Fish Physiol Biochem DOI 10.1007/s10695-009-9358-y
Partial characterization of the gene encoding myoadenylate deaminase from the teleost fish Platichthys flesus M. T. The´bault • A. Tanguy • A. L. Meistertzheim J. P. Raffin
•
Received: 9 February 2009 / Accepted: 21 May 2009 Ó Springer Science+Business Media B.V. 2009
Abstract AMP-deaminase (AMPD, EC 3.5.4.6), which catalyzes the irreversible hydrolytic deamination of AMP to IMP and ammonia, is an important energy-related enzyme. The partial genomic sequence of the gene encoding myoadenylate deaminase (AMPD1) from the teleost fish Platichthys flesus was determined. The amino acid sequence of P. flesus AMPD1 shows 82% homology with that of the teleost fish Danio rerio. Comparison of genomic sequences of P. flesus and Rattus norvegicus reveals a high degree of conservation of both sequence and structural organization. A phylogenetic analysis of AMPD sequences shows that bony fish and mammalian AMPD1s arise by duplication of a common primordial gene.
M. T. The´bault (&) A. L. Meistertzheim UMR CNRS 6539, Laboratoire des Sciences de l’Environnement Marin, Institut Universitaire Europe´en de la Mer, Universite´ de Brest, Place Nicolas Copernic, 29280 Plouzane´, France e-mail:
[email protected] A. Tanguy UMR CNRS 7144 UPMC Evolution et Ge´ne´tique des Populations Marines, Station Biologique, Universite´ Europe´enne de Bretagne, BP 74, Place Georges Teissier, 29682 Roscoff, France J. P. Raffin UMR CNRS 6197, Laboratoire de Microbiologie des Environnements Extreˆmes, Ifremer, 29280 Plouzane´, France
Keywords Myoadenylate deaminase European flounder Molecular characterization Phylogeny
Introduction Adenylate deaminase (AMPD, AMP deaminase, AMP aminohydrolase, adenosine 50 -monophosphate deaminase, EC 3.5.4.6), which catalyzes the irreversible hydrolytic deamination of AMP to IMP and ammonia, is an important energy-related enzyme. By removing AMP from the adenylate pool, AMPD is involved in stabilizing the adenylate energy charge (reviewed by Hancock et al. 2006) and in the control of the purine nucleotide cycle (reviewed by Van den Berghe et al. 1992). AMPD is found in a variety of eukaryotes, including yeast (Solliti et al. 1993) and plants (Turner and Turner 1961), but not in prokaryotes. In higher eukaryotes, tissue-specific isoforms are produced by differential expression of three genes (AMPD1, AMPD2, and AMPD3) that encode the enzyme as well as by alternative splicing of each primary transcript (reviewed by Sabina and Mahnke-Zizelman 2000). In mammals, both AMPD1 and AMPD3 genes are expressed in skeletal muscle, with a greater abundance of AMPD1 mRNA in adult fast-twitch muscle (Sabina and Mahnke-Zizelman 2000). Only a few number of fish AMPD partial sequences of putative isoforms M, L, and E and one complete sequence of Danio rerio AMPD1 cDNA
123
Fish Physiol Biochem
(BC063996) and gene have been determined. Fish AMPD isoforms display different kinetic behavior (Raffin and Leray 1980). Comparative studies on AMPD properties in muscle from different fish have led to some controversy concerning the isoform distribution (Van Waarde and Kesbeke 1981; Kaletha et al. 1991).
