Acetyl-CoA Carboxylase
Created by Reetu Mukherji
Acetyl-Coenzyme A carboxylase (ACC) is a biotin-dependent enzyme that catalyzes the first rate-limiting step in the biosynthesis of fatty acids(1). In yeast, the enzyme ACC is composed of multiple domains: the biotin carboxylase domain (BC), the biotin carboxyl carrier protein (BCCP), and the carboxyltransferase (CT) domain. The CT domain is responsible for catalyzing the carboxylation of acetyl-coenzyme A (CoA) and thereby converting it to malonyl-CoA, an important precursor of long-chain fatty acids.
The ACC CT domain (PDB ID:1od2) and its crystal growth structures were studied extensively in the model organism Saccharomyces cerevisiae (1). The CT domain is a homodimer, composed of an A and B chain each 805 amino acids long(6). Each monomer has a molecular weight of 90,944.70 Da and an isoelectric point at pH 5.84(6). In the neighboring BCCP domain, a biotin prosthetic group, linked to a lysine residue, undergoes adenosine triphosphate (ATP) dependent carboxylation catalyzed by the BC domain of the ACC(1). The CT domain catalyzes the irreversible transfer of biotin's carboxyl group to an acetyl-CoA molecule at the CT domain's active site, thus catalyzing the formation of malonyl-CoA from CO2 and acetyl-CoA(1). The resulting malonyl-CoA is used as a regulator in beta-oxidation of fatty acids and as a precursor in lipogensis, the cellular process in which energy from sugars is stored in fats.
The ACC CT domain shares similar three-dimensional structures with other enzymes involved in carboxylation reactions and metabolism pathways. On the Dali server, a Z score of 37.8 and an rmsd of 2.5 Å (APPENDIX E) suggested that methylmalonyl-CoA carboxyltransferase's 12S subunit, transcarboxylase 12S (PDB ID: 1on3), shares similar three-dimensional structure with the ACC CT domain(5). Transcarboxylase (TC) 12 catalyzes the two-step reversible transfer of a carboxyl group from methylmalonyl-CoA (MMCoA) to pyruvate, via a biotin intermediate, producing propanoyl-CoA and oxaloacetate which are metabolized via gluconneogensis and the Calvin cycle. TS 12 specifically catalyzes the transfer of a carboxyl group from MMCoA to biotin on TC 1.3S(2). The CT domain and TC 12S share biotin dependent carboxylase activity and are characteristic of the crotonase/ ClpP superfamily, which includes many acyl-CoA enzymes that catalyze reactions for fatty acid synthesis(3). The monomers of the hexameric core in TC 12S have backbone folds between the N and C domains, similar to those between the domains in each CT monomer, and share 18% sequence identity with the CT domain(2). TC 12S and the CT domain also both have adenine-binding active sites for acetyl Co-A and MMCoA respectively between beta sheets. However, they also differ significantly. For example, the phosphate interaction between the adenine base of the CoA and the enzyme is between Arg-1731, Lys-2034, and Arg-2036 on the CT domain but between Arg-35 and Gln-42 in TC 12S.
A Z score of 20.0 and a rmsd of 3.8 Å indicates that the CT domain of ACC and the glutaconyl-CoA decarboxylase alpha subunit (PDB ID:1pix) (Gcd ) also share three-dimensional structural similarities(5). Gcd , an ion pump, transfers a carboxyl group from glutaconyl-CoA to a biotin and converts glutaconyl-CoA to crotonyl-CoA. The released energy pumps sodium ions from the cytoplasm to the periplasm of cells. Gcd also consists of a dimer with the crotonase-like backbone-fold and N and C monomer domains seen in the CT domain of ACC and TC 12S structures. Similarly, the active site also lies between the two monomers. Despite their roles in different pathways, ACC CT domain, Gcd , and TC 12S share a similar immediate role of carboxylating specific coenzymes via a biotin dependent reaction. This shared function can be partially attributed to the enzymes' structural similarities.
The structure of the CT domain of Acetyl Co-A Carboxylase (PDB ID: 1od2) allows it to catalyze the carboxylation of acetyl Co-A to produce malonyl Co-A, a vital precursor of fatty acids and a regulator of fatty acid oxidation(7). The physiological oligomeric state of the CT domain is a homodimer. Each monomer subunit consists of an N and C domain, and as the two monomers align side by side to form the dimer, the N domain of each monomer is aligned with the C domain of the other monomer(1). Although there is no sequence conservation between the sub-domains, the N and C domains share a backbone fold with a central beta-beta-alpha super-helix characteristic of those seen in the crotonase/ClpP superfamily. Members of this superfamily are similarly involved in essential fatty acid synthesis and metabolism pathways. Dimerization is facilitated by the placement of an alpha helix group of the C domain, alpha6A- alpha6D, between the core and beta sheet group, Beta7A-Beta7D, of the N domain of the other monomer(1). These groups of alpha helices and beta sheets, known as "insertions," provide binding surfaces that allow oligomerization(1). Although the monomers and sub-domains do not have individual functions, when they dimerize, an important active site forms at the interface of the monomers between one of the interacting N and C domains.
