Plant sesquiterpene synthases catalyze the conversion of the linear substrate farnesyl diphosphate, FPP, into a remarkable array of secondary metabolites. These secondary metabolites in turn mediate a number of important interactions between plants and their environment, such as plant-plant, plant-insect and plant-pathogen interactions. Given the relative biological importance of sesquiterpenes and their use in numerous practical applications, the current thesis was directed towards developing a better understanding of the mechanisms employed by sesquiterpene synthases in the biosynthesis of such a diverse class of compounds. Substrate preference for sesquiterpene synthases initially isolated from Nicotiana tabacum (TEAS), Hyoscyamus muticus (HPS) and Artemisia annuna (ADS) were optimized with regards to a divalent metal ion requirement. Surprisingly, careful titration with manganese stimulated bona fide synthase activity with the native 15-carbon substrate farnesyl diphopshate (FPP) as well as with the 10-carbon substrate geranyl diphosphate (GPP). Reaction product analysis suggested that the GPP could be used to investigate early steps in the catalytic cascade of these enzymes. To investigate how structural features of the sesquiterpene synthases translate into enzymatic traits, a series of substrate and active site residue contacts maps were developed and used in a comparative approach to identify residues that might direct product specificity. The role and contribution of several of these residues to catalysis and product specificity were subsequently tested by the creation of site-directed mutants. One series of mutants was demonstrated to change the reaction product to a novel sesquiterpene, 4-epi-eremophilene, and while another series successfully transmutated TEAS into a HPS-like enzyme. This is the first report of a rational redesign of product specificity for any terpene synthase. The contact map provides a basis for the prediction of specific configurations of amino acids that might be necessary for as yet uncharacterized sesquiterpene synthases from natural sources. This prediction was tested by the subsequent isolation and validation that valencene synthase, a synthase from citrus, did indeed have the amino acid configuration as predicted. Lastly, an in vitro system was developed for analyzing the interaction between sesquiterpene synthases and the corresponding terpene hydroxylase. Development of this in vitro system is presented as a new important tool in further defining those biochemical features giving rise to the biological diversity of sesquiterpenes.

Plants employ diverse defense mechanisms to combat attack by harmful organisms. For instance, plants produce constitutive physical barriers or use chemical compounds such as specialized secondary metabolites to resist herbivore or pathogen invasion. Considering the cost-efficiency and energy balance between defense, growth and reproduction, defense reactions in plants have to be regulated temporally and spatially. As more cost-efficient strategies, plants may induce their defense response only in the presence of the attacker or restrict constitutive defenses to specific tissues or cells. In this study, we investigated aspects of the spatial regulation and induced changes of primary and secondary metabolism in roots of Arabidopsis thaliana. Roots represent important organs for anchoring plants in the soil and taking up water and nutrients. Hence, it is assumed that roots are as well protected as aerial tissues by different defense mechanisms. The first part of this work is focused on the cell-type-specific biosynthesis of volatile terpenes in Arabidopsis roots. Terpenes are the most abundant specialized metabolites in plants and play an important role in plant defense against pathogens or herbivores. Terpene biosynthetic enzyme activities are often coordinated in specific tissues and cellular compartments. Fine-scale transcriptome maps of Arabidopsis roots have shown that terpene biosynthesis is restricted to particular cell types. However, the reasons and significance of this cell-type specificity are not well understood. We hypothesized that the formation of terpene metabolites is not restricted to specific cells but can be supported by different cell types. We, therefore, probed the plasticity of the cell-specific formation of terpenes by swapping the expression of the terpene synthase (TPS) genes, TPS08, TPS13 and TPS25, between different root cell types in the respective mutant background. To investigate the ectopic expression of TPSs at different levels, quantitative real-time PCR (qRT-PCR), confocal microscopy, and gas chromatography-mass spectrometry (GC-MS) were performed. We found that terpene synthase TPS08, which produces the diterpene rhizathalene and is normally expressed in the root vascular tissue, is