Analysis of the interface between primary and secondary metabolism in catharanthus roseus cell cultures using (13)C-stable isotope feeding and coupled mass spectrometry.

Dear Editor, The tropical plant Madagascar periwinkle Catharanthus roseus (L.) G.Don is a rich source of plant-derived medicinal terpenoid indole alkaloids (TIAs), including the anti-hypertensive ajmalicine, the sedative compound serpentine, and the anticancer drugs vinblastine and vincristine. However, the latter two compounds are produced in C. roseus plants only in very low amounts. Elicitors such as hormones (e.g. jasmonates or salicylic acid) activate plant natural defense responses, including increased secondary metabolite production (El-Sayed and Verpoorte, 2007; Lackman et al., 2011). There are two different metabolic pathways for the biosynthesis of terpenoids leading to the formation of the C5 central precursor isopentenyl diphosphate (IPP), namely the classical cytosolic mevalonate pathway and the mevalonate-independent plastidic 2-C-methyl-D-erythritol 4-phosphate (MEP) pathway. The biosynthesis of C. roseus TIAs starts with the condensation of tryptamine (shikimate pathway) and secologanin (MEP pathway) to form the key intermediate strictosidine, the common backbone structure of all TIAs. However, to our knowledge, the late steps of the alkaloid pathway in C. roseus and the genes involved remain partly unknown (Rischer et al., 2006). When there is a high demand for the product and only low levels accumulate in plants, plant cell cultures have shown to be an alternative production system of valuable secondary metabolites offering controlled conditions on a large scale for the rapid reproduction of plant cells without the need for depleting natural sources (Oksman-Caldentey and Inzé, 2004). Metabolic flux analysis using 13C-labeled substrates is essential for understanding the control and regulation of metabolic networks, and it has been often observed that significant changes in C-flow are sometimes associated with only small adjustments in metabolite abundance (Ratcliffe and Shachar-Hill, 2006; Baxter et al., 2007). Whilst transgenic lines, which are one prerequisite for this approach, are available for a multitude pathways of secondary metabolism, methods for assessing metabolic fluxes, which are another prerequisite, are largely not. In this study, we performed a 13C-stable isotope feeding experiment using [C3]-pyruvate as substrate. Given that mass spectrometry does not distinguish atomic level resolution, [C3]-pyruvate was introduced into the network for enhanced sensitivity, and the labeling pattern of metabolites was subsequently measured using two complementary techniques: gas chromatography coupled to time-of-flight–mass spectrometry (GC–TOF–MS) (Lisec et al., 2006; Huege et al., 2011; Kopka et al., 2005; Luedemann et al., 2008) and ultra-performance liquid chromatography coupled to high-resolution Fourier Transform mass spectrometry (UPLC–FT–MS) (Giavalisco et al., 2009; Tohge and Fernie, 2010). High-resolution UPLC–FT–MS analysis was used for the target analysis of the secondary metabolites and TIA precursors loganic acid, loganin, and secologanin, and also the TIA ajmalicine, not detectable by GC–TOF–MS, which were here rather used to assess the importance of primary metabolite precursor supply to these pathways. Catharanthus roseus cell extracts treated for 24 h with methyl jasmonate (MeJA) or dimethyl sulfoxide (DMSO/ control) and harvested at different time points (0, 1, 4, 8, and 24 h) were first analyzed for their metabolite composition using GC–TOF–MS (Supplemental Table 1, no addition of 13C-labeled substrate). Supplemental Table 1 shows the steady-state relative levels of all measurable metabolites (53 metabolites) in C. roseus cell extracts. Significant accumulation (t-test p < 0.05) of maltose, trehalose, galactinol, glucose-6phosphate, alanine, glutamate, tyrosine, histidine, γ-amino butyric acid (GABA), urea, succinate, ascorbate, and threonate was observed in C. roseus cells after 4 h of MeJA elicitation. Only phenylalanine was shown to significantly decrease after 8 h of MeJA elicitation (t-test p < 0.05). From the total primary metabolites detected, only 15% were shown to increase after 4 h of MeJA treatment, 2% decreased, whereas the majority of metabolites (83%) remained unchanged, suggesting that the levels of primary metabolites in C. roseus cells were largely not affected by MeJA/DMSO treatment. We next determined the fate of 13C-label accumulated in primary metabolites using GC–TOF–MS and the fate of 13C-label