Biosynthetic Origins of the Isoprene Units of 4-Nerolidylcatechol in Potomorphe umbellata

A origem biossintética das unidades isoprênicas do 4-nerolidilcatecol (1), principal metabólito secundário de Potomorphe umbellata, foi investigada através da incorporação de glicose marcada com [C] e [C], e com precursores das vias mevalonoídica e triose/piruvato, incluindo mevalonolactona-[2-C] e 3-fosfato de gliceraldeído-[U-C], respectivamente. O padrão de incorporação de glicose-[1-C] em 1 foi determinado por experimentos de RMN C quantitativos. O padrão de marcação revelou que o precursor foi especificamente incorporado, e que as unidades isoprênicas presentes no 4-nerolidilcatecol são derivadas tanto do ácido mevalônico, quanto da via triose/piruvato. Os resultados também indicaram que tanto a via plastídica quanto citoplasmática são capazes de gerar unidades de difosfato de isopentenila para a biossíntese de 1.


Introduction
The terpenoids represent one of the largest classes of secondary metabolites comprising, together with the steroids, more than 30,000 compounds to date. 1 Whilst the terpenoids are known to play important roles in the development and survival of every living organism, 2 many compounds of the class have been found to accumulate at high levels in a number of plant species.They possess, as a common building block, the isoprene unit that is formally derived from isopentenyl diphosphate (IPP).Originally, the well-studied mevalonic acid (MVA) pathway was considered to be the unique biosynthetic route to IPP and was believed to operate in all organisms.5][6][7][8] The reaction sequence of the alternative pathway involves condensation of pyruvate and glyceraldehyde-3-phosphate leading to the formation of 1-desoxyxylulose-5-phosphate, the first C 5 intermediate.Several subsequent steps, including an intramolecular rearrangement, lead to IPP. 4 There is some reason to believe that the triose/pyruvate pathway is compartmentalised in the plastids of plant cells and is involved in the formation of mono-, di-, and tetraterpenoids, whilst the MVA pathway is located in the cytoplasm and is responsible for the synthesis of sesquiterpenes and sterols. 9However, substantial evidence indicates that this hypothesis of subcellular compartmentation of plant terpenoid biosynthesis may be simplistic as it does not take into account the exchange of plastidic and cytoplasmic metabolites.For example, early studies involving chamomile flowers (Chamomilla recutita) demonstrated an efficient incorporation of label from [1-13 C]-glucose into sesquiterpenes, but revealed an heterogeneous labelling pattern in the accumulated farnesyl pyrophosphate.Thus, whilst the two initial isoprene units clearly originated from the triose/pyruvate route, the third unit was derived from both the MVA and the triose/pyruvate pathways. 10he major constituent of leaves and roots of Potomorphe umbellata is the powerful antioxidant, 4-nerolidylcatechol (1). 11,12Currently, no information is available concerning the biosynthesis of either the sesquiterpene or the catechol portion of this type of compound.The aim of the present study was, therefore, to determine the route of formation of the C 15 moiety of 1 in leaves of P. umbellate through feeding experiments involving appropriately labelled precursors.Determination of the 13 C-enrichment pattern in 1 following the incorporation of label from [1-13 C]-glucose revealed that the sesquiterpene portion was derived from both the MVA and the triose/pyruvate pathways.