Materials and methods Juvenile fishes around 2-years-old were collected in the Ster estuary (Brittany, France). White muscle was dissected from each fish as described in Marchand et al. (2004). Total RNA was extracted from white muscle of Plathychtys flesus collected in Ster according to the guanidium isothiocyanate method. The cDNA was prepared using degenerated PCR primers corresponding to conserved amino acids of the AMPD peptide sequences from Aradopsis thaliana (AY133852), Saccharomyces cerevisiae (M30449), Rattus norvegicus (JO2811), and Homo sapiens (M60092). A series of 50 - and 30 -rapid amplification of cDNA ends experiments was performed to determine the cDNA sequence of the AMPD gene, using a commercial 50 /30 RACE kit (Roche). Sequence analysis and multiple alignments were performed using the FASTA program. First-strand synthesis was performed using an anchored oligo-dT primer (Table 1) and M-MLV reverse transcriptase. Several combinations of degenerated primers were used for subsequent PCR amplification (Table 1). Amplification of specific 50 UTR region of mRNA was performed using an anchor oligo-dT primer sense, a 50 UTR primer antisense designed according to the previous sequence (Table 1) and terminal transferase according to the protocol for the rapid amplification of 50 /30 cDNA ends (50 /30 RACE Kit, Roche, Mannheim, Germany). The genomic sequence of AMPD gene was amplified using different combinations of specific primers designed in the cDNA sequence (Table 1). Subsequently, the PCR products of cDNAs and genomic DNA were inserted into the pGEM-T vector (Promega, Madison, USA). The inserts were sequenced by extension from both ends using T7 and Sp6 universal primers, and internal sense and antisense primers for genomic inserts (T7 sequencing kit, Amersham Pharmacia Biotech, Uppsala, Sweden).
123
Molecular phylogenetic analyses were performed on AMPD amino acid sequences from various organisms including invertebrates and vertebrates. Amino acid sequences were aligned with Clustal W Software. Molecular phylogenetic trees were constructed using the Neighbor-Joining (NJ) algorithm in the PHYLIP software and the phylogenetic package MEGA2. Amino acid differences between sequences were corrected for multiple substitutions using a gamma correction. In this correction, a, the shape parameter of the gamma distribution, was set to 2. With a = 2, the distance between any two amino sequences, dij, is approximately equal to Dayhoff’s (1979) PAM distance per site. Phylogenetic trees were also generated using parsimony. For this analysis, amino acid changes were unweighted; thus a change from one amino acid to any other was equally probable. Support for the major nodes within both distance and parsimony trees were evaluated by bootstrapping the data; 1000 bootstrap replicates of the whole dataset were examined. Homogenization of skeletal muscle was performed as described previously (Raffin and The´bault 1991). AMPD activity was calculated on the basis of the amount of IMP produced in the reaction mixture during the incubation with AMP (The´bault et al. 2005).
Results By combining RT–PCR and 60 /30 RACE, a 1851-bp cDNA (613 amino acids) for Platichthys flesus AMPD was cloned. The sequence has been deposited in GenBank under the accession no. AY660017. From a comparison of other vertebrate AMPD sequences, we showed that the cDNA sequence corresponds to the AMPD1 gene, encoding the M isoform. Sequence analysis and subsequently BlastX analysis revealed an 82% amino acid identity with Danio rerio (BC063996.1), 77% amino acid identity with Macaca mulata (XP_001111615), and Rattus norvegicus (NP_620231.1) AMPD1s. Alignment of the nucleotide sequence of the AMPD1 gene of the teleost fish D. rerio with that of P. flesus revealed that around 600 nucleotides in the 50 end of the expected sequence were missing. A continuing high degree of homology with D. rerio exon 2 coding sequence was observed upstream of the putative P. flesus start
Fish Physiol Biochem Table 1 Primers used to amplify PfAMPD cDNA and gene No.a
Primer name
Nucleotide sequence
Directionb
1
Anchor oligo-dT primer
GACCACGCGTATCGATGTCGACT(T)10
2
Anchor primer sense
GACCACGCGTATCGATGTCGAC
3
Anchor oligo-dT primer bis
CGCTCTAGAACTAGTGGATCT(T)10
4
P1
50 -CNKCNTSHATGAAYCARAARCA-30
F
5
P2
50 -TTCCAHMGHTTYGAYAARTTYAA-30
F
718
6
P3
50 -TCSACNGATGAYCCNHTVCARTTC-30