functionally active when expressed in the epidermis or cortex, although at substantially lower levels compared to the wild type. We did not find an obvious correlation between the volatile emission level and gene transcript level of TPS08, which may be attributed to a reduced activity of the expressed TPS08-yellow fluorescent protein (YFP) fusion protein. When expression of TPS13 (producing the sesquiterpene (Z)-"-bisabolene) was directed from the cortex to the epidermis or stele, TPS13 gene expression and (Z)-"-bisabolene formation was supported by these cell types although to varying levels in comparison to wild type. TPS13-YFP fluorescent signal driven by the epidermal WER and GL3 promoters was primarily detected at the root tip. Terpene production was also observed for the (E)-"-farnesene sesquiterpene synthase TPS25 when its expression was switched from the endodermis and non-hair producing epidermal cells to hair producing epidermal cells although only a weak fluorescent signal was detected from the expressed TPS25-mGFP protein. Together, the results provide preliminary evidence for a relaxed cell specificity of terpene biosynthesis in Arabidopsis roots and suggest that tissue-specific terpene metabolite patterns could change depending on different selective pressures in rhizosphere. In the second part of this study, we performed global gene transcript profiling and primary metabolite analysis of Arabidopsis roots upon feeding by the generalist root herbivore, Bradysia (fungus gnat). In a microarray analysis, we identified 451 of 22,810 genes that were up-regulated more than 2-fold. Gene ontology (GO) analysis showed that 26% of those genes with predicted or known functions play a role in primary or secondary metabolism, while 24% are involved in cell signaling or in responses to stimulating factors, such as jasmonic acid (JA), ethylene, wounding, and oxidative stress. At the metabolite level, we observed only marginal changes of amino acid, sugar and carboxylic acid relative levels over a time course of 4 days of Bradysia feeding. There was a trend for increased levels of amino acids and the relative levels of sucrose were increased significantly ("=0.05) at the fourth day of feeding. In conclusion, the study provided evidence for the induction of genes related to primary and secondary metabolism and stress responses in Arabidopsis roots, but showed only marginal changes at the primary metabolite level. In addition, the work indicated that the formation of terpene-specialized metabolites in Arabidopsis roots is not restricted to specific cells, but can be supported by different cell types. / Master of Science

Roots and leaves are integrated structural elements that together sustain plant growth and development. Insect herbivores pose a constant threat to both above- and belowground plant tissues. To ward off herbivorous insects, plants have developed different strategies such as direct and indirect chemical defense mechanisms. Research has primarily focused on visible aboveground interactions between plants and herbivores. Root-feeding insects, although often overlooked, play a major role in inducing physical and physiological changes in plants. However, little is known about how plants deploy chemical defense against root herbivores. We have developed an Arabidopsis aeroponic culture system based on clay granulate, which provides access to root tissue and accommodates subterranean insect herbivores. Using this system, feeding performance and plant tissue damage by the root herbivore Bradysia (fungus gnat) were evaluated. Larval feeding was found to reduce Arabidopsis root biomass and water uptake. Furthermore, we have characterized a root-specific terpene synthase AtTPS08, which is responsible for the constitutive formation of the novel volatile diterpene compound, rhizathalene, in Arabidopsis roots. Rhizathalene synthase is a class I diterpene synthase that has high affinity for the substrate geranylgeranyl diphosphate (GGPP) and is targeted to the root leucoplast. Expression of the β-glucuronidase (GUS) reporter gene fused to the upstream genomic region of AtTPS08 demonstrated constitutive promoter activity in the root vascular tissue and root tips. Using the established bioassay with Arabidopsis and Bradysia larvae, in aeroponic culture we could show that roots deficient in rhizathalene synthesis were more susceptible to herbivory. Our work provides in vivo-evidence that diterpene compounds are involved in belowground direct defense against root-feeding insects. Future work is still required to improve our understanding of plant root defense. This study has provided a basis for future investigations on the biochemistry, molecular regulation and defensive function of Arabidopsis root chemicals in interaction with both above- and belowground herbivores (and pathogens). / Ph. D.

