J. Biochem. I l l , 123-128 (1992)
Human Long-Chain Acyl-CoA Synthetase: Structure and Chromosomal Location1 Takaaki Abe,' Takahiro Fujino,* Ryuichi Fukuyama," Shinsei Minoshima," Nobuyoshi Shimizu,** Hiroyuki Toh,"* Hiroyuki Suzuki,' and Tokuo Yamamoto*2 'The Tohoku University Gene Research Center, Aoba-ku, Sendai, Miyagi 981; "The Department of Molecular Biology, Keio University School of Medicine, Shinjuku-ku, Tokyo 160; and '"The Protein Engineering Research Institute, Suita, Osaka 565 Received for publication, September 6, 1991
A complementary DNA clone encoding the entire human long-chain acyl-CoA synthetase was isolated and the total 698-amino acid sequence was deduced. The amino acid sequence of human long-chain acyl-CoA synthetase shows 84.9% identity to that of rat long-chain acyl-CoA synthetase. The nucleotide sequences of the protein coding regions between human and rat long-chain acyl-CoA synthetase mRNAs are highly conserved (85.6%), whereas those of the 3' untranslated regions are less conserved (72%). The location of the human long-chain acyl-CoA synthetase gene was identified on chromosome 4 by spot hybridization of flow-sorted chromosomes. Computer-assisted homology search revealed a significant similarity of the enzyme with the enzymes of the luciferase family. Based on this similarity, the structure of human long-chain acyl-CoA synthetase can be divided into five domains: the N-terminus, two domains similar to those in enzymes of the luciferase family, a long gap region between the similar domains and the C-terminus.
Long-chain acyl-CoA synthetase [EC 6.2.1.3] plays a critical role in fatty acid metabolism in mammals. AcylCoAs produced by the enzyme are important intermediates not only for the degradation of fatty acid via fi- oxidation in mitochondria and peroxisomes but also for the production of glycerolipids and cholesteryl ester in microsomes. Acyl-CoAs are also important for the regulation of fatty acid biosynthesis. The activity of acetyl-CoA carboxylase, a rate-limiting enzyme in fatty acid biosynthesis, is negatively regulated by acyl-CoAs (1). In addition to fatty acid metabolism, acyl-CoAs are required for the acylation of many membrane proteins (2). Furthermore, a recent study by Bronfman et al. (3) revealed that acyl-CoAs modulate the activity of protein kinase C, a key enzyme in transmembrane signaling. In rat liver, long-chain acyl-CoAs synthetases localized in microsomes, mitochondria, and peroxisomes are identical with respect to molecular, catalytic, and immunological properties (4, 5). The mechanism by which long-chain acyl-CoAs synthetase is transported to the three different organelles is unknown, but this system provides an excellent model for studying the intracellular transportation of the protein. In our previous study, we reported the isolation and sequence analysis of a cDNA for rat long-chain acyl-CoA synthetase (6). The results showed that the enzyme contains 699-amino acid residues and a part of the enzyme is very similar to firefly luciferase. Based on the similarity of the reaction mechanism of the two enzymes, we have proposed that the firefly luciferase -like region found in rat 1 This work was supported by research grants from the Ministry of Education, Science and Culture of Japan, the Urakami Foundation, the Naito Foundation, and the Japan Heart Foundation. 2 To whom correspondence should be addressed.
Vol. I l l , No. 1, 1992
