DfnAEnoylReductase

DfnA is a protein that is involved in the biosynthesis of difficidin in the bacteria Bacillus amyloliquefaciens (1). The enoyl reductase (ER) domain of DfnA (PDB ID: 4CW5) catalyzes one step of the fatty acid biosynthesis cycle, a process that is essential to living organisms. The reductase functionality of DfnA ER relies upon its structural components including a TIM barrel and a flavin mononucleotide (FMN) ligand (2). DfnA ER is of particular interest to researchers concerned with the evolutionary history of fatty acid synthases (FAS) complexes such as giant fungal FAS (1). The structure and function of DfnA ER domain in Bacillus amyloliquefaciens is comparable to ER domains in other organisms such as the fungus Thermomyces lanuginosus and the bacteria Streptococcus pneumoniae.

Fatty acid biosynthesis is a metabolic pathway that converts simple precursors into long fatty acid chains. The six steps in the cycle are consistent across all kingdoms and require six distinct functional enzymes: acetyltransferase, malonyl transferase, ketoacyl synthase, ketoacyl reductase, dehydratase, and enoyl reductase (3). In most eukaryotic organisms, these enzyme domains are clustered together to form large multifunctional complexes called type I FASs (1). In bacteria and plants, monofunctional enzymes classified as type II FASs perform each step separately (1). Bacillus amyloliquefaciens is a bacterium and as such, its fatty acid biosynthesis occurs on type II enzymes including the ER domain of DfnA. ER domains catalyze the reduction of an enoyl-ACP (acyl-carrier-protein) to a saturated acyl-ACP in the sixth step of fatty acid biosynthesis (4). The final product of this iterative process is a fatty acid of 16 or 18 carbons (1).

Despite the fact that all ER domains perform the same chemical step in fatty acid biosynthesis, this functionality is achieved in diverse ways as exemplified by the wide array of ER structures. The DfnA ER domain is a member of just one family of ERs, whose structures are characterized by a TIM barrel and a critical FMN ligand (2). Other families of ER enzymes use different mechanisms and have diverse structural components including Rossman folds, NADH cofactors, and metal ions (2). Some organisms have a single ER domain, while others have multiple ER domains (4). This diversity is significant for clinical research on antibacterial medications that function as ER inhibitors because their utility would depend on their efficacy across a broad range of bacterial ER domains (4).

The DfnA ER domain is a dimer of two identical subunits, each composed of 454 residues (5). Each subunit is characterized by a TIM barrel, which consists of 8 α-helices and 8 parallel β-strands (1). The two subunits dimerize over a 1920 Å² interface that is formed in part by the TIM barrel (1). Adjacent to the TIM barrel is the substrate-binding domain: this is where the reduction of enoyl-ACP to acyl-ACP occurs (1). The FMN cofactor bound to the TIM barrel participates in the reaction by transferring electrons (1). Also located at the substrate-binding domain is an inserted α-helix (1). Other components of the secondary structure of the DfnA ER domain include an additional 4 β-strands, 9 α-helices, 3 3/10 helices, and random coils. The ER domain of DfnA has a molecular weight of 49861.17 Da and its isoelectric point is 7.85 (6).

Bioinformatics is a useful method of quantifying the similarities between protein structures. Two online resources to obtain data on proteins similar to a query are PSI-BLAST and the Dali Server. PSI-BLAST is used to identify proteins with similar primary structures to a protein of interest. The results of a BLAST search an E value for each similar protein. E values are calculated using sequence homology and a low E value indicates high sequence homology. The Dali Server is used to identify proteins with similar tertiary structures to a protein of interest. The results include a Z-score for each protein that is similar to the query. The Z-score is calculated by comparing intramolecular distances in a sum-of-pairs method. High Z-score are indicative of similar tertiary structures.

