Glanduläre Trichome und Isoprenoidbiosynthese

Leaves of tobacco (Nicotiana tabacum) are covered with glandular trichomes that produce sucrose esters and diterpenoids in varying quantities, depending on cultivar type. The bicyclic diterpene Z‐abienol is the major labdanoid present in some oriental tobacco cultivars, where it constitutes a precursor of important flavours and aromas. We describe here the identification and characterization of two genes governing the biosynthesis of Z‐abienol in N. tabacum. As for other angiosperm labdanoid diterpenes, the biosynthesis of Z‐abienol proceeds in two steps. NtCPS2 encodes a class‐II terpene synthase that synthesizes 8‐hydroxy‐copalyl diphosphate, and NtABS encodes a kaurene synthase‐like (KSL) protein that uses 8‐hydroxy‐copalyl diphosphate to produce Z‐abienol. Phylogenetic analysis indicates that NtABS belongs to a distinct clade of KSL proteins that comprises the recently identified tomato (Solanum habrochaites) santalene and bergamotene synthase. RT‐PCR results show that both genes are preferentially expressed in trichomes. Moreover, microscopy of NtCPS2 promoter‐GUS fusion transgenics demonstrated a high specificity of expression to trichome glandular cells. Ectopic expression of both genes, but not of either one alone, driven by a trichome‐specific promoter in transgenic Nicotiana sylvestris conferred Z‐abienol formation to this species, which does not normally produce it. Furthermore, sequence analysis of over 100 tobacco cultivars revealed polymorphisms in NtCPS2 that lead to a prematurely truncated protein in cultivars lacking Z‐abienol, thus establishing NtCPS2 as a major gene controlling Z‐abienol biosynthesis in tobacco. These results offer new perspectives for tobacco breeding and the metabolic engineering of labdanoid diterpenes, as well as for structure–function relationship studies of terpene synthases.

Many plant species have evolved specialized organs dedicated to the production of a restricted number of secondary metabolites. These organs have secretory tissues which can lead to very significant accumulations of products, in the range of mg per g of fresh weight. These natural cell factories are therefore interesting targets for metabolic engineering. Plant glandular trichomes in particular have attracted interest because of the relative ease to isolate them and to analyse the compounds they produce because they are secreted onto the leaf surface. Depending on the species, trichomes can produce a variety of metabolites. Terpenoids however are particularly well represented and have been used by humans in a variety of industries, including as aroma, fragrance and pharmaceutical ingredients. Tobacco trichomes produce diterpenoids in large amounts and were therefore chosen as a model system for engineering the biosynthesis of this important class of compounds. We present here our strategy and first results, which bode well for the future of glandular trichomes as engineered natural cellular factories.