long-chain acyl-CoA synthetase may constitute a region where ATP and the carboxyl group of a fatty acid interact to form acyl-AMP. The level of the long-chain acyl-CoA synthetase mRNA is high in tissues performing active fatty acid oxidation (heart) or triglyceride synthesis (liver and adipose tissue). We have also shown that the level of the mRNA in liver is remarkably increased by feeding a lipogenic diet. In order to extend our knowledge of the structure, function, and regulation of long-chain acyl-CoA synthetase, we have isolated and characterized a cDNA encoding human long-chain acyl-CoA synthetase. With this cDNA we have determined the chromosomal location of the enzyme. Based on the similarities of structure and reaction mechanism of long-chain acyl-CoA synthetase to the enzymes of the luciferase family, we discuss the functional domains of the enzyme. MATERIALS AND METHODS Materials—All restriction enzymes, T4 DNA ligase, Escherichia coli DNA ligase, DNA polymerase I (E. coli), large fragment of E. coli DNA polymerase I, T4 DNA polymerase, terminal deoxynucleotidyl transferase, Rous associated virus 2 reverse transcriptase, mung bean nuclease, exonuclease HI, RNase H (E. coli), human placenta RNase inhibitor, and calf intestine alkaline phosphatase were purchased from Takara Shuzo (Kyoto). T7 DNA polymerase (Sequenase™) was from United States Biochemical. Human liver poly(A)+ RNA was obtained from Stratagene. The 32P-labeled nucleotides and "S-labeled nucleotides were products of either Amersham or Du PontNew England Nuclear. General Methods—Preparation of plasmid DNA, restric-
123
124 tion enzyme digestions, screening of the cDNA library, agarose gel electrophoresis, DNA blotting, and hybridization were performed by standard procedures (7). DNA probes were labeled with 3 T by random octanucleotide priming (8). To sequence the entire cDNA insert, the cDNA insert was shortened successively by exonuclease III (9) and subcloned into pUC vectors. Chain termination sequencing (20) was performed on denatured supercoiled plasmid DNA (11) with specific oligonucleotide primer and
T. Abe et al. T7 DNA polymerase. Both strands were sequenced entirely. The DNA sequence was analyzed on an NEC PC9801RA5 with programs from Software Development (Tokyo) or SciSoft. cDNA Cloning—The cDNA library was constructed from human liver poly(A)+ RNA by the method of Gubler and Hoffman (12) with slight modifications. Briefly, a synthetic primer-adapter consisting of a NotI site and a homopolymeric tract ( 5 ' C T C T A G A G G C G G C C G C T J O was used in-
-13TCAACACAGGACA -1 ATGCAAGCCCATGAGCTGnCCGGTATTnCGAATGCCAfi«a6GTTGACnCCGACAGTACGTGCGTACTCTTCC6ACCAACACGCTT 9 0 M Q A H E L F R Y F R M P E L V D F R Q Y V R T L P T H T L 3 0 ATGGGCTTCGGAGCTnTGCAGCACTCACCACCTTnGGTACGCCACSAGACCCAAACCCaGAAGCCGCCATGCGACCTCTCCATGCAG 180 N G F G A F A A L T T F W Y A T R P K P L K P P C 0 L S M Q 6 O TCACTGGAAGTGGCGGGTAeTGGTGGTGCACGAA6ATCCSCACTACTTGACASCGACGAGCCCnGGTGTATTTCTATGATGATGTCACA 270 S V E V A G S G G A R R S A L L D S D E P L V Y F Y D 0 V T 9 O ACATTATACGAAGGTTTCUGAGeGGAATACAGGTGTCAAATAATGeCCCnGTTTAGGaCTCGGAAACCAGACCAACCCTATGAATGG 360 T L Y E G F Q R G I Q V S N N G P C L G S R K P D Q P Y E M I T O CTTTCATATAAACAGGTTGCAGAAnGTCGGAGTGCATAGGCTCACCACTGATCCAGAAGeGCnCAAGACTGCCCCAGATCAGnCAn 450 L S Y K Q V A E L S E C I G S A L 1 Q K G F K T A P D Q F I 1 S 0 GGCATCTTTGCTCAAAATAGACCTGAGTG6STGAnATTGAACA«GATGCnTGCnAnCG*TGGTGATCGTTCCACTTTATGATACC 540 G 1 F A Q H R P E W V I I E Q G C F A Y S H V I V P L Y 0 T 1 8 O CTTGGAAATGAAGCaTCACGTACATAfiTCAACAAAGCTGAACTaCTaGGTTTTTGnGACAAGCCAGAGAAGGCCAAACTCnATTA 630 L G N E A 1 T Y I V N K A E L S L V F V 0 K P E K A K L L L 2 1 0 GAGGGTGTAGAAAATAAGnAATACCAGGCmAAMTCATAGTTGTCATGGATGCCTACGGCAGTGAACTGGTGGAACGAGGCCAGAGG 720 E G V E N K L I P G L K I I V V M D A Y G S E L V E R G Q R 2 4 0 TGTGGGeTGGAAGTCACCAGCATGAAGGCGATGGAGGACaGGGAAGAGCCAACAGACGGAAGCCCAAGCCTCCAGCACaGAAGATCn 810 C G V E V T S H K A H E D L G R A N R R K P K P P A P E D L 2 7 0 GCAGTAATTTGTTTCACAAGTGGAACTACAGGCAACCCCAAAGGAGCMTGGTCACTCACCGAAACATAGTGAGCGAnGnCAGCTTTT 900 A V I C F T S G T T G N P K G A H V T H R N I V S 0 C S A F 3 O 0 GTGAAAGaWCAGAGAATACAGTCAATCCnGCCCAGATGATACTTTGATATCTTTCnGCCTCTCGCCCATATGTnGAGAGAGnGTA 990 V K A T E N T V N P C