The fungal FAS from Thermomyces lanuginosus (PDB ID: 4V58) consists of two chains, each including six subunits, that perform the six enzymatic reactions required in fatty acid biosynthesis (3). While fungal FAS is a type I multienzyme complex and the DfnA ER domain is a monofunctional type II enzyme, comparisons can be made between DfnA ER and the ER subunit of the fungal FAS. A search on the Dali Server provided a Z-score of 26.3 for the comparison of the ER domain of DfnA to the ER subunit of the fungal FAS (7). A Z-score above 2 indicates that the two proteins have tertiary structure similarity. However, a BLAST comparison of these two structures led to no significant result (8). This data suggests that while the two proteins have highly similar tertiary sequences, their primary sequences diverge. The fungal ER domain includes a TIM barrel with an α-helical insertion, as well as an FMN cofactor (3). These features of the substrate-binding region align with that of the DfnA ER domain. In addition, information about the reaction that takes place in the fungal ER domain is better-studied than for DfnA. In the fungal ER, NADP+ binds between the α-helical insert and the TIM barrel during the reduction (3). It is hypothesized that during the first step of the reaction, NADP+ is oxidized and released, and in the second step, the enoyl-ACP substrate binds and is reduced (3). The tertiary structural similarities between the DfnA ER and the fungal ER present the possibility that they follow the same mechanism. As DfnA performs the same enoyl reduction process as fungal ER, it is possible that DfnA ER follows the same two-step reaction. However, neither structural nor functional similarity dictates that the mechanism has to be identical, so further research into the mechanistic process at the ER domain of DfnA must be pursued in order to definitively characterize the roles of particular amino acid residues, the TIM barrel, and the FMN ligand.

The DfnA ER domain is interesting to researchers concerned with the evolutionary history of fungal FAS. The catalytic domains of the fungal FAS complex derived from the type I bacterial fatty acid synthases over the course of evolution (1). But, the evolutionary source of the scaffolding matrix that is found in fungal FAS is unknown. Scaffolding elements in the FAS facilitate the interactions between catalytic domains in the complex (1). Evidence suggests that the DfnA ER domain and fungal FAS share an evolutionary ancestor. Comparison of the crystal structures of the ER domains of DfnA and the scaffolding in fungal FAS reveals significant similarities that are not seen in other ER domains (1).

The enoyl-ACP reductase (FabK) from the bacteria Streptococcus pneumoniae (PDB ID: 2Z6J) is a type II monofunctional enzyme that reduces enoyl-ACP to acyl-ACP in the fatty acid biosynthesis pathway (9). A BLAST search produced a value of 0.62 for the comparison of the primary structures of DfnA ER and FabK ER (8). An E value below 0.05 indicates high sequence homology between proteins, so this E value indicates that the primary structures are not highly similar. A search using the Dali Server to compare DfnA ER to FabK ER provided a Z score of 32.3 (7). This result indicates that the tertiary structures are quite similar between these two proteins. The FabK ER is a dimer of two identical subunits (9). Each subunit contains a TIM barrel and has an FMN cofactor, but does not have an α-helical insertion in the substrate-binding domain (9). In addition, the FabK ER has three other ligand types: calcium ion, (4s)-2-methyl-2,4-pentanediol, and a modified acetic acid (10). The FabK ER domain is very similar to DfnA ER as they both are homodimers that contain a TIM barrel and FMN at the substrate-binding site. However, FabK ER does not have an α-helical insertion in that region and it has multiple ligands that DfnA does not have. It is possible that these two bacterial type I FAS enzymes diverged from a common ancestral enzyme with a TIM barrel and FMN, and over the course of evolution, DfnA ER acquired an α-helical insertion and FabK ER acquired other ligands. The comparison of these ER domains suggests that the TIM barrel and FMN cofactor are essential to the catalytic reduction in the fatty acid biosynthesis cycle. It is possible that the α-helical insertion in DfnA ER and the additional ligands in FabK ER are not necessary for the enoyl reductase functionality, because these components are not found in both proteins. The FabK ER catalytic reaction depends on NADP+, however the detailed mechanism has not been determined (9). Further studies into both the DfnA ER domain and the FabK ER domain could reveal the similarities and differences between the reduction mechanisms.

The ER domain of DfnA in Bacillus amyloliquefaciens is characterized by a TIM barrel, inserted α-helix, and FMN ligand at the substrate-binding site. Further research into the mechanism would reveal the roles of each structural component in fatty acid biosynthesis. This domain has a highly similar tertiary structure to that of the fatty acid synthase from Thermomyces lanuginosus as well as that of the ER domain from Streptococcus pneumoniae, as revealed by BLAST and Dali Server searches.