyces cerevisiae D3 ; D2 -Enoyl-CoA isomerase, J. Biol. Chem. 273 (1998) 33184–33191. D.S.M. Schor, E.A. Struys, B.M. Hogema, K.M. Gibson, C. Jakobs, Development of a stable-isotope dilution assay for gamma-aminobutyric acid (GABA) transaminase in isolated leukocytes and evidence that GABA and beta-alanine transaminases are identical, Clin. Chem. 47 (2001) 525–531. M.B. Shapiro, P. Senapathy, RNA spice junctions of different classes of eukaryotes: sequence statistics and functional implications in gene expression, Nucleic Acids Res. 15 (1987) 7155– 7174. J. Jager, M. Moser, U. Sauder, J.N. Jansonius, Crystal structures of Escherichia coli aspartate aminotransferase in two conformations. Comparision of an unliganded open and two liganded closed forms, J. Mol. Biol. 239 (1994) 285–305. P. Guidetti, E. Okuno, R. Schwarcz, Characterization of rat brain kynurenine aminotransferase I and II, J. Neurosci. Res. 50 (1997) 457–465. J.W. Olney, O.L. Ho, V. Rhee, Cytotoxic effects of acidic and sulphur containing amino acids on the infant mouse central nervous system, Exp. Brain Res. 14 (1971) 61–76. J.W. Olney, T. De Gubareff, J.F. Collins, Stereospecificity of the gliotoxic and anti-neurotoxic actions of alpha-amino-adipate, Neurosci. Lett. 19 (1980) 277–282. D.S. Casper, R.L. Trelstad, L. Reif-Lehrer, Glutamate-induced cellular injury in isolated chick embryo retina: M€ uller cell localization of initial effects, J. Comp. Neurol. 209 (1982) 79–90. C. Vallat, F. Rivier, H. Bellet, B. Magnan de Bornier, H. Mion, B. Echenne, Treatment with vigabatrin may mimic a-aminoadipic aciduria, Epilepsia 37 (1996) 803–805. A.J. Manders, C.G. von Oostrom, J.M. Trijbels, F.J. Rutten, W.J. Kleijer, a-Aminoadipic aciduria and persistence of fetal haemoglobin in an oligophrenic child, Eur. J. Pediatr. 136 (1981) 51–55. M. Candito, C. Richelme, P. Parvy, C. Dageville, A. Appert, S. Bekri, D. Rabier, P. Chambon, R. Mariani, P. Kamoun, Abnormal a-aminoadipic acid excretion in a newborn with a defect in platelet aggregation and antenatal cerebral haemorrhage, J. Inher. Metab. Dis. 18 (1995) 56–60. B. Plecko, S. St€ ockler-Ipsiroglu, E. Paschke, W. Erwa, E. Struys, C. Jakobs, Pipecolic acid elevation in plasma and cerebrospinal fluid of two patients with pyridoxine-dependent epilepsy, Ann. Neurol. 48 (2000) 121–125.