P D 0 T L I S F L P L A H H F E R V V 3 3 O GAGTGTGTAATGCTGTGTCATGeAGCTAAAATCGGATTTTTCCAAGGAGATATCAGGaGaCATGGATGACCTCAAGGTGCTTCAACCC 1080 E C V H L C H G A K 1 G F F Q G D I R L L N D 0 L K V L Q P 3 6 O AaGTCnCCCCGTGGnCCAAGAaGCTGAACCG6ATGTTTGACCGAATTTTCGGACAAGCAAACACCACGCTGAAGCGATGGCTCTTG 1170 T V F P V V P R L L N R H F D R I F G q A N T T L K R W L L 3 9 0 GACTTTGCCTCCAAGAGGAAAGAAGCAGAGCnCGCAGCGeUTCATCAGAAACAACAGCCTGTGeeACCGGCTGATCnCCACAAAGTA 1260 D F A S K R K E A E L R S G I I R N K S L W D R L I F H K V 4 2 O CACTCGAGCCTGGGCGGAAGAGTCCGGCTGATGGTGACAGGAGCCGCCCCGGTGTaGCCAaGTGCTGACGnCCTCAGAGCAGCCaG 1350 Q S S L G G R V R L M V T G A A P V S A T V L T F L R A A L 4 5 O GGCrcTCAGTTrrATGAAG6ATACGGACAGACAGAGTGCACTGCCGGGTGaGCCTAACCATGCaGGAGACTGGACCGCAGGCCATGn 1440 G C q F Y E G Y G Q T E C T A G C C L T M P G D H T A G H V 4 8 0 GGGGCCCCGATGCCGTGCAATTTGATAAAACnGnGATGTGGAAGAAATGMTTACATGGCTGCC6AGGGCGAGGGCGAGGTGTGTGTG 1530 G A P H P C N L 1 K L V 0 V E E H N Y M A A E G E G E V C V 5 1 0 AAAGGGCCAAATGTATnCAGGGCTACnGAAGeACCCAGCGAAAACA6CAGAAGCTnGGACAAAGACGGCTGGTTACACACAGGGGAC 1620 K G P H V F Q G Y L K D P A K T A E A L D K D G W L H T 6 0 5 4 0 AnGGAAAATGGnACCAAATGGCACCnGAAMnATCGACC6GAAAAAeCACATATTTAAGaGeCACAAGGAGAATACATAGCCCCT 1710 1 G K H L P N G T L K 1 I D R K K H I F K L A Q G E Y I A P 5 7 O GAAAAGAnGAAMTATaACATGCGAAGTGAGCaGnGCTCAGGTGTTTGTCCACGeAGAAAGCaGCAGGCATTTCTCAnGCAAn 1800 E K I E H I Y M R S E P ' V A Q V F V H G E S L Q A F L I A I 6 O 0 GTGGTACCAGATGnGAGACAnATGnCCTGGGCCCAAAAGAGAGGATTTGAAGGGTCGTTrGAGGAAaGTGCAGAAATAAGGATGTC 1890 V V P D V E T L C S W A Q K R G F E G S F E E L C R N K 0 V 6 3 O AAAAAAGaATCCTCGAAGATATGGTGAGACTTGGeAAGGATTCTGeTCTGAAACCATTTGMCAGGTCAAAGGCATCACATTGCACCa 1980 K K A I L E O H V R L G K D S G L K P F E Q V K G I T L H P 6 6 0 GMnATTnCTATCGACMTGGCCnCTGACTCCXACAATGAAGGCGAAAAGGCCAGAGCTGCGGAACTATTTCAGGTCGCAGATAGAT 2070 E L F S I D N G L L T P T H K A K R P E L R N Y F R S Q I D 6 9 0 GACaCTAnCCACTATUAGGTTTAGTGTGAAGAAGAAA6aCAGAGGAAATGGCACAGnCCACAATCTCTTaCCTGCTGATGGCCT 2160 O L Y S T I K V * 698 TCATGnGnAAnnGAATACAGCAAGTGTAGGGAAGGAAGCGnCGTGTTrGACrTGTCCAnCGGGGnCnCTCATAeGAATGCTA 2250 GAGGAAACAGMCACCGCCnACAGTCACCTCATGnGCAGACUTGTnATGGTAATACACACTTTCCAAAATGAGCCnAAAAAnGT 2340 AAA6GG6ATAnATAAATGTGCTAAGnATTTGAGACnCCTCAGTnAAAAAGTG66TrnAAATCTTCTGTCTCCaGCTTTTCTAAT 2430 CAAfiGGenAGGACTTTGaATCTCTGAGATGTnGCTACTTGCTGCAAAnCTGCAGCTGTCTGaGaCTAAAGAGTACAGTGCACTA 25?0 GAGGSAAGTGnCCCTTTAAAAATAAGAACAACTGTCCTGGaGGAGAATaCACAAeCGGACCAGAGATCTTTTTAAATCCCTGCTACT 2610 GTCCCTTCTCACAGGCAnWCAGAACCCnCTGAnCGTAAGG6nACGAAACTCATGnCTTCTCCAGTCCCCTGTGGTTTCTGnG6 2700 AGCATAAGCTnCCAGTAAGCGGGAGGGCAGATCCAAaCAGAACCATGCAGATAAGGAGCaaGGCAAATGGGTGCTCATCAGAACGC 2790 GT6eAnCTCnTOTGGaVGAAT6CTCnGGAaCG6naCCAG6CaGATTCCCCSACTCCATCCTTTTTCAGGG6TTATTTAAAAA 2880 TCTGCCTTAGAnCTATAGTGAAGACAAGCATTTCAAGAAAGAGTTACCTGGATCAGCCATGaCAGaGTGACGCCTGAATAACTGTa 2970 ACTTTATCnCACTGMCCAaCACTCTGTGTAAAGGCCAACAGATTTTTAATGTGGTTnCATATCAAAAGATCATGnGGGAnAACT 3060 TGCCTTTTTCCCCAAAAAATAAACTCTCAGGCAAGCATTTCTTTAAA6CTAnAAGGeAeTATATACnGAGTACnATT6AAATRCACA 3150 GTAATAAGCAAATGncnATAATGCTACCTGATTTCTATGAAATGTGTTTGACAAGCCAAAAnCTAGGATGTAGAAATCTGGAAAGn 3240 CATTTCCTGGGAnCACncTCCAGGGATTTTTTAAAGnAATTTGGGAAATTAACAGCAGnCACnTAnGTGAGTCTTTGCCACATT 3330 TGACTGAAnGAGCTGTCATnGTACAXrjA/WGO\GCTGTTTTGGGGTCTGTGAGAGTACATGTATTATATA«AGCACAACAGGGCn 3420 GCACTAAAGAAnGTCAnGTAATAACACTACTTGGTAGCCTAACnMTATATGTAnCnAAnGCACAAAAAGTCAATAATTTGTCA 3510 CCnGGGGTTTTGAATGTTTGCTnAAGTGTTGGCTATnCTATGTTnATAAACCAAAACAAAATTTCCAAAAACAATGAAGGAAACCA 3600
Fig. 1. Nucleotide sequence of the cDNA corresponding to human longchain acyl-CoA gynthetase mRNA and the predicted amino acid sequence of the protein. Two potential polyadenylation signals in the 3' untranslated region are underlined. Two potential mRNA destabilizing signals in the 3' untranslated region are indicated by overlines and underlines. J. Biochem.