Stinkbugs (Pentatomidae) and other agricultural pests such as bark beetles and flea beetles are known to synthesize terpenoids as aggregation pheromones. Knowledge of the genes and enzymes involved in pheromone biosynthesis may allow engineering of pheromone biosynthetic pathways in plants to develop new forms of trap crops and agricultural practices for pest management. The harlequin bug, Murgantia histrionica, a specialist pest on crucifer crops, produces the sesquiterpene, murgantiol, as a male-specific aggregation pheromone. Similarly, the southern green stink bug, Nezara viridula, a generalist pest worldwide on soybean and other crops, releases sesquiterpene cis-/trans-(Z)-α-bisabolene epoxides as male-specific aggregation pheromone. In both species, enzymes called terpene synthases (TPSs) synthesize precursors of the aggregation pheromones, which are sesquipiperitol and (Z)-α-bisabolene as the precursor of murgantiol and cis-/trans-(Z)-α-bisabolene epoxide, respectively. We hypothesized that enzymes in the family of cytochrome P450 monooxygenases are involved in the conversion of these precursors to the final epoxide products. This study investigated the tissue specificity and sequence of these conversions by performing crude enzyme assays with protein extracts from male tissues. Furthermore, candidate P450 genes were selected by RNA-sequencing and co-expression analysis and the corresponding recombinant proteins tested for enzyme activity. To engineer the pheromone biosynthetic enzymes in plants, transient expression of the TPSs of both stink bugs was performed in Nicotiana benthamiana leaves. Both sesquipiperitol and (Z)-α-bisabolene were found to be produced and emitted from inoculated N. benthamiana leaves. Future work will implement stable transformation to engineer murgantiol biosynthesis in crucifer trap crops and develop similar approaches for pheromone engineering of other agricultural pests. / Master of Science / Stinkbugs including the harlequin bug, Murgantia histrionica and southern green stinkbug, Nezara viridula, are major agricultural pests in the US and worldwide. To control these pests with alternative pest management strategies, we have proposed to develop trap crops that emit pheromones to lure the insects away from crop fields. To establish pheromone biosynthesis in plants, we investigated the corresponding enzymatic steps in both stink bugs. We show that terpene synthase enzyme from both stink bugs can be transformed into plants for the engineering of pheromones in trap crops. With identification of P450 genes in pheromone biosynthesis enhanced trap crops can be made.

Specialized metabolites have important roles as infochemicals in inter- and intraspecific interactions of insects. A particularly abundant class of specialized metabolites are terpenes, which are released by many members of taxonomically diverse insect lineages as pheromone and defense compounds. Despite the broad occurrence of terpenes in insects, knowledge of their biosynthesis remains limited compared to that in other forms of life. Terpenes are biosynthetically produced by the action of terpene synthase (TPS) enzymes. While insects lack TPS enzymes found in plants and microbes, there is growing evidence that insect TPS proteins have evolved independently from isoprenyl diphosphate synthase (IDS) enzymes in core terpene metabolism. To gain deeper insight into the transition from IDS to TPS function, I have explored the genomic and functional evolution of TPS enzymes in representatives of major insect lineages. First, I investigated evolutionary and functional relationships of TPS enzymes with roles in pheromone biosynthesis in pentatomids (stink bugs) including the invasive and economically critical pests Nezara viridula (Southern green stink bug) and Halyomorpha halys (brown marmorated stink bug). I also performed a comprehensive phylogenetic analysis of TPS genes in species across the broader order of piercing-sucking insects (Hemiptera), which provided evidence for an ancient emergence of TPS function in this group of insects. To gain a better understanding of core structural determinants of insect TPS evolution, we next defined distinct IDS catalytic motifs that are consistently substituted in enzymes with TPS function. These sequence characteristics were used to make predictions of TPS functionality in a large dataset of insect proteins. I determined the evolutionary dynamics of predicted and known TPS and IDS enzymes through extensive phylogenetic analysis to make top-level inferences about the distribution and evolution of TPS function in insects. Using this knowledge, I further explored functional transitions and subfunctionalization of TPS genes in the large order of beetles (Coleoptera), and more specifically, in species of the lady beetle family (Cocinellidae) including the globally invasive pest, Harmonia axyridis. Comparative genome analyses and IDS/TPS gene functional characterizations revealed gene duplication patterns and enzyme transitions that suggest TPS function evolved in part through processes of subfunctionalization and bifunctional enzymatic states. Additionally, this study provided the first experimental evidence for the mitochondrial localization of terpene metabolism in insects. Lastly, I identified putative TPS enzymes in the American cockroach, Periplaneta americana, and conducted an investigation into their catalytic activity. I found first evidence for TPS enzymatic activity in