Characterization of the human gene encoding a-aminoadipate aminotransferase (AADAT)q Denise L.M. Goh,a,b Ankita Patel,c George H. Thomas,a,c Gajja S. Salomons,d Danielle S.M. Schor,d Cornelis Jakobs,d and Michael T. Geraghtya,* a
Department of Pediatrics, McKusick-Nathans Institute of Genetic Medicine, Johns Hopkins University School of Medicine, Blalock 10-08, 600 North Wolfe Street, Baltimore, MD 21287, USA b Department of Paediatrics, National University of Singapore, Singapore c Kennedy Krieger Institute, Baltimore, MD, USA d Department of Clinical Chemistry, Metabolic Unit, VU University Medical Center, Amsterdam, The Netherlands Received 19 February 2002; received in revised form 25 April 2002; accepted 25 April 2002
Abstract In mammals, the conversion of a-aminoadipate to a-ketoadipate by a-aminoadipate aminotransferase (AADAT) is an intermediate step in lysine degradation. A gene encoding for a-aminoadipate aminotransferase and kynurenine aminotransferase activites had been previously identified in the rat (KAT/AadAT). We identified the human gene (AADAT) encoding for AADAT. It has a 2329 bp cDNA, a 1278 bp open-reading frame, and is predicted to encode 425 amino acids with a mitochondrial cleavage signal and a pyridoxal-phosphate binding site. AADAT is 73% and 72% identical to the mouse and rat orthologs, respectively. The genomic structure spans 30 kb and consists of 13 exons. FISH studies localized the gene to 4q32.2. Two transcripts ( 2:9 and 4:7 kb) were identified, with expression highest in liver. Bacterial expression studies confirm that the gene encodes for AADAT activity. The availability of the DNA sequence and enzyme assay will allow further evaluation of individuals suspected to have defects in this enzyme. Ó 2002 Elsevier Science (USA). All rights reserved. Keywords: a-Aminoadipate aminotransferase; Human; Chromosome 4; 2-Aminoadipate-2-oxoglutarate aminotransferase; Pyridoxine; AADAT; KAT; Lysine
1. Introduction a-Aminoadipate is a six-carbon homolog of glutamate. The L -isomer is an intermediate in the metabolism of D L -lysine and L -hydroxylysine as well as a component of the precursor to penicillin and cephalosporin [1–3] The D -isomer is a constituent of some natural b-lactam antibiotics [3]. In humans, lysine is the main source of a-aminoadipate. Lysine, an essential amino acid in humans, is degraded when there is an excess beyond that required for protein synthesis (Fig. 1). The saccharopine pathway is the major route for the catabq Note. Gene symbol AADAT has been approved by HUGO. Sequence data from this article have been deposited with the GenBank Data Libraries under Accession No. AF481738. * Corresponding author. Fax: +410-614-9246. E-mail address: [email protected] (M.T. Geraghty).
olism of L -lysine and occurs in the liver. The pipecolate pathway is used primarily for D -lysine catabolism, though in the brain, it is said to be the only pathway for L -lysine degradation [4,5]. The two pathways converge at a-aminoadipate semialdehyde, which is subsequently converted to a-aminoadipate by a-aminoadipate semialdehyde dehydrogenase. Degradation of a-aminoadipate to a-ketoadipate involves a transamination reaction mediated by a-aminoadipate aminotransferase (AADAT) (EC.2.6.1.39). We had previously identified and characterized two other enzymes involved in lysine degradation viz. a-aminoadipate semialdehyde synthase (AASS) and a-aminoadipate semialdehyde dehydrogenase-phosphopantetheinyl transferase (AASD-PPT) [6,7]. In mammals, a-aminoadipate is converted to a-ketoadipate through a transamination reaction mediated by a-aminoadipate aminotransferase (AADAT). Human
1096-7192/02/$ - see front matter Ó 2002 Elsevier Science (USA). All rights reserved. PII: S 1 0 9 6 - 7 1 9 2 ( 0 2 ) 0 0 0 3 7 - 9
D.L.M. Goh et al. / Molecular Genetics and Metabolism 76 (2002) 172–180
173
Fig. 1. Lysine metabolic pathways.
liver has been shown to have two such enzymes (AADAT-I and AADAT-II) [8]. Bovine kidney has appreciable amount of AADAT activity but no kynurenine aminotransferase (KAT) activity [9]. In the rat, AADAT activity has been found in liver and kidney tissues [10– 16]. A gene (KAT/AadAT also known as KAT2) coding for a soluble rat kidney aminotransferase with both KAT and AADAT activity has been identified [17]. The cDNA contained a single 1275 bp open-reading frame (ORF) encoding for a soluble protein of 425 amino acid residues that appeared to be structurally homologous to aspartate aminotransferase in its pyridoxal 50 -phosphate binding domain. RNA blot analysis of rat tissues revealed a single transcript of approximately 2.1 kb in kidney, liver, and brain. Expression studies confirm that the gene encodes for both KAT and AADAT enzymatic activities as well as activity with tryptophan and aspartate. The rat KAT/AadAT is highly homologous to mouse Kat2 [18]. The full-length cDNA of mouse Kat2 also encodes for 425 amino acids, spans 30 kb, and is composed of 13 exons. It is expressed mainly in kidney and to a lesser amount in liver and brain. In Saccharomyces cerevisiae, a protein with similar function has also been described. ARO8 encodes for a protein known as aromatic aminotransferase I [19,20]. It is involved in phenylalanine and tyrosine biosynthesis and is able to use glutamate, phenylalanine, tyrosine, and tryptophan as amino do-
nors and phenylpyruvate, hydroxyphenylpyruvate, 2oxoglutarate, and pyruvate as amino acceptors. It is also able to use 2-aminoadipate, methionine, and leucine as amino donors and with the corresponding oxoacids as amino acceptors. We report the identification and characterization of the human gene encoding a-aminoadipate aminotransferase (AADAT).