125
Human Long-Chain Acyl-CoA Synthetase stead of oligo(dT)12-18 to prime the synthesis of the first strand of cDNA. After the replacement synthesis of the second strand, the cDNA was cleaved with Notl and then ligated to Bluescript vector (Stratagene) which had been cleaved with Notl and EcoRV. The human long-chain acyl-CoA synthetase cDNAs were screened at low stringency with "P-labeled probe prepared from the cDNA for rat long-chain acyl-CoA synthetase (6). Screening of l x 10s clones yielded two positive clones. The longest clone, designated pHACSl, was subjected to restriction enzyme mapping and its nucleotide sequence was determined. Computer-Assisted Homology Search—Programs for weak homology detection (13) and homology matrix construction (14) were used. According to the pairwise alignments made by the program for weak homology detection and the detailed outputs of homology matrices, multiple alignments were constructed. The significance of the detected sequence similarity was analyzed by statistical testing with 100 randomized sequence pairs (25). Differences and alignment scores among the aligned sequences were calculated, considering a continuous gap as a single substitution regardless of its length (14). Chromosomal Mapping—Human metaphase chromosomes were prepared from the human B-lymphoblast line GM00130B (46,XY). These cells showed an apparently normal karyotype. Preparation, staining, and sorting of metaphase chromosomes were described previously (16, 17). Fifty thousand chromosomes of each type were sorted to a small spot on nitrocellulose filter disks and treated for a hybridization. Assignment of the chromosomes in the sorted fraction was made according to the method of Lebo et oL (18) and independently confirmed with chromosome -
specific DNA probes (16, 19). RESULTS AND DISCUSSION Isolation and Characterization of Human Long-Chain Acyl-CoA Synthetase cDNA—A near full-length cDNA for human long-chain acyl-CoA synthetase was isolated from the human liver cDNA library by cross-hybridization with the cDNA for rat long-chain acyl-CoA synthetase. Figure 1 shows the nucleotide sequence of the human long-chain acyl-CoA synthetase cDNA and deduced amino acid sequence of the protein. An initiator methionine codon (nucleotide positions 1-3) with a sequence matching Kozak's consensus sequence (20) begins an open reading frame of 2,094 bp, corresponding to 698 amino acids (Mr 78,942). The human enzyme is shorter than rat enzyme by one amino acid. The degree of sequence identity between the human and rat enzymes is 84.9%; a gap was counted as one substitution (Fig. 2). When conservative replacements are included in the calculation, the similarity of the human enzyme to the rat enzyme is 94.8%. The nucleotide sequence of the protein coding region of the human long-chain acyl-CoA synthetase mRNA is highly conserved between the two species (85.6%), whereas that of the 3' noncoding region is less conserved (72%). Two potential mRNA destabilizing signals (AUUTJA) (21) are localized at nucleotide positions 2872 and 3357, suggesting the possibility that the mRNA may undergo a rapid degradation. Chromosomal Location of the Human Long-Chain Acyl-CoA Synthetase Gene—The human long-chain acylCoA synthetase cDNA was used for spot-hybridization of flow-sorted human chromosomes from B-lymphoblast line
119
1 HACS
MQAHELFRYF(WPEL\raFRQWRTLPTNTU«FGAFAALnFWYATRPKPUPPCOLSMQSVEVA6-SSGARRSALLDSOEPLVYFYDDVTTLYEGFqRGI0VSNKGPCLGSRKPDQPYE ************************************************ • * #++*.*. *
RACS
#.+
************
*******************************
HACS RACS
VGAPMPCNYIKimEDMMYiyWffiEGEVCTKGANVFICGYLKOPARTAEALDIUKWL 481
600
600 HACS RACS
698 IWPDVETLCSWAQKreFEGSFEELCRNICDVICMILEDWRLGKOSGUCPFEQVKGITLHPELFSI^ ******* * a * * * * * * * . * * * * * * * * * * * . ********.***.•***********..************** a . * * * * * * * * * * * * * * * * . * * * * * * . IWPOVEILPSWAqWffiF^SFEELCRHiaJINWaEDMVICLGICMAGLKPFEQVKGIAVHPELFSIOHGLLTPTUAKRPELRNYFRSQIOELYSTIICI
601 699 Fig. 2. Alignment of the similar amino acid sequences of human (HACS) and rat (RACS) long-chain acyl-CoA synthetases. The two proteins are aligned to obtain maximum identity. Residues identical with those of human long-chain acyl-CoA synthetase are indicated with asterisks. Conservative substitutions are dotted. Vol. I l l , No. 1, 1992
126
T. Abe et al.