Blattodea as the most anciently diverging order of terpene-emitting insects and made inferences on the relationship of these enzymes to characterized IDS and TPS proteins in other insects. Our findings in the American cockroach point to the potential independent evolution of TPS function in blattodean cockroaches and termites in types of IDS ancestors. This work significantly advances our understanding of the evolution, functional diversity, and biochemical properties of TPS enzymes in insects, highlighting their recurring pattern of parallel evolution from IDS ancestors and its significance as a model for the emergence of novel specialized functions in core metabolic enzymes. / Doctor of Philosophy / Insects use many types of chemicals for purposes of communication and defense. Terpenes represent a common and diverse class of natural chemicals, which are used by insects to send pheromone signals and to protect themselves from predators. Terpenes also occur in other kingdoms of life. For example, in plants, they are especially widespread, forming a large portion of their essential oil and floral scent compounds. In contrast to plants and other organisms, not much is known about how insects produce terpenes. Terpenes are made by proteins called terpene synthase (TPS) enzymes. TPS enzymes have been traditionally associated with plants and microbes but have not been found in any insect species. Instead, there is growing evidence that insects have developed their own versions of these enzymes, known as isoprenyl diphosphate synthase (IDS)-type TPS enzymes, from proteins with essential functions in metabolism. To learn more about these unique insect enzymes, we explored their evolution and activity in species of several different groups of insects. First, we investigated TPS enzymes that are required for the biosynthesis of pheromones in stink bugs including two agriculturally important pests, the Southern green stink bug and the brown marmorated stink bug. This research showed that the ability to make terpenes might be quite ancient in this group of insects compared to TPS enzymes in other insects. Next, we examined the protein structure of insect TPS enzymes and determined features that are characteristic for these types of enzymes. This information was used to predict the occurrence of TPS proteins and their evolution across many different groups of insects. In particular, I found evidence for the emergence of TPS enzymes in lady beetles, with a focus on the invasive Asian lady beetle, which emits a terpene pheromone for aggregation. My research suggested that lady beetle TPS enzymes evolved through a process called subfunctionalization, where genes duplicate and progressively split their ancestral functions with new features to evolve novel functions. This study also provided the first evidence that insects might produce terpenes in their mitochondria, a part of the cell known for energy production. Finally, I discovered potential TPS enzymes in the American cockroach. My investigation showed that cockroaches and termites, both part of the oldest-diverging group of terpene-releasing insects, may have independently developed their own TPS enzymes from different ancestor proteins. Overall, this research helps us understand how insects produce chemical compounds important to their biology and ecology and how these abilities have evolved over time. This knowledge can be useful in agriculture, pest control, and for our understanding of insect biology.

Alstroemeria is an important cut flower and its breeding has been developed focused on aesthetic characteristics and vase life longevity, but little is known about its scent. Five different genotypes were assessed including the non scented cultivars ‘Rebecca’ and ‘Samora’ and the scented cultivars, ‘Sweet Laura’, ‘Ajax’ and the species A. caryophyllaea. The scented Alstroemerias emitted the terpenoids: isocaryophyllene and ocimene as the major floral volatile compounds. Characterization of an Alstroemeria TPS (ALSTER) was based on four ESTs previously found in A. cv ‘Rebecca’. Rapid amplification of cDNA ends (RACE) was performed and the full length ORF was used for characterizations of the genomic organization and amino acid sequences (phylogenetic analysis). ALSTER genomic region contains five introns and six exons. This unique genomic organization classified ALSTER as a member of the class III with a merged 5-6 exon. The deduced amino acid sequence was classified into the subfamily TPS-b. A functional analysis showed enzymatic activity of ALSTER with geranyl diphosphate (GPP) and the monoterpene myrcene was the only product obtained. Gene expression evaluated through real time and semi q RT-PCR on eight different stages of development (SO to S7) showed high expression of ALSTER at around S2 - S4 in the scented Alstroemerias, coinciding with high scent emission perceived and also with the maturation of reproductive organs. Evaluations through surveys focused on level of liking of floral scent, were performed finding positive correlations between floral scent liking and floral appearance liking and between floral scent liking and floral scent intensity. Finally, 17 new lines of A. caryophyllaea were evaluated in terms of their morphology, phenology and productivity. Although none of them were suitable for the market because of their low productivity, short stems and small flowers, they were all scented and identified as promising starting points for breeding purposes.