F
1552
7
P4
50 -TTRAAYTTRTCRAADCKDTGGAA-30
R
718
8
P5
50 -GAAYTGBADNGGRTCATCNGTSGA-30
R
1552
9
50 -UTR primer antisense
CTGATGGTCAGGGCGCGGTACAGACC
R
28
10
30 -UTR primer sense
ATCAAGAGGTCTTATCGTGTGGACGC
F
568
11
1AMPDF
50 -ATGGACGACTTTGAGCTGTCCAAAGG-30
F
1 895
0
Position of 50 -end in cDNA
F
0
532
12
AMPDF4
5 -ATCTACGGCTGCAACCCCGACGAGTGGAG-3
F
13
AMPDF1
50 -GAGTGGAGCAAGCTGGCCGGCTGGTT-30
F
916
14
AMPDF5
50 -GTTGTGTCTCTCTCCACAGACCCCATGCA-30
F
1540
15 16
1AMPDR AMPDR2
50 -TGTCGCAGGTGCTGAGCTTGAAGACC-30 50 -CTGGTGTTTGTCTGCTGCCGCGGC GTC-30
R R
1623 271
17
AMPDR1
50 -AACCAGCCGGCCAGCTTGCTCCACTC-30
R
916
18
AMPDR5
50 -TGCATGGGGTCTGTGGAGAGAGACACAAC-30
R
1540
19
AMPDR6
50 -TACTCAGCCTTCAGGCCTTCCTTGATGCG-30
R
1825
a
Primers 4–8 were degenerated primers. Primers 11–19 were non-degenerated primers used for genomic sequence study
b
F is forward and R is reverse direction
codon, suggesting a longer P. flesus AMPD1 ORF. However, attempts to clone longer 50 sequences in the flounder by 50 RACE were unsuccessful. The 30 untranslated (UTR) sequence comprises 248-bp not including the polyA-tail. The partial sequence of the P. flesus AMPD1 gene (PfAMPD1) was deposited in the GenBank database (Accession number AY660016). The genomic sequence contains 12 coding exons of 160, 220, 130, 195, 132, 164, 127, 164, 121, 174, 111, and 153 bp, respectively, separated by 11 introns of 82, 323, 1151, 827, 229, 419, 160, 199, 252, 249, and 381 bp, respectively (Fig 1a). All intron borders of P. flesus AMPD gene start and end with the consensus GT and AG splicing signals. Predicted amino acid homologies for individual exons between rat and flounder range from 50% (i.e., exon 5) to 90% (i.e., exon 9) (Fig. 1c). Comparison of rat and flounder AMPD1 genes, shows that exon sizes have been precisely conserved between the two species, with the exception of exons 5, 10, and 16. Exon 16, the last exon, is 19 nucleotides
shorter in the flounder (Fig. 1a). Intron size ranges from quite small (i.e., 82 bp; intron 5) to 1.15 kb (intron 7). The length of individual introns is not well conserved between flounder and rat AMPD1 genes (Fig. 1a). For five of the eleven introns compared (introns 7, 8, 10, 13, and 14), the rat intron size is smaller than its flounder homolog. Only the intron 5 is markedly shorter in the flounder. Examination of the P. flesus gene showed that the highly conserved motif SLSTDDP (nt 516–522) involved in the active site of the enzyme (Mahnke-Zizelman and Sabina 1992) and the conserved sequence EPLMEEYAIAAQVFK (nt 530–544) involved in the ATP-binding of AMPDs (Mahnke-Zizelman and Sabina 1992) are localized in exons 14 and 15, respectively. Thirty AMPD amino acid sequences that spanned several kingdoms (vertebrates, invertebrates, unicellular eukaryotes, and plants) were selected for molecular phylogenetic analysis. The phylogenetic tree we obtained shows each of the 30 proteins segregating into one of five major AMPD variants (AMPD1,
123
Fish Physiol Biochem Fig. 1 Physical structure of the Platichthys flesus and Rattus norvegicus AMPD1genes. a P. flesus AMPD1 gene. b R. norvegicus AMPD1 gene. Exons are numbered and depicted by vertical bars. Introns are numbered and represented by horizontal lines. Exon and intron sizes are given below. Arrows refer to the oligonucleotide primers used to amplify the gene (' forward, b reverse). c Predicted amino acid homology between individual exons of P. flesus and the comparable regions of R. norvegicus AMPdeaminase (from Sabina et al. 1990)
AMPD3, AMPD2, AMPD2-like, and ‘‘ancestral’’ AMPDs) (Fig. 2). The AMPD1 gene from mammalian species forms a monophyletic group separate from the bony fish group. The phylogenetic tree we obtained clearly shows that the PfAMPD1 gene is distinct from Tetraodon nigroviridis, Danio rerio, and Salmo salar AMPD1 sequences. The tree shows that AMPD1 is present exclusively in mammals and fish, whereas AMPD3 is present in tetrapods, birds, fishes, and tunicates. AMPD2 is a very large and diversified group which has evolved, presumably through gene duplication events, to vertebrate AMPD2 and invertebrate AMPD2-like. ‘‘Ancestral AMPDs’’ is a multigene group that falls in unicellular eukaryotes and plants. Genetic distance between the different phylogenetic groups appeared to be relatively high, especially between species displaying ‘‘ancestral’’ forms.