GM00130B (46,XY). Human chromosomes were separated into 16 or 17 fractions when stained with Hoechst 33258 (Fig. 3A) or propidium iodide (Fig. 3C), respectively. Spot-hybridization with flow-sorted chromosomes stained with Hoechst 33258 revealed a strong hybridization signal in fraction B, which contains chromosomes 3 and 4 (Fig. 3B). Hybridization with flow-sorted chromosomes stained with propidium iodide (Fig. 3D) revealed a strong hybridization signal in fraction c, which contains chromosome 4. These results indicate that the long-chain acyl-CoA synthetase gene is localized on chromosome 4. Amino Acid Sequence Similarity of Long-Chain AcylCoA Synthetase with the Enzymes of the Luciferase Family—In a previous paper, we showed that the sequence of amino acid residues 458 to 591 of rat long-chain acyl-CoA synthetase was similar to a part of firefly luciferase (6). Based on the similarity of the reactions catalyzed by long-chain acyl-CoA synthetase and firefly luciferase, we have proposed that the luciferase-like region of long-chain acyl-CoA synthetase might constitute a region where ATP and the carboxyl group of fatty acid interact to form acyl-AMP. Recently, Toh (22) reported that the N-terminal halves of gramicidin S synthetase 1 and tyrocidine synthetase 1 are similar to plant 4-coumarate:CoA ligase and luciferase from click beetle and firefly. Since long-chain acyl-CoA synthetase is similar to firefly luciferase, we compared the human long-chain acyl-CoA synthetase se-
quence with those of the enzymes of the luciferase family (22), luciferases from firefly (23) and click beetle (24), Bacillus brevis gramicidin S synthetase 1 (25), B. brevis tyrocidine synthetase 1 (26), and 4-coumarate:CoA ligases from parsley (27). In this analysis, we used computer programs for weak homology detection (13) and homology matrix construction (14). We found that all these enzymes of the luciferase family have sequence similarities with human long-chain acyl-CoA synthetase. Among these enzymes, we found that click beetle luciferase (green) is most similar to human long-chain acyl-CoA synthetase. Figure 4 shows the sequence similarity between human long-chain acyl-CoA synthetase and click beetle luciferase. Within the long-chain acyl-CoA synthetase sequence, there are two domains similar to click beetle luciferase (designated LSI and LS2 in Fig. 4A). The degrees of sequence identity of domains LSI and LS2 in the human long-chain synthetase sequence to the corresponding regions of click beetle luciferase are 25.1 and 26.2%, respectively (Fig. 4B). When conservative replacements are included in the calculation, the degrees of sequence similarity of domains LSI and LS2 to the corresponding regions of click beetle luciferase are 46.4 and 46.1%, respectively. We also detected 25.5% similarity (including conservative changes) between the N-termini of the two enzymes (residues 1-46 of human long-chain acyl-CoA synthetase sequence, Fig. 4B); however, this value seems to be not significant. Based
Flow karyotype of GM130B with H33258 600
Row karyotype of (^1308 with PrI 700
j9 -12
Human Long-Chain Acyl-CoA Synthetase: Structure and Chromosomal Location1 Takaaki Abe,' Takahiro Fujino,* Ryuichi Fukuyama," Shinsei Minoshima," Nobuyoshi Shimizu,** Hiroyuki Toh,"* Hiroyuki Suzuki,' and Tokuo Yamamoto*2 'The Tohoku University Gene Research Center, Aoba-ku, Sendai, Miyagi 981; "The Department of Molecular Biology, Keio University School of Medicine, Shinjuku-ku, Tokyo 160; and '"The Protein Engineering Research Institute, Suita, Osaka 565 Received for publication, September 6, 1991
A complementary DNA clone encoding the entire human long-chain acyl-CoA synthetase was isolated and the total 698-amino acid sequence was deduced. The amino acid sequence of human long-chain acyl-CoA synthetase shows 84.9% identity to that of rat long-chain acyl-CoA synthetase. The nucleotide sequences of the protein coding regions between human and rat long-chain acyl-CoA synthetase mRNAs are highly conserved (85.6%), whereas those of the 3' untranslated regions are less conserved (72%). The location of the human long-chain acyl-CoA synthetase gene was identified on chromosome 4 by spot hybridization of flow-sorted chromosomes. Computer-assisted homology search revealed a significant similarity of the enzyme with the enzymes of the luciferase family. Based on this similarity, the structure of human long-chain acyl-CoA synthetase can be divided into five domains: the N-terminus, two domains similar to those in enzymes of the luciferase family, a long gap region between the similar domains and the C-terminus.
