AmericanChemicalSociety.com

A promising route for the generation of new sustainable biofuels is to adapt biosynthetic pathways that directly produce fuel grade hydrocarbons. Although such pathways has been known for a long time, the enzymes that comprise them have not been studied to a large extent (1,2). The cer1 gene from A .thaliana has been proposed to encode an aldehyde decarbonylase protein that can convert long chain aldehydes to long chain alkanes (1). Very recently, another enzyme from cyanobacteria has been shown to possess aldehyde decarbonylase activity towards long chain aldehydes (3). This is a very unusual reaction, and a detailed mechanistic investigation of these enzymes may facilitate engineering of the enzymes to produce shorter chain, fuel grade hydrocarbons,

Studies on Cer1 Protein

Our studies have been directed toward expression and mechanistic studies of these two enzymes. Previous genetic studies strongly implicated the cer1 gene from A. thaliana as an aldehyde decarbonylase (2). cer1 encodes a 67 kDa integral membrane protein that has an 8-His iron binding motif common to membrane-bound diiron desaturases, which include fatty acid desaturases and alkane hydroxylases (4). Attempts to overexpress the A. thaliana gene proved problematic, so a synthetic gene was designed to include optimized codons for yeast and E. coli, affinity purification using a His-tag and antibody recognition motifs. Since Cer1 is a eukaryotic membrane protein, attempts for expression were first made to express it in yeast (S. Cerevisae). Standard yeast expression vectors did not yield any detectable band in western blotting analysis, presumably due to very low expression levels. Efforts are currently being made to increase the expression levels in yeast.

Initial attempts to express cer1 gene from pET28 vector in E. coli were not successful. This might be due to the fact that E.Coli does not have the proper machinery to target Cer1 protein to the membrane for proper folding, resulting in protein degradation inside the cell. To overcome this problem, fusion protein partners were used. These fusion partners include mystic protein and maltose binding protein, which were shown to successfully target their N-terminal fusions to the membrane. Western blot analysis showed, Cer1 protein can be over-expressed at high levels in fusion to both mystic and maltose binding proteins, although presence of lower molecular weight bands indicates proteolytic degradation. Currently we are exploring assay conditions for Cer1. Since this enzyme is an 8-His motif containing diiron enzyme, we will also examine the possibility that the enzyme has a desaturase or hydroxylase activity.

Studies on ADC Protein

It was recently shown that cyanobacteria possess an enzyme that converts long chain aldehydes to alkanes and alkenes (3). The sequence similarity of this cyanobacterial aldehyde decarbonylase protein (CAD) to ribonucleotide reductases suggested that this protein is a member of the nonheme diiron superfamily. In fact, the crystal structure of an ortholog of CAD from Prochlorococcus marinus has been solved as part of a structural genomics effort without any assigned function. This is a soluble protein is similar to the R2 domain of E. coli ribonucleotide reductase, with a non-heme diiron center at the active site ligated by histidine and carboxylate residues. A synthetic gene encoding this protein was used to over-express the CAD in E. coli and it has been purified in quite high yields (~150 mg/L). CAD as purified has a broad UV absorbance at 350 nm, typical of non-heme proteins due to a histidine-to-metal charge transfer band. Initial spectroscopic studies showed that the protein is ~30% iron loaded as purified. Treatment of the protein by high concentration of EDTA did not decrease the iron content, suggesting that the iron is tightly bound. Expression in the presence of ferrous ammonium sulfate increased the iron content of the protein by ~50 % as purified. Current studies are focused on the conditions need to activate the protein for activity. Future studies will focus on the metal requirement, the possible involvement of molecular oxygen in the reaction and the substrate specificity of the enzyme.

References

1. Aarts, M. G., Keijzer, C. J., Stiekema, W. J., and Pereira, A. (1995) Molecular characterization of the CER1 gene of arabidopsis involved in epicuticular wax biosynthesis and pollen fertility, Plant Cell 7, 2115-2127.

2. Cheesbrough, T. M., and Kolattukudy, P. E. (1984) Alkane biosynthesis by decarbonylation of aldehydes catalyzed by a particulate preparation from Pisum sativum, Proc Natl Acad Sci U S A 81, 6613-6617.

3. Schirmer, A., Rude, M. A., Li, X., Popova, E., and del Cardayre, S. B. Microbial biosynthesis of alkanes, Science 329, 559-562.

4. Shanklin, J., and Cahoon, E. B. (1998) Desaturation and Related Modifications of Fatty Acids1, Annu Rev Plant Physiol Plant Mol Biol 49, 611-641