Each monomer consists of an 805 amino acid sequence. The secondary structure of the CT domain consists of beta-sheets, alpha-helices, random coils, and turns. The N domain of each monomer consists of eight alpha-helices and helices groups and 13 beta-sheets and sheet groups(1). The C domain consists of eight alpha-helices and helices groups and 12 beta-sheets or beta sheet groups(1). Random coils and beta turns form among the alpha-helices and beta-sheets throughout each monomer. The CT domain consists of two active sites. The substrate, acetyl Co-A, binds to an active site which lies in between the beta-sheets (Beta5, Beta7, Beta9, and Beta11) of the beta-beta-alpha superhelix of the C domain of one monomer and the beta-beta-alpha superhelix of the N domain of the other monomer. Alpha6 helices from each domain form two walls around the active site that provide additional binding and stabilizing surfaces for acetyl Co-A.
The acetyl Co-A substrate binds to an active site mostly associated with the N domain at the dimer interface. The N1 and N6 atoms of acetyl Co-A's adenine base bind to the active site by recognizing and forming hydrogen bonds with a chain of residues following Beta7 in the N domain(5). The phosphate groups of the Co-A are positioned near Arg-1731 on the N domain and Lys-2034 and Arg-2036 on the C domain, and most likely form hydrogen bonds with their side chains. Out of the three residues, Arg-1731, whose amine group hydrogen-bonds with acetyl Co-A's phosphate group, plays the most important role at the enzyme's active site, since its mutation yields in a 14 fold increase in the Km of the substrate binding(1). The Co-A's thiol group is positioned in the cavity between the two domains and its panthotheine arm rests near the surface of a small beta-sheet of the N domain. Minor and insignificant conformational changes occur when Co-A binds to the enzyme(1).The other active binding site of CT, on the other N domain on the different monomer, binds to an adenine base in a similar way. However, electron density observation of the adenine base alone suggests that the acetyl Co-A molecule is disfigured at this active site.
A prosthetic carboxybiotin group, transferred from the biotin carboxyl carrier protein (BCCP) domain of an ACC, associates with a small beta-sheet on the C domain across from the N domain with the acetyl Co-A(5). Arg-1954 in the C domain recognizes the carboxyl group of carboxybiotin, most likely due to hydrogen bonding between the side chain amine of Arg and the carboxyl group of carboxybiotin. Carboxybiotin's role is to provide a carboxyl group for the CT domain to transfer to the acetyl Co-A in the active site, thus producing malonyl Co-A which is used as a precursor in fatty acid synthesis(7). Experimental results suggest the N1 atom of biotin as the necessary base to initiate the catalytic activity of the enzyme(1).
The commercial herbicides haloxyfop and diclofop inhibit ACC activity by competing with acetyl Co-A at the same active site and forming haloxyfop or diplofop complexes (PDB ID: 1uyr and PDB ID: 1uys respectfully)6. When bound, the pyridyl ring of haloxyfop lies between Tyr-1738 on the N domain and Phe1956 on the C domain, forming pi-pi interactions. Its trifluromethyl group lies over Trp-1924 and near the side chains of Val-1967, Ile-1974, and Val-2025. One of haloxyfop's carboxylate oxygen atoms hydrogen-bonds to the backbone amides of Ala-1627 and Ile-1735. Haloxyfop's methyl group forms van der Waal interactions with the side chains of Ala-1627 and Leu-1705. Upon binding, a large conformational change occurs to the original, physiological conformation where the side chains of Tyr-1738 and Phe-1956 move away from the binding pocket and become beta-stacked with the pyridyl ring of haloxyfop. This new conformation helps cover the hydrophobic core of the CT dimer interface. A trifluromethylpyridyl group is also inserted deep into the hydrophobic interface(5). Diclofop binds similarly with the CT domain with minor conformational changes that differ slightly from those found in the haloxyfop-CT domain complex. These drug-complexes are studied to eventually apply them to inhibiting human ACC's as a solution to lipid-induced insulin resistance and type 2 diabetes(3).
The CT domain and methylmalonyl-CoA (MMCoA) transcarboxylase (TC) 12S (PDB ID:1on3) share biotin dependent carboxylase activity and have similar beta-beta-alpha super-helices, tertiary structure, and crotonase-superfamily backbone folds(2). Similar to how acetyl Co-A binds to the active site in the CT domain, in TC 12S, the N1 and N6 of the MMCoA's adenine base hydrogen bond to the backbone carbonlyls of Ile-145 and Gly-141 respectfully. The Co-A phosphates of the MMCoA substrate also interacts with an argenine residue, Arg35, similar to the Arg1731 interaction seen in the CT domain. Although six monomers make up the 12S domain, the active site of the enzyme is also located between the monomers, as seen in the CT domain, with similar Co-A binding patterns(2).