Long-chain acyl-CoA synthetase [EC 6.2.1.3] plays a critical role in fatty acid metabolism in mammals. AcylCoAs produced by the enzyme are important intermediates not only for the degradation of fatty acid via fi- oxidation in mitochondria and peroxisomes but also for the production of glycerolipids and cholesteryl ester in microsomes. Acyl-CoAs are also important for the regulation of fatty acid biosynthesis. The activity of acetyl-CoA carboxylase, a rate-limiting enzyme in fatty acid biosynthesis, is negatively regulated by acyl-CoAs (1). In addition to fatty acid metabolism, acyl-CoAs are required for the acylation of many membrane proteins (2). Furthermore, a recent study by Bronfman et al. (3) revealed that acyl-CoAs modulate the activity of protein kinase C, a key enzyme in transmembrane signaling. In rat liver, long-chain acyl-CoAs synthetases localized in microsomes, mitochondria, and peroxisomes are identical with respect to molecular, catalytic, and immunological properties (4, 5). The mechanism by which long-chain acyl-CoAs synthetase is transported to the three different organelles is unknown, but this system provides an excellent model for studying the intracellular transportation of the protein. In our previous study, we reported the isolation and sequence analysis of a cDNA for rat long-chain acyl-CoA synthetase (6). The results showed that the enzyme contains 699-amino acid residues and a part of the enzyme is very similar to firefly luciferase. Based on the similarity of the reaction mechanism of the two enzymes, we have proposed that the firefly luciferase -like region found in rat 1 This work was supported by research grants from the Ministry of Education, Science and Culture of Japan, the Urakami Foundation, the Naito Foundation, and the Japan Heart Foundation. 2 To whom correspondence should be addressed.
Vol. I l l , No. 1, 1992
long-chain acyl-CoA synthetase may constitute a region where ATP and the carboxyl group of a fatty acid interact to form acyl-AMP. The level of the long-chain acyl-CoA synthetase mRNA is high in tissues performing active fatty acid oxidation (heart) or triglyceride synthesis (liver and adipose tissue). We have also shown that the level of the mRNA in liver is remarkably increased by feeding a lipogenic diet. In order to extend our knowledge of the structure, function, and regulation of long-chain acyl-CoA synthetase, we have isolated and characterized a cDNA encoding human long-chain acyl-CoA synthetase. With this cDNA we have determined the chromosomal location of the enzyme. Based on the similarities of structure and reaction mechanism of long-chain acyl-CoA synthetase to the enzymes of the luciferase family, we discuss the functional domains of the enzyme. MATERIALS AND METHODS Materials—All restriction enzymes, T4 DNA ligase, Escherichia coli DNA ligase, DNA polymerase I (E. coli), large fragment of E. coli DNA polymerase I, T4 DNA polymerase, terminal deoxynucleotidyl transferase, Rous associated virus 2 reverse transcriptase, mung bean nuclease, exonuclease HI, RNase H (E. coli), human placenta RNase inhibitor, and calf intestine alkaline phosphatase were purchased from Takara Shuzo (Kyoto). T7 DNA polymerase (Sequenase™) was from United States Biochemical. Human liver poly(A)+ RNA was obtained from Stratagene. The 32P-labeled nucleotides and "S-labeled nucleotides were products of either Amersham or Du PontNew England Nuclear. General Methods—Preparation of plasmid DNA, restric-
123
124 tion enzyme digestions, screening of the cDNA library, agarose gel electrophoresis, DNA blotting, and hybridization were performed by standard procedures (7). DNA probes were labeled with 3 T by random octanucleotide priming (8). To sequence the entire cDNA insert, the cDNA insert was shortened successively by exonuclease III (9) and subcloned into pUC vectors. Chain termination sequencing (20) was performed on denatured supercoiled plasmid DNA (11) with specific oligonucleotide primer and
T. Abe et al. T7 DNA polymerase. Both strands were sequenced entirely. The DNA sequence was analyzed on an NEC PC9801RA5 with programs from Software Development (Tokyo) or SciSoft. cDNA Cloning—The cDNA library was constructed from human liver poly(A)+ RNA by the method of Gubler and Hoffman (12) with slight modifications. Briefly, a synthetic primer-adapter consisting of a NotI site and a homopolymeric tract ( 5 ' C T C T A G A G G C G G C C G C T J O was used in-