123
The total activity of AMPD in P. flesus white muscle was 160 ± 46 lmol min-1 gww-1 and was similar to that of rainbow trout white muscle (The´bault et al. 2005).
Discussion From a comparison of other vertebrate AMPD sequences, we showed that the genomic sequence we characterized in the flounder P. flesus corresponds to the AMPD1 gene, encoding the M isoform. On average, the PfAMPD amino acid sequence shares 82 and 77% homology with sequences from another teleost fish, Danio rerio, and a mammal, Macaca mulata, respectively. Alignments among the predicted amino acid sequences encoded by all known
Fish Physiol Biochem
Fig. 2 A phylogenetic tree showing the most likely relationship between representative AMPD amino acid sequences. Branch lengths are proportional to estimates of evolutionary change. Then number associated with each internal branch is the local bootstrap probability, which is an indicator of confidence. The sequences are Platichthys flesus (AY660017), Homo sapiens AMPD1 (NP_000027), Macaca mulata AMPD1 (XP_001111615), Sus scrofa AMPD1 (ABR26259), Rattus norvegicus AMPD1 (NP_620231), Mus musculus AMPD1 (NP_001028475), Danio rerio hypothetical protein LOC39 3867 (NP_957187), Tetraodon nigroviridis (CAG01709), Salmo salar AMPD1 (NM_001141677), Xenopus tropicalis AMPD3 (NP_001025687), Gallus gallus similar to AMPdeaminase (XP_420973), Homo sapiens AMPD3 (NM_00
0480), Rattus norvegicus AMPD3 (EDM17861), Mus musculus AMPD3 (NP_033797), Ciona intestinalis AMPD3 (XM_00211 9875), Homo sapiens AMPD2 (NP_631895), Rattus norvegicus AMPD2 (NP_001095151), Danio rerio AMPD2 (XM_69 4936), Hydra magnipapillata AMPD2 (XM_002169916), Strongylocentrus purpuratus hypothetical protein (XP_792 615), Lumbriculus variegatus AMPD (EU624494), Mesenchytraeus solifugus AMPD (EU624492), Caenorhabditis elegans (NP_001040752), Drosophila melanogaster isoform D (NP_72 7741), Anopheles gambiae (XP_310497), Neurospora crassa (EAA36130), Saccharomyces cerevisiae YJM789 (EDN6 4358), Arabidopsis thaliana (NP_565886), and Oryza sativa probable AMP-deaminase (Q84NP7)
AMPD genes indicate strong amino acid identities in the carboxyl two-thirds of the enzyme, while the amino one-third displays much lower homologies, as previously shown by Bausch-Jurken and Sabina (1995) and Mahnke-Zizelman and Sabina (1992). Previous studies of endogenous and recombinant AMPD activities from mammals have suggested that N-terminal amino acid residues play a significant role in the structural/functional properties of the enzyme and influence regulatory properties of the enzyme (Bausch-Jurken and Sabina 1995). The structural organization of the AMPD1 gene has been remarkably conserved over the 450 million
years since the divergence of bony fish and tetrapods (Takahashi et al. 2000). Comparison of the flounder AMPD1 gene with that of mammals reveals that intron size varies much more than exon size. Moreover, 30 -flanking sequences were markedly larger in the flounder. The presence of an unusually short 30 untranslated region in human and rat genes suggests this structural motif may have been preserved for some purpose and raises the question of whether this transcript may have and altered stability, translatability, or subcellular localization (Sabina et al. 1990). The phylogenetic tree we obtained corroborates the bony fish cladogram based on data from Cossins
123
Fish Physiol Biochem
