Structure-function relationships in wax producing enzymes

Abstract

Wax esters (WEs) are esters of fatty acids and fatty alcohols. They can cover a broad range of physical properties, which makes them especially interesting for industrial applications, including additives in cosmetics and high class lubricants. Biosynthesis of WEs is a widespread feature in nature and is carried out by two essential classes of enzymes. One of them is the class of fatty acyl reductases (FARs), which reduce acyl-CoAs or acyl-acyl carrier proteins (acyl-ACPs) to the corresponding fatty alcohols. The second one is the class of wax synthases (WSs), which esterify fatty alcohols with acyl CoAs, yielding WEs. The majority of FARs and WSs described to date exhibit a broad substrate range, resulting in wax blends of heterogeneous compositions. With respect to a commercial production of WEs in genetically modified plants, a defined WE blend is desired rather than a mixture of WEs. Enzymes with improved substrate specificities, tailored for the production of individual WE blends, are one way to overcome these issues. However, the required knowledge about structure-function relationships in FARs and WSs for the construction of respective enzymes is lacking to date. The aim of the present thesis was thus to elucidate structural determinants of substrate specificity in FARs and WSs. To date, crystal structures are neither available from FARs nor from WSs. Hence, four FARs (DmFAR1 from Drosophila melanogaster, MaFAR1 from Marinobacter aquaeolei, MmFAR1 from Mus musculus and TcFAR1 from Tribolium castaneum), two WSs (AbWSD1 from Acinetobacter baylyi and MmAWAT2 from Mus musculus) and a soluble diacylglycerol O-acyltransferase (AtDGAT3 from Arabidopsis thaliana) were studied in order to obtain structural insights. The activity of DmFAR1 was first described in the present work. It produces tetracosanol and hexacosanol upon expression in yeast, while expression in E. coli yielded tetradecanol, hexadecanol, hexadecenol and octadecenol. DmFAR1, MaFAR1, MmFAR1, AbWSD1, MmAWAT2 and AtDGAT3 were expressed in and purified from E. coli. Although this resulted in aggregated proteins in most cases, AbWSD1 was obtained in a quaternary structure corresponding to a trimer, when expressed as a fusion protein with the bacterial trigger factor. The respective fusion protein reproducibly formed crystals, which diffracted to 2.1 Å. Attempts to soak the crystals with iodine led to disruption of the crystals. Hence, the respective selenoprotein was produced and applied to crystallisation screens, which were ongoing at the end of this thesis. In order to elucidate substrate specificity determining structures in MmAWAT2, comparative studies of MmAWAT2 and MmDGAT2, which share ~ 70 % homology, were carried out. Both enzymes are acyltransferases and are capable of synthesising both, TAGs and WEs. Furthermore, both enzymes show distinct substrate specificities with respect to WE synthesis. Two predicted, neighboured hairpin forming transmembrane (TM) domains were identified to have an influence on the substrate specificity of MmAWAT2. Chimeric enzyme variants of MmAWAT2 carrying the respective section of the MmDGAT2 sequence showed a severely altered acyl chain incorporation pattern into WEs as compared to both, MmAWAT2 and MmDGAT2. Furthermore, respective variants showed an altered ratio of produced WEs and TAGs. This phenotype was also exhibited by the MmAWAT2 single amino acid exchange variant N36R, carrying a mutation in the part of the sequence which encodes the two predicted TM domains. Thus, the predicted TM domains of MmAWAT2 seem to have a role in substrate specificity determination of the enzyme. This work provides further insights into structure function relationships concerning substrate specificity in DGAT2-type acyltransferases. Furthermore, the successful crystallisation of a WS might pave the way for an extensive comprehension of this class of enzymes.