-13TCAACACAGGACA -1 ATGCAAGCCCATGAGCTGnCCGGTATTnCGAATGCCAfi«a6GTTGACnCCGACAGTACGTGCGTACTCTTCC6ACCAACACGCTT 9 0 M Q A H E L F R Y F R M P E L V D F R Q Y V R T L P T H T L 3 0 ATGGGCTTCGGAGCTnTGCAGCACTCACCACCTTnGGTACGCCACSAGACCCAAACCCaGAAGCCGCCATGCGACCTCTCCATGCAG 180 N G F G A F A A L T T F W Y A T R P K P L K P P C 0 L S M Q 6 O TCACTGGAAGTGGCGGGTAeTGGTGGTGCACGAA6ATCCSCACTACTTGACASCGACGAGCCCnGGTGTATTTCTATGATGATGTCACA 270 S V E V A G S G G A R R S A L L D S D E P L V Y F Y D 0 V T 9 O ACATTATACGAAGGTTTCUGAGeGGAATACAGGTGTCAAATAATGeCCCnGTTTAGGaCTCGGAAACCAGACCAACCCTATGAATGG 360 T L Y E G F Q R G I Q V S N N G P C L G S R K P D Q P Y E M I T O CTTTCATATAAACAGGTTGCAGAAnGTCGGAGTGCATAGGCTCACCACTGATCCAGAAGeGCnCAAGACTGCCCCAGATCAGnCAn 450 L S Y K Q V A E L S E C I G S A L 1 Q K G F K T A P D Q F I 1 S 0 GGCATCTTTGCTCAAAATAGACCTGAGTG6STGAnATTGAACA«GATGCnTGCnAnCG*TGGTGATCGTTCCACTTTATGATACC 540 G 1 F A Q H R P E W V I I E Q G C F A Y S H V I V P L Y 0 T 1 8 O CTTGGAAATGAAGCaTCACGTACATAfiTCAACAAAGCTGAACTaCTaGGTTTTTGnGACAAGCCAGAGAAGGCCAAACTCnATTA 630 L G N E A 1 T Y I V N K A E L S L V F V 0 K P E K A K L L L 2 1 0 GAGGGTGTAGAAAATAAGnAATACCAGGCmAAMTCATAGTTGTCATGGATGCCTACGGCAGTGAACTGGTGGAACGAGGCCAGAGG 720 E G V E N K L I P G L K I I V V M D A Y G S E L V E R G Q R 2 4 0 TGTGGGeTGGAAGTCACCAGCATGAAGGCGATGGAGGACaGGGAAGAGCCAACAGACGGAAGCCCAAGCCTCCAGCACaGAAGATCn 810 C G V E V T S H K A H E D L G R A N R R K P K P P A P E D L 2 7 0 GCAGTAATTTGTTTCACAAGTGGAACTACAGGCAACCCCAAAGGAGCMTGGTCACTCACCGAAACATAGTGAGCGAnGnCAGCTTTT 900 A V I C F T S G T T G N P K G A H V T H R N I V S 0 C S A F 3 O 0 GTGAAAGaWCAGAGAATACAGTCAATCCnGCCCAGATGATACTTTGATATCTTTCnGCCTCTCGCCCATATGTnGAGAGAGnGTA 990 V K A T E N T V N P C P D 0 T L I S F L P L A H H F E R V V 3 3 O GAGTGTGTAATGCTGTGTCATGeAGCTAAAATCGGATTTTTCCAAGGAGATATCAGGaGaCATGGATGACCTCAAGGTGCTTCAACCC 1080 E C V H L C H G A K 1 G F F Q G D I R L L N D 0 L K V L Q P 3 6 O AaGTCnCCCCGTGGnCCAAGAaGCTGAACCG6ATGTTTGACCGAATTTTCGGACAAGCAAACACCACGCTGAAGCGATGGCTCTTG 1170 T V F P V V P R L L N R H F D R I F G q A N T T L K R W L L 3 9 0 GACTTTGCCTCCAAGAGGAAAGAAGCAGAGCnCGCAGCGeUTCATCAGAAACAACAGCCTGTGeeACCGGCTGATCnCCACAAAGTA 1260 D F A S K R K E A E L R S G I I R N K S L W D R L I F H K V 4 2 O CACTCGAGCCTGGGCGGAAGAGTCCGGCTGATGGTGACAGGAGCCGCCCCGGTGTaGCCAaGTGCTGACGnCCTCAGAGCAGCCaG 1350 Q S S L G G R V R L M V T G A A P V S A T V L T F L R A A L 4 5 O GGCrcTCAGTTrrATGAAG6ATACGGACAGACAGAGTGCACTGCCGGGTGaGCCTAACCATGCaGGAGACTGGACCGCAGGCCATGn 1440 G C q F Y E G Y G Q T E C T A G C C L T M P G D H T A G H V 4 8 0 GGGGCCCCGATGCCGTGCAATTTGATAAAACnGnGATGTGGAAGAAATGMTTACATGGCTGCC6AGGGCGAGGGCGAGGTGTGTGTG 1530 G A P H P C N L 1 K L V 0 V E E H N Y M A A E G E G E V C V 5 1 0 AAAGGGCCAAATGTATnCAGGGCTACnGAAGeACCCAGCGAAAACA6CAGAAGCTnGGACAAAGACGGCTGGTTACACACAGGGGAC 1620 K G P H V F Q G Y L K D P A K T A E A L D K D G W L H T 6 0 5 4 0 AnGGAAAATGGnACCAAATGGCACCnGAAMnATCGACC6GAAAAAeCACATATTTAAGaGeCACAAGGAGAATACATAGCCCCT 1710 1 G K H L P N G T L K 1 I D R K K H I F K L A Q G E Y I A P 5 7 O GAAAAGAnGAAMTATaACATGCGAAGTGAGCaGnGCTCAGGTGTTTGTCCACGeAGAAAGCaGCAGGCATTTCTCAnGCAAn 1800 E K I E H I Y M R S E P ' V A Q V F V H G E S L Q A F L I A I 6 O 0 GTGGTACCAGATGnGAGACAnATGnCCTGGGCCCAAAAGAGAGGATTTGAAGGGTCGTTrGAGGAAaGTGCAGAAATAAGGATGTC 1890 V V P D V E T L C S W A Q K R G F E G S F E E L C R N K 0 V 6 3 O AAAAAAGaATCCTCGAAGATATGGTGAGACTTGGeAAGGATTCTGeTCTGAAACCATTTGMCAGGTCAAAGGCATCACATTGCACCa 1980 K K A I L E O H V R L G K D S G L K P F E Q V K G I T L H P 6 6 0 GMnATTnCTATCGACMTGGCCnCTGACTCCXACAATGAAGGCGAAAAGGCCAGAGCTGCGGAACTATTTCAGGTCGCAGATAGAT 2070 E L F S I D N G L L T P T H K A K R P E L R N Y F R S Q I D 6 9 0 GACaCTAnCCACTATUAGGTTTAGTGTGAAGAAGAAA6aCAGAGGAAATGGCACAGnCCACAATCTCTTaCCTGCTGATGGCCT 2160 O L Y S T I K V * 698 TCATGnGnAAnnGAATACAGCAAGTGTAGGGAAGGAAGCGnCGTGTTrGACrTGTCCAnCGGGGnCnCTCATAeGAATGCTA 2250 GAGGAAACAGMCACCGCCnACAGTCACCTCATGnGCAGACUTGTnATGGTAATACACACTTTCCAAAATGAGCCnAAAAAnGT 2340 AAA6GG6ATAnATAAATGTGCTAAGnATTTGAGACnCCTCAGTnAAAAAGTG66TrnAAATCTTCTGTCTCCaGCTTTTCTAAT 2430 CAAfiGGenAGGACTTTGaATCTCTGAGATGTnGCTACTTGCTGCAAAnCTGCAGCTGTCTGaGaCTAAAGAGTACAGTGCACTA 25?0 GAGGSAAGTGnCCCTTTAAAAATAAGAACAACTGTCCTGGaGGAGAATaCACAAeCGGACCAGAGATCTTTTTAAATCCCTGCTACT 2610 GTCCCTTCTCACAGGCAnWCAGAACCCnCTGAnCGTAAGG6nACGAAACTCATGnCTTCTCCAGTCCCCTGTGGTTTCTGnG6 2700 AGCATAAGCTnCCAGTAAGCGGGAGGGCAGATCCAAaCAGAACCATGCAGATAAGGAGCaaGGCAAATGGGTGCTCATCAGAACGC 2790 GT6eAnCTCnTOTGGaVGAAT6CTCnGGAaCG6naCCAG6CaGATTCCCCSACTCCATCCTTTTTCAGGG6TTATTTAAAAA 2880 TCTGCCTTAGAnCTATAGTGAAGACAAGCATTTCAAGAAAGAGTTACCTGGATCAGCCATGaCAGaGTGACGCCTGAATAACTGTa 2970 ACTTTATCnCACTGMCCAaCACTCTGTGTAAAGGCCAACAGATTTTTAATGTGGTTnCATATCAAAAGATCATGnGGGAnAACT 3060 TGCCTTTTTCCCCAAAAAATAAACTCTCAGGCAAGCATTTCTTTAAA6CTAnAAGGeAeTATATACnGAGTACnATT6AAATRCACA 3150 GTAATAAGCAAATGncnATAATGCTACCTGATTTCTATGAAATGTGTTTGACAAGCCAAAAnCTAGGATGTAGAAATCTGGAAAGn 3240 CATTTCCTGGGAnCACncTCCAGGGATTTTTTAAAGnAATTTGGGAAATTAACAGCAGnCACnTAnGTGAGTCTTTGCCACATT 3330 TGACTGAAnGAGCTGTCATnGTACAXrjA/WGO\GCTGTTTTGGGGTCTGTGAGAGTACATGTATTATATA«AGCACAACAGGGCn 3420 GCACTAAAGAAnGTCAnGTAATAACACTACTTGGTAGCCTAACnMTATATGTAnCnAAnGCACAAAAAGTCAATAATTTGTCA 3510 CCnGGGGTTTTGAATGTTTGCTnAAGTGTTGGCTATnCTATGTTnATAAACCAAAACAAAATTTCCAAAAACAATGAAGGAAACCA 3600
Fig. 1. Nucleotide sequence of the cDNA corresponding to human longchain acyl-CoA gynthetase mRNA and the predicted amino acid sequence of the protein. Two potential polyadenylation signals in the 3' untranslated region are underlined. Two potential mRNA destabilizing signals in the 3' untranslated region are indicated by overlines and underlines. J. Biochem.