and Crawford (2005) indicating that Pleuronectiformes are more closely related to Tetraodontiformes than to Cyprinoformes and Salmonids. Our results show that the AMPD1 and AMPD3 genes arise by duplication of a common primordial gene. AMPD3 gene is expressed in heart, slow-twitch skeletal muscles and non-muscle tissues whereas AMPD1 is expressed at high level in fast-twitch skeletal muscle (Morisaki et al. 1990). The level of AMPD activity in bony fish white muscles is similar to that in mammals whereas a steady decrease of 3–4 fold is measured between some cartilagineous fish (Raffin and The´bault 1994). Our previous detailed study on AMPD in white muscle from two cartilagineous fish has shown that the difference in enzyme activity between the two species was correlated with the appearance in Rajidae of the AMPD1 gene product found in mammals (The´bault et al. 2005). The absence of both AMPD1 and AMPD3 sequences for invertebrates suggests that the corresponding isoforms could be present in chordates only. Few data are available concerning the AMPD2 gene. To our knowledge, only a partial sequence of AMPD2 gene has been recently reported in the teleost fish Danio rerio (accession number: BC107820) confirming the existence of the three AMPD variants in bony fish. Our phylogenetic tree shows that vertebrate AMPD2 and invertebrate AMPD2-like genes aroused by duplication of a common primordial gene. Lack of crustacean and mollusc AMPD sequences did not permit a more detailed analysis of their position in the tree. Our attempts to clone AMPD sequences by 50 RACE from the Japanese oyster Crassostrea gigas were unsuccessful. Very low AMPD activities were found in invertebrate muscles in which ATP was necessary to detect the activity (Raffin and The´bault 1994). A non-specific AMPD was isolated from the snail Helix pomatia (Stankiewicz 1986). The ability to deaminate ADP, ATP, and NADH in addition to AMP makes the snail AMPD the only example described so far in animals, suggesting that the mollusc AMPD gene could constitute an isolated branch in the invertebrate group (Stankiewicz 1986). ‘‘Ancestral AMPDs’’ is a multigene group that falls in unicellular eukaryotes and plants. Genetic distance between the different phylogenetic groups appeared to be relatively high, especially between species displaying ‘‘ancestral’’ forms. AMPD from yeast was activated by ATP and specific for AMP (Solliti et al. 1993). Plant AMPDs
123
form a group that has originated much earlier than the diversification of invertebrates and vertebrates. AMPD enzymes exhibit unique physical and regulatory behaviors. In Arabidopsis thaliana, the enzyme assembles as a physiological dimer, contrary to AMPD enzymes of the animal kingdom that are tetramers (Han et al. 2006). In this article we report the characterization of a partial genomic sequence of white muscle AMPD in P. flesus and examine the phylogenic relationships of PfAMPD. Our data provide a basis for further study of the regulation of the PfAMPD1 gene. The determination of the genomic sequence now makes possible to search for functional polymorphism to compare the genetic structure of flounders exposed to various environmental factors.
Acknowledgments We would like to thank M.S. M’Boumba for her excellent technical assistance. We applied the FLAE (first–last-author emphasis) approach combined with EC (equal contribution): ? (1st and 2nd authors) for the sequence of authors (Tscharntke et al. 2007).