2. Materials and methods 2.1. Cloning of the human AAAT The BLAST algorithm was used to scan human databases of The National Center for Biotechnology Information (NCBI) (http://www.ncbi.nlm.nih.gov/) and the Institute of Genomic Research (TIGR) (http://www. tigr.org/docs/tigr-scripts/nhgi_scripts/tgi_blast.pl?organism ¼ Human) for sequences homologous to the rat KAT/AadAT cDNA (GenBank Acession No. Z50144). A human EST contig containing the putative full-length human AADAT cDNA was assembled. Alignment of the products of human AADAT, mouse Kat2, rat KAT/ AadAT (also known as Kat2), and yeast ARO8 was performed using MacVector Version 6.23. Further analysis for molecular weight, pI, and functional
174
D.L.M. Goh et al. / Molecular Genetics and Metabolism 76 (2002) 172–180
domains in the predicted human protein product was done using the Compute pI/Mw tool, PROSITE and PSORTII programs, respectively (http://www.expasy.ch/ tools/pi_tool.html, http://www.expasy.ch/prosite/ and http://psort.nibb.ac.jp/). The sequence of the ORF of the human AADAT cDNA was confirmed by gene specific amplification, followed by direct sequencing of the amplified DNA. The ORF was amplified from human liver cDNA (Clontech, Palo Alto, CA, USA) by PCR using oligonucleotides containing gene specific sequences and the restriction endonuclease sites SalI and NotI, respectively. (50 -GCCGTCGACCATGAATTACGCACGGT TCATCA CGGC and 50 -CAGGCGGCCGCTCATAA AGATTC T TTTATAAGTTGTGC). The BLAST algorithm was used to search the high throughput genomic sequence (HTGS) database at NCBI for sequences identical to the human AADAT cDNA. A human genomic DNA BAC clone containing the full-length cDNA was identified (Homo sapiens clone RP11-6E9, GenBank Accession No. AC084866). Using the BLAST 2 sequences algorithm, the intron– exon boundaries, 50 UTR, and 30 UTR were identified. 2.2. Chromosomal localization Genomic DNA of clone RP11-25G11 (GenBank Accession No. AC022631) was purified (Qiagen, Valencia, CA) and used in standard fluorescence in situ hybridization to further localize AADAT [21]. 2.3. Transcript analysis and functional characterization of human AAAT Tissue expression analysis was performed using multi-tissue Northern blots (Clontech, Palo Alto, CA, USA). The blots were hybridized with an ½a-32 P labeled probe containing the full-length AADAT ORF, washed according to manufacturer’s instructions, and subjected to autoradiography at )70 °C. 2.4. Functional characterization of human AAAT The ORF PCR product was subcloned into a bacterial expression vector (pMBPT: gift of B.V. Geisbrecht, Johns Hopkins University, Baltimore, MD). The construct was sequenced to verify the nucleotide sequence. The vector pMBPT was derived from the pMAL2 vector (New England BioLabs, MA, USA) by inserting the SalI and NotI restriction sites into the multiple cloning site [22]. Further, the tobacco etch virus protease recognition site (ENLYFQ/G) was inserted 50 and in-frame with the SalI site. The protein was over-expressed in bacteria and AADAT activity was measured using a modification of the procedure used by Schor et al. [23] to study GABA-T activity. Briefly, in a total volume of
200 ll containing complete buffer (50 mmol/L potassium phosphate buffer (pH 8.0), supplemented with 0.25 mmol/L dithiothreitol, 0.05 mmol/L disodium EDTA, and 0.1 mmol/L pyridoxal-5-phosphate), 5 lg of bacterial lysate was incubated for 45 min at 37 °C with 70 nmol a-ketoglutaric acid and 1240 nmol substrate 15N-a aminoadipate. The formation of 15N-glutamic acid was quantified by GC-MS using 2 nmol [2; 3; 3; 4; 4-2 H5 ]glutamic acid as internal standard.