Southern Magnolia (Magnolia grandiflora) is a primitive tree species that has attracted attention because of its horticultural distinctiveness, the wealth of natural products associated with it, and its evolutionary position as a basal angiosperm. Terpenoid constituents were determined from Magnolia leaves and flowers. Magnolia leaves constitutively produced two major terpenoids, andamp;acirc;-cubebene and germacrene A. However, upon wounding Magnolia leaves biosynthesized a significant array of monoand sesquiterpenoids, including andamp;acirc;-pinene, trans-andamp;acirc;-ocimene, andamp;aacute;-gurjunene, andamp;acirc;-caryophyllene and andamp;acirc;-cubebene, along with fatty acid derivatives such as cis-jasmone, for up to 19 hours after treatment. Flowers were also examined for their emission of terpene volatiles prior to and after opening, and also in response to challenge by Japanese beetles. Opened and un-opened flowers constitutively emitted a blend of monoterpenes dominated by andamp;acirc;-pinene and cis-andamp;acirc;-ocimene. However, the emission levels of monoterpenes such as verbenone, geraniol, and citral, and sesquiterpenes such as andamp;acirc;-cubebene, andamp;aacute;-farnesene, and andamp;acirc;-caryophyllene were significantly elevated in the emissions of the beetle-challenged flowers. Three cDNAs corresponding to terpene synthase (TPS) genes expressed in young Magnolia leaves were isolated and the corresponding enzymes were functionally characterized in vitro. Recombinant Mg25 converted FPP (C15) predominantly to andamp;acirc;-cubebene, while Mg17 converted GPP (C5) to andamp;aacute;-terpineol. Efforts to functionally characterize Mg11 were unsuccessful. Transcript levels for all 3 genes were prominent in young leaf tissue and significantly elevated for Mg25 and Mg11 mRNAs in stamens. A putative N-terminal signal peptide of Mg17 targeted the reporter GFP protein to both chloroplasts and mitochondria when transiently expressed in epidermal cells of Nicotiana tabacum leaves. Phylogenetic analyses indicated that Mg25 and Mg11 belonged to the angiosperm sesquiterpene synthase subclass TPS-a, while Mg17 aligned more closely to the angiosperm monoterpene synthase subclass TPS-b. Unexpectedly, intron/exon organizations for the three Magnolia TPS genes were different from one another and from other well characterized terpene synthase gene sets. The Mg17 gene consists of 6 introns arranged in a manner similar to many other angiosperm sesquiterpene synthases, but Mg11 contains only 4 introns, and Mg25 has only a single intron near the 5 terminus of the gene. Our results suggest that much of the structural diversity observed in the Magnolia TPS genes may have occurred by means other than intron-loss from a common ancestor TPS gene. Costunolide is a sesquiterpene lactone widely recognized for its diverse biological activities, including its bitter taste in lettuces, and as a precursor to the more potent pharmacological agent parthenolide. A lettuce EST database was screened for cytochrome P450 genes that might be associated with sesquiterpene hydroxylation. Five ESTs were selected based on sequence similarity to known sesquiterpene hydroxylases and three of them (Ls7108, Ls3597 and Ls2101) were successfully amplified as fulllength cDNAs. To functionally characterize these cDNAs, they were co-expressed along with a germacrene A synthase and a cytochrome P450 reductase in yeast. Based on product profile comparisons between the three different lines to the control line, only the Ls7108-harboring line produced unique compounds. Neither of the other lines showed a new product peak. The more abundant, polar product generated by the Ls7108-containing line was purified and identified as a 12-acetoxy-germacrene by NMR analysis. In vitro studies using Ls7108 microsomal proteins did not yield the 12-acetoxy-germacrene A, but the putative germacra-1(10),4,11(13)-trien-12-ol intermediate. Catalytic activity of the Ls7108 microsomal enzyme was NADPH, pH and time dependent. Our results demonstrate that Ls7108 is a lettuce cytochrome P450 which catalyzes the hydroxylation of a methyl group of the isopropenyl substituent of germacrene A, generating germacra-1(10),4,11(13)-trien-12-ol, and that when this mono-hydroxylated sesquiterpene is synthesized in yeast, an endogenous yeast enzyme further modifies the germacrenol compound by acetylation of the alcohol group at the C-12 position.

