Characterization of Recombinant Long-Chain Rat Acyl-CoA Synthetase Isoforms 3 and 6: Identification of a Novel Variant of Isoform 6 Cynthia G. Van Horn, ‡ Jorge M. Caviglia, § Lei O. Li, Shuli Wang, Deborah A. Granger, and Rosalind A. Coleman* Department of Nutrition, UniVersity of North Carolina, Chapel Hill, North Carolina 27599 ReceiVed October 26, 2004 ABSTRACT: The metabolism of long-chain fatty acids in brain and their incorporation into signaling molecules such as diacylglycerol and LPA and into structural components of membranes, including myelin, requires activation by long-chain acyl-CoA synthetase (ACSL). Because ACSL3 and ACSL6 are the predominant ACSL isoforms in brain, we cloned and characterized these isoforms from rat brain and identified a novel ACSL6 clone (ACSL6_v2). ACSL6_v2 and the previously reported ACSL6_v1 represent splice variants that include exon 13 or 14, respectively. Homologue sequences of both of these variants are present in the human and mouse databases. ACSL3, ACSL6_v1, and ACSL6_v2 with Flag-epitopes at the C-termini were expressed in Escherichia coli and purified on Flag-affinity columns. The three recombinant proteins were characterized. Compared to ACSL4, another brain isoform, ACSL3, ACSL6_v1, and ACSL6_v2 showed similarities in kinetic values for CoA, palmitate, and arachidonate, but their apparent K m values for oleate were 4- to 6-fold lower than for ACSL4. In a direct competition assay with palmitate, all the polyunsaturated fatty acids tested were strong competitors only for ACSL4 with IC 50 values of 0.5 to 5 µM. DHA was also strongly preferred by ACSL6_v2. The apparent K m value for ATP of ACSL6_v1 was 8-fold higher than that of ACSL6_v2. ACSL3 and the two variants of ACSL6 were more resistant than ACSL4 to heat inactivation. Despite the high amino acid identity between ACSL3 and ACSL4, rosiglitazone inhibited only ACSL4. Triacsin C, an inhibitor of ACSL1 and ACSL4, also inhibited ACSL3, but did not inhibit the ACSL6 variants. These data further document important differences in the closely related ACSL isoforms and show that amino acid changes near the consensus nucleotide binding site alter function in the two splice variants of ACSL6. Acyl-CoA synthetase catalyzes the formation of fatty acyl- CoA’s from ATP, CoA, and long-chain fatty acids. After this essential activation step, the acyl-CoA’s enter multiple degradative pathways including -oxidation and fatty acid retroconversion, or multiple synthetic pathways including glycerolipid synthesis, phospholipid reacylation, and cho- lesterol ester formation (3, 4). Five rat ACSL 1 isoforms, each encoded by a separate gene, have been described. These isoforms differ in fatty acid preference, subcellular location (5), and regulation (6). We previously characterized the liver isoforms, ACSL1, ACSL4, and ACSL5, and found that triacsin C inhibits only ACSL1 and ACSL4 and that thiazolidinediones inhibit ACSL4 exclusively (7). In human fibroblasts, triacsin C inhibits de novo PL and TAG synthesis 83% and 93%, respectively, but does not impair phospholipid reacylation (8). In primary hepatocytes triacsin C inhibits the de novo synthesis of triacylglycerol 73%, but inhibits -oxidation by only 33% (9). Similarly in primary hepato- cytes, troglitazone inhibits oleate incorporation into triacylg- lycerol and oxidation products 50% and 20%, respectively (10). These indirect data suggest that the individual ACSL isoforms channel fatty acids toward separate metabolic fates and, more specifically, that ACSL1 and ACSL4 are linked to de novo glycerolipid synthesis. We did not include rat ACSL3 or ACSL6 (formerly ACS2 2 ) in our previous study (7) because neither had been identified in liver. However, ACSL6 and ACSL3 are the predominant isoforms in rat brain (11, 12) and may play roles in regulating brain fatty acid metabolism similar to those proposed for liver. Characterization of the brain ACSL isoforms, particularly their susceptibility to specific inhibitors, could well be important for studies of neuronal lipid metabolism (13) and contribute to our understanding of acyl- CoA channeling in Alzheimer disease, depression, and alcoholism (14). In this study we characterize three ACSL isoforms that we cloned from rat brain: ACSL3 and two splice variants of ACSL6, ACSL6_v1 and ACSL6_v2. * To whom correspondence should be addressed. Tel: 919-966-7213. Fax: 919-966-7216. E-mail: rcoleman@unc.edu. ‡ Current address: Department of Biochemistry, Rm. 239 Nutrition Research Center, Wake Forest University Health Sciences Medical Center Blvd, Winston-Salem, NC 27157. § Current address: Department of Nutritional Sciences, 96 Lipman Drive, Rutgers, The State University of New Jersey, New Brunswick, NJ 08901. 1 Abbreviations: ACSL, acyl-CoA synthetase; DHA, docosahexaeno- ic acid; EPA, eicosapentaenoic acid; FACL, fatty acyl-CoA ligase; mLACS, mouse acyl-CoA synthetase; NEM, N-ethylmaleimide; PHAX, phytanoyl-CoA R-hydroxylase; PL, phospholipid; TAG, triacylglycerol. 2 The nomenclature for the acyl-CoA synthetases has been changed. Rat ACS and human FACL, previously termed long-chain acyl-CoA synthetase or fatty acid CoA ligase, are now both termed ACSL. The numbering system has been changed so that rat ACS1, human FACL1, and human FACL2 are all now ACSL1, and the former rat ACS2 and human FACL6 are both ACSL6 (see http://www.gene.ucl.ac.uk/ nomenclature/genefamily/acs.html). 1635 Biochemistry 2005, 44, 1635-1642 10.1021/bi047721l CCC: $30.25 © 2005 American Chemical Society Published on Web 01/11/2005