3. Results 3.1. Cloning of human AADAT The full-length AADAT cDNA was derived using the rat cDNA (GenBank Accession No. Z50144) and a query in a blast search (NCBI). It is 2329 bp long and predicted to have an ORF of 1278 bp (GenBank Accession No. AF481738) (Fig. 2). The sequence of the ORF was confirmed by gene specific amplification using PCR with a liver cDNA library (Clontech) as template, followed by direct sequencing of the amplified DNA. The predicted initiation codon (ATG at nt +1) corresponds to the start site determined for the rat and mouse homologues. There was no further ATG found upstream of this predicted initiation codon and an inframe termination codon (TGA) was found upstream at nt )93. The translation termination codon (TGA) was found at nt +1276, in the same position as several human ESTs as well as in the rat and mouse homologs (GenBank Accession Nos. Z50144, AF072376). A polyadenylation signal (AATAAA) was found at nt +1960. The 50 UTR and 30 UTR of the gene were derived using two methods: (i) Assembly of a contig consisting of ESTs showed that multiple clones had similar 30 UTR and 50 UTRs. (GenBank Accession Nos. AI798080, AA404282, and BF221663, AA488185); (ii) In addition, the human cDNA is homologous to the rat and mice orthologs. These findings suggest that we have identified the 50 UTR and 30 UTR of the human gene. The cDNA is predicted to encode 425 amino acids with a calculated molecular weight of 50.8 kDa and an isoelectric point of 5.03. The protein is also predicted to have a mitochondrial leader cleavage signal (SRG — PK) at amino acid position 27 and the expected pyridoxalphosphate binding site at amino acid position 260. The human protein is 73% identical to the mouse KAT-II protein, 72% identical to the rat KAT-II protein, and 29% identical to the yeast counterpart (Fig. 3). Alignment analysis also showed that the pyridoxal phosphatebinding site was conserved across all four species. The human AADAT spans 30 kb and consists of 13 exons divided by 12 introns (Fig. 4). The exons range from 62 to 750 bp and all intron–exon boundaries conformed to the 50 AG to 30 GT rule [24].
D.L.M. Goh et al. / Molecular Genetics and Metabolism 76 (2002) 172–180
175
Fig. 2. Nucleotide and predicted amino acid sequences of human AADAT. Nucleotides are numbered starting with +1 at the predicted initiation codon (ATG). The deduced amino acid sequence is shown in single-letter code below the nucleotide sequence. The predicted mitochondrial cleavage signal is boxed. The residues involved in cofactor binding are shaded and the lysine reisdue information in the formation of an aldimine bond with pyridoxal 50 -phosphate (PLP) is indicated in bold [25].
3.2. Chromosomal localization FISH studies done confirmed the localization of the gene to chromosome 4q32.2 (Fig. 5). This is consistent
with our BLAST results, which also mapped the gene to chromosome 4. In addition, this region is syntenic to the mouse chromosome 8, which contains the locus of the mouse homolog.
176
D.L.M. Goh et al. / Molecular Genetics and Metabolism 76 (2002) 172–180
Fig. 3. Sequence alignment of human AADAT (Hs AADAT), mouse KAT2 (Mm Kat2), rat KAT2 (Rn Kat2), and S. pombe aromatic aminotransferase I (ARO8) proteins. Residues that are identical (dark gray) or similar (light gray) are shown. The lysine residue that forms the aldimine bond with PLP is indicated by a star. The residues involved in cofactor binding are indicated by arrowheads [22].