Results and Discussion
Labelled, putative precursors of the sesquiterpene portion of 4-nerolidylcatechol (1), namely, [3-14 C]-R,Smevalonolactone and [U- 14 C]-glyceraldehyde-3-phosphate, were administrated to the leaves of Potomorphe umbellata, together with [U-14 C]-glucose, which is a precursor of both mevalonolactone and glyceraldehyde-3-phosphate.Freshly harvested leaves were incubated, separately, with 2 μCi of each of these precursors for 6, 24, 48 and 72 h, after which time the leaf material was extracted with ethanol and analysed by HPLC.Radiolabel from all precursors was incorporated into 1 (Table 1).Specific and/or non-specific incorporations of ≥ 1% were obtained within 72 h of feeding with [3-14 C]-R,S-mevalonolactone and [U- 14 C]glyceraldehyde-3-phosphate but, as would be expected, a considerably lower percentage incorporation (0.1 -0.3%) was attained with [U-14 C]-glucose.
In order to determine the specificity of incorporation, and hence the possible biosynthetic origin of the isoprenoid moieties, experiments involving the incorporation of [1-13 C]-glucose into 4-nerolidylcatechol (1) were carried out.Different patterns of 13 C-enrichment in 1 would be expected depending on the pathway followed. 10,13,14Thus, the conversion of [1-13 C]-glucose into IPP via the MVA pathway would lead to the incorporation of label at positions 2, 4 and 5 of each isoprene unit.On the other hand, if [1-13 C]-glucose is converted into pyruvate, the resulting C 5 units derived from the triose/pyruvate pathway should be labelled at positions 1 and 5 (Figure 1).Whichever of the two pathways is extant, no label should appear at position 3 of any of the isoprene units if [1- 13 C]-glucose is specifically incorporated into 1.
In feeding experiments involving [1-13 C]-glucose as a putative precursor, eight young leaves of P. umbellata were fed for 72 h with 100 μL of a 10% solution of [1-13 C]glucose, and then extracted with ethanol.The ethyl acetate soluble portion of the ethanol extract was fractioned by column chromatography over silica gel followed by TLC to afford 4-nerolidylcatechol (1).The 13 C NMR spectra of this sample of 1 and of 4-nerolidylcatechol at natural abundance sample were measured in CDCl 3 under identical conditions.The relative abundances of 13 C at individual carbon atoms were calculated by comparison of the integration values for the peaks in the labelled sample with those in the natural abundance sample.The values were normalised to a reference value of 1.1% for 13 C abundance at the carbon with the lowest 13 Cenrichment (Table 2).Enhancement of the signals was significant and varied in the range 2.9 -6.9%.Most importantly, no enrichment of the signals at C-3', C-7' or C-11' of 1 was observed indicating that label was not incorporated into position 3 of any of the isoprene units (Figure 1).This demonstrates that the label from [1-13 C]glucose was incorporated specifically into 1, and not via catabolic processes followed by re-incorporation, within the time period of incubation.Furthermore, a significant enrichment of the remaining carbons (C-2, 4-6) was found (relative 13 C abundance of 1.5%).This general labelling might be due to complex metabolic turnover of [1-13 C]glucose with following statistical distribution of the carbon atoms in various positions of intermediates of carbohydrate metabolism.
As shown in Table 2, feeding experiment with [1-13 C]glucose resulted in clear enhancements of the signals at positions 1, 2, 4 and 5 of all of the isoprene units of the nerolidol moiety.Such a pattern of specific incorporation is consistent with the simultaneous operation of the MVA and triose/pyruvate pathways in the biosynthesis of the sesquiterpenoid portion of 4-nerolidylcatechol (1) in P. umbellata, and suggests an exchange of plastidic and cytoplasmic IPP units.These results corroborate previous studies 9 that provided evidence of significant differences in the degree of IPP exchange depending on the type and developmental stage of the tissue, thus giving rise to flexible compartmentalisation of terpenoid biosynthesis in plants.

Plant material
Leaves of Potomorphe umbellata employed in the feeding experiments with 14 C-labelled precursors were collected from plants grown under a 16 h day regime in greenhouses at the Institute of Biological Chemistry (WSU) in Pullman, USA.Plants used in experiments involving [1-13 C]-glucose were grown under a natural daynight regime in greenhouse at the Institute of Chemistry (UNESP) in Araraquara, Brazil.

Administration of [ 14 C]-labelled compounds to leaves of P. umbellata
Young leaves of P. umbellata were excised and fed for 6, 24, 48 and 72 h with 100 μL of a solution containing 2 μCi of either [3-14 C]-R,S-mevalonolactone, [U-14 C]-glyceraldehyde-3-phosphate or [U-14 C]-Dglucose using one leaf for each feeding experiment.The leaf was frozen in liquid nitrogen, dried overnight under vacuum and extracted, with sonication, for 15 min with ethanol (2 x 10 mL).The solvent was evaporated, the residue was resuspended in methanol (1 mg mL -1 ) and a 10 μL aliquot analysed by HPLC (elution programmelinear gradient of methanol : water from 65:35 to 85:15 in 10 min, followed by isocratic elution with 85:15 for a further 10 min; flow rate -1 mL min -1 ; detection -280 nm).Fractions (1 mL) of the eluate were collected over a 44 min period and submitted to liquid scintillation counting.

Table 2 .
Percentage 13 C abundance in 4-nerolidylcatechol (1) isolated after incorporation of [1-13 C]-D-glucose in leaves of Potomorphe umbellata Referenced to CDCl 3 ; b Calculated from comparison 3 of relative intensities of signals in 13 C NMR spectra of labelled and unlabelled 4-nerolidylcatechol. a