Cette thèse s’est focalisée sur les lavandes et en particulier sur trois lavandes retrouvées en France : la lavande fine (Lavandula angustifolia Miller), la lavande aspic (Lavandula latifolia Medik) et leur hybride le lavandin (Lavandula x intermedia). Capables de synthétiser de grandes quantités de composés organiques volatils (COV) et plus particulièrement des terpènes volatils tel les mono- et sesquiterpènes, les lavandes sont utilisées depuis l’antiquité par l’Homme pour les propriétés médicinales et aromatiques de ces terpènes, composants de l’huile essentielle de lavande. De par l’importance économique et écologique des terpènes volatils, ce travail de thèse présente différents aspects d’étude. Dans un premier temps la relation entre la diversité chimique, géographique et génétique de la lavande fine a été analysée sur l’ensemble de son aire de répartition. Les résultats ont mis en évidence des populations de lavande fine très différentes autant chimiquement que génétiquement au bord de leur aire de répartition. Dans un deuxième temps, l’impact du stress hydrique au cours du temps sur les contenus en terpènes volatils a été évalué sur la lavande aspic, le lavandin et six populations de lavande fine. Les résultats ont montré des tolérances différentielles en fonction des espèces et des populations ; ainsi le lavandin est plus rapidement affecté par le stress hydrique que la lavande fine. Les contenus terpéniques n’ont été que faiblement impactés par le stress hydrique et cela à des états hydriques différents selon les espèces et les populations. Malgré la grande diversité de réponse selon les composés, l’intensité du stress hydrique et les plantes ; les terpènes de la voie du camphre (bornéol, camphène et camphre) sont ceux qui présentent les plus grandes variations entre les plantes stressées et témoins. Il apparaissait alors intéressant d’étudier la voie de biosynthèse du camphre. Dans ce cadre-là, nous avons identifié et caractérisé la bornyl diphosphate synthase capable de former le bornéol à partir du bornyl diphosphate. L’ensemble de ces travaux permettent de mieux appréhender les relations entre production de terpènes volatils et environnement ainsi que de donner des outils génétiques afin de poursuivre ces investigations / The PhD was focused on lavenders and precisely on lavenders present in France: the fine lavender (Lavandula angustifolia Miller), the spike lavender (Lavandula latifolia Medik) and their hybrid the lavandin (Lavandula x intermedia). Skilled to synthetize huge organic volatils coumpounds (COV) amount and in particular volatils terpene such mono- and sesquiterpenes, lavenders are used by human from antiquity for medicinal and aromatic properties of these volatils terpenes, lavender essential oil is composed of. Due to economical and ecological volatils terpenes importance, several study aspects is considered in the PhD. In a first hand, on all the fine lavender’s repartition area, relationship between chemical, geographical and genetical diversities was assessed. Results showed chemical and genetical significant different populations, at the border of fine lavender repartition area. In a second hand, hydric stress impact over time on volatiles terpenes content was assessed on the spike lavender, lavandin and six fine lavender populations. Results put in evidence differential tolerances by species and populations; thus lavandin is more quickly affected by hydric stress than the fine lavender. Terpenes contents were slightly impacted by hydric stress and with different states amoung species and populations. Despite huge answer diversities amoung compounds, hydric stress intensity and plants; camphre pathway terpenes (borneol, comphene and camphre) are those which have the more important variations among stressed and controlled plants. Therefore study camphre biosynthesis pathway emerged. In this context, we have identified and characterized the bornyl diphosphate synthase able to produce the borneol from the bornyl diphosphate. These works allow a better understanding of relationships between volatils terpenes production and environment as well as give genetical tools to proceed to further investigations