D.L.M. Goh et al. / Molecular Genetics and Metabolism 76 (2002) 172–180
177
Fig. 4. (A) Genomic structure of AADAT. Vertical lines and numbers indicate exons. Horizontal lines between exons represent introns. The relative sizes of each exon and intron are drawn to scale. (B) Intron–exon boundaries of AADAT genomic DNA. Exon and intron sequences are represented by upper and lower cases, respectively. The sizes of the exons were determined by sequencing.
Fig. 5. FISH. Metaphase cell showing FISH of BAC clone RP11-25G11 (red signals).
178
D.L.M. Goh et al. / Molecular Genetics and Metabolism 76 (2002) 172–180
Fig. 6. Northern blot analysis. Mouse multiple tissue Northern blots were hybridized with a ½a-32 P labeled probe (full-length ORF). Numbers on the left represent standard molecular weight markers.
Fig. 7. AADAT enzyme assay. pMBP ¼ vector alone; pMPB + AADAT ¼ vector with a full length ORF insert.
3.3. Transcript analysis and functional characterization of human AAAT Northern blot analysis performed using a probe containing the ORF revealed the existence of two species of mRNA (Fig. 6). There was a major signal at 2:9 kb and a minor signal at 4:7 kb. Expression was highest in liver but was also seen at much lower levels in heart, brain, kidney, pancreas, prostate, testis, and ovary. Bacterial expression studies confirm that the gene encodes aminoadipate aminotransferase activity (Fig. 7).
4. Discussion The degradation of a-aminoadipate to a-ketoadipate involves a transamination reaction mediated by AADAT.
We report here the identification and characterization of a gene that encodes for the human AADAT enzyme. AADAT is localized to chromosome 4q32.2, a region that is syntenic to the mouse chromosomal locus of the mouse homolog, Kat2 [18]. Both genes span 30 kb and contain 13 exons of similar sizes. The genomic organization of the rat Kat2 is currently not known. The full length AADAT cDNA is very similar to the rat and mouse homologs [17,18]. All three have similar translational initiation and termination sites and are predicted to encode for 425 amino acids. The predicted human AADAT protein is 73% identical to the mouse KAT II enzyme, 72% identical to the rat KAT II enzyme, and 29% identical to the yeast aromatic aminotransferase I enzyme. All four proteins contain a pyridoxal phosphate-binding site, the sequence of which is highly conserved across all four species [25]. This is consistent with the knowledge that most transamination reactions are pyridoxine (B6) dependent. The human AADAT enzyme is also predicted to contain a mitochondrial cleavage signal, suggesting that AADAT is a mitochondrial protein. AADAT activity has been found in the mitochondrial fraction of human, cow, and rat extracts [8–10,13,16]. In addition, the downstream conversion of a-ketoadipate to glutaryl-CoA occurs in the mitochondria. Thus, it is possible that AADAT encodes for a protein found in the mitochondria. Further subcellular localization studies are needed to prove that AADAT is a mitochondrial protein. We suspect that AADAT codes for the protein previously purified and identified as AADAT-II [8]. AADAT-II is found in human liver and has similar molecular weight and pI as
D.L.M. Goh et al. / Molecular Genetics and Metabolism 76 (2002) 172–180
the predicted protein product of AADAT. It has a low Km value for aminoadipate (0.25 mmol/L), while also showing activity towards tryptophan and kynurenine. The rat KAT II enzyme, in addition to its activity with aminoadipate, could also use kynurenine (Km 0.95 mmol/L), tryptophan, and aspartate as substrates [17]. This suggests that human AADAT may also be capable of using these substrates and experiments are underway to test this hypothesis. Northern blot analysis of human AADAT revealed the presence of two transcripts ( 2:9 and 4:7 kb). Northern blot analysis of mouse Kat2 showed the presence of