125
Human Long-Chain Acyl-CoA Synthetase stead of oligo(dT)12-18 to prime the synthesis of the first strand of cDNA. After the replacement synthesis of the second strand, the cDNA was cleaved with Notl and then ligated to Bluescript vector (Stratagene) which had been cleaved with Notl and EcoRV. The human long-chain acyl-CoA synthetase cDNAs were screened at low stringency with "P-labeled probe prepared from the cDNA for rat long-chain acyl-CoA synthetase (6). Screening of l x 10s clones yielded two positive clones. The longest clone, designated pHACSl, was subjected to restriction enzyme mapping and its nucleotide sequence was determined. Computer-Assisted Homology Search—Programs for weak homology detection (13) and homology matrix construction (14) were used. According to the pairwise alignments made by the program for weak homology detection and the detailed outputs of homology matrices, multiple alignments were constructed. The significance of the detected sequence similarity was analyzed by statistical testing with 100 randomized sequence pairs (25). Differences and alignment scores among the aligned sequences were calculated, considering a continuous gap as a single substitution regardless of its length (14). Chromosomal Mapping—Human metaphase chromosomes were prepared from the human B-lymphoblast line GM00130B (46,XY). These cells showed an apparently normal karyotype. Preparation, staining, and sorting of metaphase chromosomes were described previously (16, 17). Fifty thousand chromosomes of each type were sorted to a small spot on nitrocellulose filter disks and treated for a hybridization. Assignment of the chromosomes in the sorted fraction was made according to the method of Lebo et oL (18) and independently confirmed with chromosome -
specific DNA probes (16, 19). RESULTS AND DISCUSSION Isolation and Characterization of Human Long-Chain Acyl-CoA Synthetase cDNA—A near full-length cDNA for human long-chain acyl-CoA synthetase was isolated from the human liver cDNA library by cross-hybridization with the cDNA for rat long-chain acyl-CoA synthetase. Figure 1 shows the nucleotide sequence of the human long-chain acyl-CoA synthetase cDNA and deduced amino acid sequence of the protein. An initiator methionine codon (nucleotide positions 1-3) with a sequence matching Kozak's consensus sequence (20) begins an open reading frame of 2,094 bp, corresponding to 698 amino acids (Mr 78,942). The human enzyme is shorter than rat enzyme by one amino acid. The degree of sequence identity between the human and rat enzymes is 84.9%; a gap was counted as one substitution (Fig. 2). When conservative replacements are included in the calculation, the similarity of the human enzyme to the rat enzyme is 94.8%. The nucleotide sequence of the protein coding region of the human long-chain acyl-CoA synthetase mRNA is highly conserved between the two species (85.6%), whereas that of the 3' noncoding region is less conserved (72%). Two potential mRNA destabilizing signals (AUUTJA) (21) are localized at nucleotide positions 2872 and 3357, suggesting the possibility that the mRNA may undergo a rapid degradation. Chromosomal Location of the Human Long-Chain Acyl-CoA Synthetase Gene—The human long-chain acylCoA synthetase cDNA was used for spot-hybridization of flow-sorted human chromosomes from B-lymphoblast line
119
1 HACS
MQAHELFRYF(WPEL\raFRQWRTLPTNTU«FGAFAALnFWYATRPKPUPPCOLSMQSVEVA6-SSGARRSALLDSOEPLVYFYDDVTTLYEGFqRGI0VSNKGPCLGSRKPDQPYE ************************************************ • * #++*.*. *
RACS
#.+
************
*******************************
HACS RACS
VGAPMPCNYIKimEDMMYiyWffiEGEVCTKGANVFICGYLKOPARTAEALDIUKWL 481
600
600 HACS RACS
698 IWPDVETLCSWAQKreFEGSFEELCRNICDVICMILEDWRLGKOSGUCPFEQVKGITLHPELFSI^ ******* * a * * * * * * * . * * * * * * * * * * * . ********.***.•***********..************** a . * * * * * * * * * * * * * * * * . * * * * * * . IWPOVEILPSWAqWffiF^SFEELCRHiaJINWaEDMVICLGICMAGLKPFEQVKGIAVHPELFSIOHGLLTPTUAKRPELRNYFRSQIOELYSTIICI
601 699 Fig. 2. Alignment of the similar amino acid sequences of human (HACS) and rat (RACS) long-chain acyl-CoA synthetases. The two proteins are aligned to obtain maximum identity. Residues identical with those of human long-chain acyl-CoA synthetase are indicated with asterisks. Conservative substitutions are dotted. Vol. I l l , No. 1, 1992
126
T. Abe et al.
GM00130B (46,XY). Human chromosomes were separated into 16 or 17 fractions when stained with Hoechst 33258 (Fig. 3A) or propidium iodide (Fig. 3C), respectively. Spot-hybridization with flow-sorted chromosomes stained with Hoechst 33258 revealed a strong hybridization signal in fraction B, which contains chromosomes 3 and 4 (Fig. 3B). Hybridization with flow-sorted chromosomes stained with propidium iodide (Fig. 3D) revealed a strong hybridization signal in fraction c, which contains chromosome 4. These results indicate that the long-chain acyl-CoA synthetase gene is localized on chromosome 4. Amino Acid Sequence Similarity of Long-Chain AcylCoA Synthetase with the Enzymes of the Luciferase Family—In a previous paper, we showed that the sequence of amino acid residues 458 to 591 of rat long-chain acyl-CoA synthetase was similar to a part of firefly luciferase (6). Based on the similarity of the reactions catalyzed by long-chain acyl-CoA synthetase and firefly luciferase, we have proposed that the luciferase-like region of long-chain acyl-CoA synthetase might constitute a region where ATP and the carboxyl group of fatty acid interact to form acyl-AMP. Recently, Toh (22) reported that the N-terminal halves of gramicidin S synthetase 1 and tyrocidine synthetase 1 are similar to plant 4-coumarate:CoA ligase and luciferase from click beetle and firefly. Since long-chain acyl-CoA synthetase is similar to firefly luciferase, we compared the human long-chain acyl-CoA synthetase se-
quence with those of the enzymes of the luciferase family (22), luciferases from firefly (23) and click beetle (24), Bacillus brevis gramicidin S synthetase 1 (25), B. brevis tyrocidine synthetase 1 (26), and 4-coumarate:CoA ligases from parsley (27). In this analysis, we used computer programs for weak homology detection (13) and homology matrix construction (14). We found that all these enzymes of the luciferase family have sequence similarities with human long-chain acyl-CoA synthetase. Among these enzymes, we found that click beetle luciferase (green) is most similar to human long-chain acyl-CoA synthetase. Figure 4 shows the sequence similarity between human long-chain acyl-CoA synthetase and click beetle luciferase. Within the long-chain acyl-CoA synthetase sequence, there are two domains similar to click beetle luciferase (designated LSI and LS2 in Fig. 4A). The degrees of sequence identity of domains LSI and LS2 in the human long-chain synthetase sequence to the corresponding regions of click beetle luciferase are 25.1 and 26.2%, respectively (Fig. 4B). When conservative replacements are included in the calculation, the degrees of sequence similarity of domains LSI and LS2 to the corresponding regions of click beetle luciferase are 46.4 and 46.1%, respectively. We also detected 25.5% similarity (including conservative changes) between the N-termini of the two enzymes (residues 1-46 of human long-chain acyl-CoA synthetase sequence, Fig. 4B); however, this value seems to be not significant. Based
Flow karyotype of GM130B with H33258 600
Row karyotype of (^1308 with PrI 700
j9 -12