Terpenes are secondary metabolites produced by plants and they have promising roles in plant defense and pharmaceuticals. They are synthesized by terpene synthases and these enzymes are part of a complex plant metabolic pathway. Diterpene biosynthesis requires co-expression of class II and class I diterpene synthases (diTPSs) to convert geranylgeranyl diphosphate (GGPP), the common precursor, into a C20 intermediate substrate. These substrates then use cytochrome p450s (CYPs) as their final steps to form diterpene scaffolds. CYPs are monooxygenases that change the redox status of their substrates into final diterpene products. Medicago truncatula was used as my model organism to investigate how legumes synthesize these secondary metabolites to contribute to crop defense improvement in the future. Seven diTPSs - MtTPS17, MtTPS18, MtTPS19, MtTPS37, MtTPS38, MtTPS39, and MtTPS40 - in M. truncatula have been identified. MtTPS38 was found to produce ent-CPP and MtTPS37 used ent-CPP to yield ent-kaurene. Combinatorial expression showed that MtTPS38 and MtTPS37 react together to produce ent-kaurene, a precursor for an important plant hormone gibberellin (GA). CYPs have also been discovered to be clustered around MtTPS19, suggesting the possibility of MtTPS19 utilizing these CYPs for downstream reactions.

Plants produce secondary (or specialized) metabolites to respond to a variety of environmental changes and threats. Especially, volatile compounds released by plants facilitate short and long distance interaction with both beneficial and harmful organisms. Comparatively little is known about the organization and role of specialized metabolism in root tissues. In this study, we have investigated the root-specific formation and function of volatile terpenes in the model plant Arabidopsis. As one objective, we have characterized the two root-specific terpene synthases, TPS22 and TPS25. Both enzymes catalyze the formation of several volatile sesquiterpenes with (E)-β-farnesene as the major product. TPS22 and TPS25 are expressed in the root in distinct different cell type-specific patterns and both genes are induced by jasmonic acid. Unexpectedly, both TPS proteins are localized to mitochondria, demonstrating a subcellular localization of terpene specialized metabolism in compartments other than the cytosol and plastids. (E)-β-Farnesene is produced at low concentrations suggesting posttranslational modifications of the TPS proteins and/or limited substrate availability in mitochondria. We hypothesize that the mitochondrial localization of TPS22 and TPS25 reflects evolutionary plasticity in subcellular compartmentation of TPS proteins with emerging or declining activity. Since (E)-β-farnesene inhibits Arabidopsis root growth in vitro, mitochondrial targeting of both proteins may fine tune (E)-β-farnesene concentrations to prevent possible autotoxic or inhibitory effects of this terpene in vivo. We further investigated the role of volatile terpenes in Arabidopsis roots in interaction with the soil-borne oomycete, Pythium irregulare. Infection of roots with P. irregulare causes emission of the C11-homoterpene (or better called C4-norterpene) 4,8-dimethylnona-1,3,7-triene (DMNT), which is a common volatile induced by biotic stress in aerial parts of plants but was not previously known to be produced in plant roots. We demonstrate that DMNT is synthesized by a novel, root-specific pathway via oxidative degradation of the C30-triterpene, arabidiol. DMNT exhibits inhibitory effects on P. irregulare mycelium growth and oospore germination in vitro. Moreover, arabidiol and DMNT biosynthetic mutants were found to be more susceptible to P. irregulare infection and showed higher rates of Pythium colonization in comparison to wild type plants. Together, our studies demonstrate differences and plasticity in the metabolic organization and function of terpenes in roots in comparison to aboveground plant tissues. / Ph. D.