two transcripts (2.0 and 1.8 kb) as well as an additional 5.0 kb transcript [18]. The 5.0 kb transcript is thought to be an alternatively spliced transcript or part of an overlapping gene. The rat, on the other hand, has only a single transcript of approximately 2.1 kb [17]. The 2:9 kb human AADAT transcript is consistent with the size of the predicted cDNA. The origin of the larger 4:7 kb transcript is unknown. It may thus represent a partially unspliced transcript, an alternatively polyadenylated transcript, an alternatively spliced transcript or part of an overlapping gene. The relative abundance of the two human transcripts was also different in different tissues. The 2:9 kb transcript was predominant in liver, kidney, and testis while both transcripts appear to have similar abundance in the heart, brain, pancreas, and prostate. Expression in general was highest in the liver for humans and in the kidney for rat and mouse [17,18]. AADAT was demonstrated to encode for an aminotransferase capable of converting a-aminoadipic acid to a-ketoadipate, a function also demonstrated in rat KAT II and bovine AADAT [9,17,26]. Yeast deletion strains (aro8 and aro9) have been previously described [19]. We, however, could not reproduce the yeast phenotype identified by these authors in strains obtained from ResGen (Huntsville, AL, USA). The lack of observed phenotype in these knock-outs is perhaps the result of differences in strain background and the extra amino acid auxotrophic requirements which provide an alternative nitrogen source other than lysine. Hence, complementation studies using the human ortholog were not done. Deficiency of AADAT is expected to produce a-aminoadipic aciduria, with possible accumulation of other lysine and tryptophan catabolites. a-Aminoadipic acid is a specific gliotoxic agent both in vivo and in vitro [27– 29]. This suggests that individuals with AAAT deficiency may present with neurological impairment, though the precise phenotype is hard to predict. The majority of reported cases of a-aminoadipic aciduria are associated with a-ketoadipic aciduria. These cases are likely to represent defects in a-ketoadipic dehydrogenase, the subsequent enzyme to AADAT in lysine and tryptophan catabolism. Vigabatrin may also cause a-aminoadipic aciduria [30]. To our knowledge, there are only two reported cases of isolated a-aminoadipic aciduria. One
179
case involved a mentally retarded girl with dysmorphism and persistent fetal hemoglobinemia [31]. A normal degradation of D L -aminoadipate in fibroblasts of this patient makes a primary defect in AADAT unlikely. Candito et al. [32] described the second case, a male child with a-aminoadipic aciduria, defect in platelet aggregation and antenatal cerebral hemorrhage. No enzymic studies were done, hence, a defect in AADAT is not excluded. We hypothesized that a deficiency of a-aminoadipic semialdehyde dehydrogenase (AASD), a-aminoadipic semialdehyde dehydrogenase-phosphopantetheinyl transferase (AASS-PPT) or AADAT might lead to elevation of pipecolic acid. Plecko et al. [33] found significant elevation of pipecolic acid in the plasma and cerebrospinal fluid of two patients with B6-responsive seizures. We sequenced the cDNA of four patients with isolated pipecolic acidemia (2 with B6 responsive seizures and two with other neurological phenotypes) but were unable to find any mutations in either AADAT or AAsD-PPT. In conclusion, we report the identification and characterization of the gene that encodes for the human AADAT enzyme. To date, we have not identified any individuals with a defect in this enzyme. However, the availability of the AADAT DNA sequence and a specific enzyme assay will allow us to further evaluate individuals suspected to have defects in this enzyme.
Acknowledgments We thank Silvy J.M. van Dooren and Erwin E.W. Jansenfor their expert laboratory support.
