Research Accomplishment Reports 2010

Unraveling the Catalytic Specificity of Terpene Hydroxylases and Engineering Sesquiterpene Hydroxylation in Plants

J. Chappell
Department of Plant and Soil Sciences

 

Non-Technical Summary

Terpenes, which represent a complex array of chemical compounds that are essential for many aspects of plant growth and development, are of considerable importance for their nutritional contributions to animals and humans, and are an important source of natural products used in agriculture and medicine. Hence, it is not surprising that the biochemistry and molecular biology of terpene biosynthesis has been intensively studied. Nonetheless, our appreciation for many of the mechanistic features of these enzymes involved in these biosynthetic pathways is still very limited.

The current application addresses this drawback by focusing on one particular class of enzymes, terpene hydroxylases, and even more narrowly, to the family of sesquiterpene hydroxylases. These enzymes decorate terpene hydrocarbon skeletons with one or several hydroxyl substituents in very specific patterns that impart biological activities to the sesquiterpene products and serve as handles for further in vivo modifications. Recent studies suggest that the specificity for these biosynthetic reactions reside within specific regions of the enzymes themselves.

Hence, the first objective of the current application is to more precisely define the structural elements of sesquiterpene hydroxylases that regulate and control their catalytic activities. We propose to accomplish this by interconverting the catalytic specificity of one sesquiterpene hydroxylase into that of a closely related hydroxylase based on an iterative and rational mutagenesis program designed upon a combination of structural and molecular comparisons between the two enzymes. Ultimately, we aim to create new plant traits, like enhanced insect and disease resistance, and the biosynthesis of high-value natural products in plants based upon the manipulation of terpene metabolism in plants.

Our second objective, therefore, will further recent advances in the genetic engineering of sesquiterpene metabolism by comparing strategies for introducing expression of unique terpene hydroxylases in transgenic plants, and evaluating these plants for the biosynthesis of new, biologically active sesquiterpenes.

2010 Project Description

The long-term goals of this project are to determine the contributions of sesquiterpenes to plant growth and development and to apply this knowledge toward engineering plants that have increased production of sesquiterpenoids important for agricultural, medical, and other industrial applications.

As one approach toward achieving these goals, we aim to understand sesquiterpene biosynthesis by identifying the residues and peptide regions that control sesquiterpene hydroxylase specificity. And for this purpose we have focused our attention to two particular cytochrome P450 enzymes, epi-aristolochene hydroxylase (EAH) and premnaspirodiene oxygenase from Hyoscyamus muticus (HPO). Both hydroxylases rely on native sesquiterpene scaffolds generated by terpene synthases. EAH catalyzes the successive hydroxylation of 5-epi-aristolochene, first at the C1 position followed by the second hydroxylation at the C3 position generating capsidiol. In contrast, HPO catalyzes the successive hydroxylation at the C4 position of premnaspirodiene to yield the ketone solavetivone. Both capsidiol and solavetivone possess antimicrobial activities and, because their production in planta is pathogen inducible, they are considered to be phytoalexins. EAH and HPO are 81% identical at the level of their amino acid sequence comparison and we supposed one means of defining their catalytic specificities would be to interconvert one into the other. For instance, what amino acids of HPO need to be mutated to convert its catalytic specificity to that of EAH.

More specifically, could we make reciprocal mutants in HPO, change particular amino acids to those found in EAH, and observe a change in HPO to an EAH like enzyme? We succeeded in this endeavor and made the following key observations. Combined mutations V366S (valine to serine mutation at position 366 of HPO), V482I and A484I were sufficient to change the regio-specificity of the initial hydroxylation from C2 of aristolochene to C1. Further mutations of I294V and F296V improved overall catalytic efficiency at C1 about 2-fold. However, no successive hydroxylation at C3 was observed in these in vitro reactions. Additional mutations V109I and H238L did however uncover the capacity of the mutant HPO to synthesize capsidiol in addition to 1β-hydroxyl aristolochene. The mutant HPO activity however was not an exact replication of authentic EAH activity. Although the mutant HPO was about 2 times more active than EAH, it only generated about 10% of its reaction product as capsidiol while the wild type EAH generates 75-80% of its reaction products as capsidiol.

2010 Impact

During the last year, we have focused on 2 efforts. The first was to perform the reciprocal mutagenesis of EAH to convert it to an HPO-like enzyme. For this reason, we created mutations in many of the analogous sites as for the HPO to EAH conversion work described above, as well as sites identified by examining molecular models of the EAH and HPO enzymes. Analogous to position F296 in HPO, V298 when changed to phenylalanine (F) in EAH greatly enhanced overall catalytic activity, as did mutants S368V, I484V and S368V, I486A. In the earlier mutagenesis work with HPO, positions analogous to 368, 484 and 486 were more associated with the regio-specificity of the enzyme rather than overall catalytic activity. When additional residues at positions 52, 107, 109, and 209 of EAH were mutated to the corresponding residues found in HPO, the mutant enzymes appeared to take on more of the successive nature of catalysis and specifically for generation of the ketone product solavetivone. These particular mutants were selected because of the prior work with the corresponding positions in HPO and because of their putative position within the molecular models of EAH. Confirmation of a role for these residues in the successive hydroxylation activity of the EAH mutant was provided by examining their catalytic activity against alternative substrates. For instance, mutants D107E; I109V, E209G; E52L, E209G; or E209G in combination with S368V and S482V all exhibited significant successive hydroxylation of 1β-hydroxy-epi-aristolochene to its ketone form.

Our second major effort over the last year has been focused on the isolation of yet other sesquiterpene hydroxylases that exhibit regio-specificities different than EAH or HPO, and in particular those associated with parthenolide biosynthesis.

It is our hope that the identification of hydroxylases which hydroxylate at different positions around the sesquiterpene scaffolds can be used in additional work to further improve our understanding of what structural features of these enzymes control regio-specificity. And if that were possible, then perhaps we would be able to predictably create hydroxylases via mutagenesis that decorated sesquiterpene scaffolds at particular carbon positions, and thus create novel molecules.

Our efforts have focused on the hydroxylases associated with parthenolide biosynthesis Magnolia grandiflora. RNA was isolated from young developing leaves and subject to 454 pyrosequencing. Over 270,000 reads yielded approximately 58 million bp with an average read of 215 bp. About one-third of the sequences assembled into contigs, which were depositing into a NCBI blast server platform. The sequence information was then screened for possible hydroxylase like genes using the EAH and HPO gene sequences, along with several other well described terpene hydroxylating enzymes. Three candidate genes have been identified and are slated for functional screening of substrate and reaction product specificity after expression in our yeast expression system.

2010 Publications

Faraldos, J.A., Wu, S., Chappell, J. and Coates R.M. (2010) Doubly Deuterium-Labeled Patchouli Alcohol from Cyclization of Singly Labeled [2-(2)H(1)]Farnesyl Diphosphate Catalyzed by Recombinant Patchoulol Synthase. J Am Chem Soc. 132:2998-3008