References Bausch-Jurken MT, Sabina RL (1995) Divergent N-terminal regions in AMP deaminase and isoform-specific catalytic properties of the enzyme. Arch Biochem Biophys 321: 372–380 Cossins AR, Crawford DL (2005) Fish as models for environmental genomics. Nat Rev Genet 6:324–333 Dayhoff MO (1979) Atlas of protein sequence and structure, vol 5. National Biomedical Research Foundation, Silver Springs, ND Han BW, Bingman CA, Mahnke DK, Bannen RM, Bednarek SY, Sabina RL, Phillips GN Jr (2006) Membrane association, mechanism of action and structure of Arabidopsis embryonic factor 1 (FAC1). J Biol Chem 281:14939– 14947 Hancock CR, Brault JJ, Terjung RL (2006) Protecting the cellular energy state during contractions: role of AMP deaminase. J Physiol Pharmacol 57:17–29 Kaletha K, The´bault MT, Raffin JP (1991) Comparative studies on heart and skeletal muscle AMP-deaminase from rainbow trout (Salmo gairdneri). Comp Biochem Physiol 99B:751–754 Mahnke-Zizelman DK, Sabina R (1992) Cloning of human AMP deaminase isoform E cDNAs. Evidence for a third AMPD gene exhibiting alternatively spliced 50 -exons. J Biol Chem 267:20866–20877 Marchand J, Quiniou L, Riso R, The´bault MT, Laroche J (2004) Physiological cost of tolerance to toxicants in the European flounder Platichthys flesus, along the Atlantic coast of France. Aquat Toxicol 70:327–343
Fish Physiol Biochem Morisaki T, Sabina RL, Holmes EW (1990) Adenylate deaminase. A multigene family in humans and rats. J Biol Chem 265:11482–11486 Raffin JP, Leray C (1980) Comparative study on AMP deaminase in gill, muscle and blood of fish. Comp Biochem Physiol 67B:533–540 Raffin JP, The´bault MT (1991) A specific AMP deaminase assay and its application to tissue homogenates. Comp Biochem Physiol 99B:125–127 Raffin JP, The´bault MT (1994) Amplification of myoadenylate deaminase during evolution. A comparative study. CR Acad Sci 317:386–391 Sabina RL, Mahnke-Zizelman DK (2000) Towards an understanding of the functional significance of N-terminal domain divergence in human AMP deaminase isoforms. Pharmacol Ther 87:279–283 Sabina RL, Morisaki T, Clarke P, Eddy R, Shows TB, Morton CC, Holmes E (1990) Characterization of the human and rat myoadenylate deaminase genes. J Biol Chem 265: 9423–9433 Solliti P, Merkler DJ, Estupinan B, Schramm VL (1993) Yeast AMP deaminase. Catalytic activity in Schizosaccharomyces pombe and chromosomal location in Saccharomyces cerevisiae. J Biol Chem 268:4549–4555
Stankiewicz A (1986) Non specific snail muscle adenylate deaminase: simplified purification, characterization and use for the preparation of deamino derivates of NAD, NADH and AMP-P(NH)P. Enzyme 36:187–196 Takahashi K, Rooney AP, Nei M (2000) Origins and divergence times of mammalian class IIMHC gene clusters. Heredity 91:198–204 The´bault MT, Izem L, Leroy JP, Gobin E, Charrier G, Raffin JP (2005) AMP-deaminase in elasmobranch fish: a comparative histochemical and enzymatic study. Comp Biochem Physiol 141B:472–479 Tscharntke T, Hochberg ME, Rand TA, Resh VH, Krauss J (2007) Author sequence and credit for contributions in multiauthored publications. PloS Biol 5:13–14 Turner DH, Turner JF (1961) Adenylic deaminase of pea seeds. Biochem J 79:143–147 Van den Berghe G, Bontemps F, Vibcebt MF, Van den Berghe F (1992) The purine nucleotide cycle and its molecular defects. Prog Neurobiol 39:547–561 Van Waarde A, Kesbeke F (1981) Regulatory properties of AMP-deaminase from lateral red muscle and dorsal white muscle of goldfish, Carassius auratus (L). Comp Biochem Physiol 69B:413–423
123