References [1] V.W. Rodwell, in: D.M. Greenberg (Ed.), Metabolic Pathways, Academic Press, New York, 1967, pp. 223–223. [2] R.A. Hiles, K. Triebwasser, C.K. Triebwasser, L.M. Henderson, The degradation of hydroxy-L -lysine in liver via its phosphate ester, Biochem. Biophys. Res. Commun. 41 (1970) 662–668. [3] S.W. Queener, N. Neuss, in: R.B Morin, M. Gorman (Eds.), The Chemistry and Biology of b-Lactam Antibiotics, Academic Press, New York, 1982, p. 1. [4] Y.F. Change, Lysine metabolism in rat brain: the pipecolic acidforming pathway, J. Neurochem. 30 (1978) 347. [5] E. Giacobini, Y. Nomura, T. Schmidt-Glenewinkel, Pipecolic acid: origin, biosynthesis and metabolism in brain, Cell. Mol. Biol. 26 (1980) 135. [6] K.A. Sacksteder, B.J. Biery, J.C. Morrell, B.K. Goodman, B.V. Geisbrecht, R.P. Cox, S.J. Gould, M.T. Geraghty, Identification of the a-aminoadipic semialdehyde synthase gene, which is defective in familial hyperlysinemia, Am. J. Hum. Genet. 66 (2000) 1736–1743. [7] V. Praphanphoj, K.A. Sacksteder, S.J. Gould, G.H. Thomas, M.T. Geraghty, Identification of the a-aminoadipic semialdehyde dehydrogenase-phosphopantetheinyl transferase gene, the human ortholog of the yeast LYS5 gene, Mol. Genet. Metab. 72 (2001) 336–342.
180
D.L.M. Goh et al. / Molecular Genetics and Metabolism 76 (2002) 172–180
[8] E. Okuno, M. Tsujimoto, M. Nakamura, R. Kido, 2-Aminoadipate-2-oxoglutarate aminotransferase isoenzymes in human liver: a plausible physiological role in lysine and tryptophan metabolism, Enzyme Protein 47 (1993) 136–148. [9] D.R. Deshmukh, S.M. Mungre, Purification and properties of 2aminoadipate: 2-oxoglutarate aminotransferase from bovine kidney, Biochem. J. 261 (1989) 761–768. [10] Y. Nakatani, M. Fujioka, K. Higashino, a-Aminoadipate aminotransferase of rat liver mitochondria, Biochim. Biophys. Acta 198 (1970) 219–228. [11] M.C. Tobes, M. Mason, L -Kynurenine aminotransferase and L -aaminoadipate aminotransferase. I. Evidence for identity, Biochem. Biophys. Res. Commun. 62 (1975) 390–397. [12] M.C. Tobes, M. Mason, a-Aminoadipate aminotransferase and kynurenine aminotransferase. Purification, characterization, and further evidence for identity, J. Biol. Chem. 252 (1977) 4591–4599. [13] F. Takeuchi, H. Otsuka, Y. Shibata, Purification, characterization and identification of rat liver mitochondrial kynurenine aminotransferase with alpha-aminoadipate amniotransferase, Biochim. Biophys. Acta 743 (1983) 323–330. [14] M.R. Mawal, D.R. Deshmukh, Alpha-aminoadipate aminotransferase and kynurenine aminotransferase activities from rat kidney, J. Biol. Chem. 266 (1991) 2573–2575. [15] M.R. Mawal, D.R. Deshmukh, Purification and properties of aaminoadipate aminotransferase from rat kidney, Prep. Biochem. 21 (1991) 63–73. [16] M.R. Mawal, A. Mukhopadhyay, D.R. Deshmukh, Purification and properties of alpha-aminoadipate aminotransferase from rat liver and kidney mitochondria, Prep. Biochem. 21 (1991) 151–162. [17] R. Buchli, D. Alberatu-Giani, P. Malherbe, C. Kohler, C. Broger§, A. Cesura, Cloning and functional expression of a soluble form of kynurenine/alpha-aminoadipate aminotransferase from rat kidney, J. Biol. Chem. 270 (1995) 29330–29335. [18] P. Yu, D.M. Mosbrook, D.A. Tagle, Genomic organization and expression analysis of mouse kynurenine aminotransferase II, a possible factor in the pathophysiology of Huntington’s disease, Mamm. Genome 10 (1999) 845–852. [19] I. Iraqui, S. Vissers, M. Cartiaux, A. Urrestarazu, Characterisation of Saccharimyces cerevisiae ARO8 and ARO9 genes encoding aromatic aminotransferases I and II reveals a new aminotransferase subfamily, Mol. Gen. Genet. 257 (1998) 238–248. [20] A. Urrestarazu, S. Vissers, I. Iraqui, M. Grenson, Phenylalanineand tyrosine-auxotrophic mutants of Saccharomyces cerevisias impaired in transamination, Mol. Gen. Genet. 257 (1998) 230– 237. [21] D. Pinkle, J. Landegent, C. Collins, J. Fuscoe, R. Segraves, J. Lucas, J. Gray, Fluorescence in situ hybridization with human
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33]
chromosome-specific libraries: detection of trisomy 21 and translocations of chromosome 4, Proc. Natl. Acad. Sci. USA 85 (1988) 9138–9142. B.V. Geisbrecht, D. Zhu, K. Schulz, K. Nau, J.C. Morrell, M. Geraghty, H. Schulz, R. Erdmann, S.J. Gould, Molecular characterization of Saccharomyces cerevisiae D3 ; D2 -Enoyl-CoA isomerase, J. Biol. Chem. 273 (1998) 33184–33191. D.S.M. Schor, E.A. Struys, B.M. Hogema, K.M. Gibson, C. Jakobs, Development of a stable-isotope dilution assay for gamma-aminobutyric acid (GABA) transaminase in isolated leukocytes and evidence that GABA and beta-alanine transaminases are identical, Clin. Chem. 47 (2001) 525–531. M.B. Shapiro, P. Senapathy, RNA spice junctions of different classes of eukaryotes: sequence statistics and functional implications in gene expression, Nucleic Acids Res. 15 (1987) 7155– 7174. J. Jager, M. Moser, U. Sauder, J.N. Jansonius, Crystal structures of Escherichia coli aspartate aminotransferase in two conformations. Comparision of an unliganded open and two liganded closed forms, J. Mol. Biol. 239 (1994) 285–305. P. Guidetti, E. Okuno, R. Schwarcz, Characterization of rat brain kynurenine aminotransferase I and II, J. Neurosci. Res. 50 (1997) 457–465. J.W. Olney, O.L. Ho, V. Rhee, Cytotoxic effects of acidic and sulphur containing amino acids on the infant mouse central nervous system, Exp. Brain Res. 14 (1971) 61–76. J.W. Olney, T. De Gubareff, J.F. Collins, Stereospecificity of the gliotoxic and anti-neurotoxic actions of alpha-amino-adipate, Neurosci. Lett. 19 (1980) 277–282. D.S. Casper, R.L. Trelstad, L. Reif-Lehrer, Glutamate-induced cellular injury in isolated chick embryo retina: M€ uller cell localization of initial effects, J. Comp. Neurol. 209 (1982) 79–90. C. Vallat, F. Rivier, H. Bellet, B. Magnan de Bornier, H. Mion, B. Echenne, Treatment with vigabatrin may mimic a-aminoadipic aciduria, Epilepsia 37 (1996) 803–805. A.J. Manders, C.G. von Oostrom, J.M. Trijbels, F.J. Rutten, W.J. Kleijer, a-Aminoadipic aciduria and persistence of fetal haemoglobin in an oligophrenic child, Eur. J. Pediatr. 136 (1981) 51–55. M. Candito, C. Richelme, P. Parvy, C. Dageville, A. Appert, S. Bekri, D. Rabier, P. Chambon, R. Mariani, P. Kamoun, Abnormal a-aminoadipic acid excretion in a newborn with a defect in platelet aggregation and antenatal cerebral haemorrhage, J. Inher. Metab. Dis. 18 (1995) 56–60. B. Plecko, S. St€ ockler-Ipsiroglu, E. Paschke, W. Erwa, E. Struys, C. Jakobs, Pipecolic acid elevation in plasma and cerebrospinal fluid of two patients with pyridoxine-dependent epilepsy, Ann. Neurol. 48 (2000) 121–125.