Metabolome profiling of floral scent production in Petunia axillaris

Highlights

• Concentrations of scent compounds of Petunia flowers increased at night.

• Concentration changes of precursors after G6P were similar to those of scent compounds.

• Biosynthetic regulation after G6P results from, at least in part, substrate concentration.

• SAM showed contrary changes in concentrations to SAH, methionine and scent compounds.

• Metabolism of G6P to 6PG and metabolism of shikimic acid are reduced in the weakly scented line.

Abstract

Emission of floral scent benzenoid/phenylpropanoid compounds in Petunia axillaris increases significantly at night, a change that is primarily determined by the endogenous concentration of these compounds in the corolla. Among wild type P. axillaris plants, there are lines that emit different amounts of scent. To understand how the nocturnal rhythm of floral scent concentrations is controlled, the concentration profiles of metabolites in the scent biosynthetic pathway in two lines of P. axillaris, a strongly scented line and a weakly scented line, are reported. In the strongly scented line, the concentration of a series of compounds from glucose-6-phosphate (G6P) to the scent compounds changed synchronously. In the weakly scented lines, the concentrations of some metabolites including 6-phosphogluconate (6PG) and downstream metabolites of shikimic acid were remarkably lower, suggesting a reduction in metabolism of G6P to 6PG and the metabolism of shikimic acid in the weakly scented line. Nocturnal increases in the concentrations of sucrose, fructose, and glucose were not found in strongly scented line. Nocturnal increases in concentrations of S-adenosylhomocysteine (SAH) and methionine and reductions in the concentrations of S-adenosylmethionine (SAM), a methylation donor to benzenoid-skeletons, were observed only in strongly scented line. It is concluded that the biosynthetic regulation of each step from G6P to the volatile scent benzenoids is performed by, at least in part, concentrations of substrates, and the regulation also affects concentrations of SAM cycle compounds.

Introduction

Due to several excellent properties including ease of growth, short life cycle, relatively compact plant size, and variation in shape and colouration, Petunia plants are often used for physiological studies (Gerats and Vandenbussche, 2005). Strong floral scent is a prominent feature of some Petunia plants, and thus they have become model plants for scent studies (Schuurink et al., 2006). Petunia axillaris, which is widely distributed in temperate South America (Ando et al., 2001), emits scent compounds predominantly at night (Verdonk et al., 2003, Oyama-Okubo et al., 2005), and the mechanisms of regulating floral scent biosynthesis using this property of P. axillaris is a current emphasis. The floral scent compounds of P. axillaris are all benzenoids, and methyl benzoate and iso-eugenol are major components. The nocturnal emission of these floral scent compounds is synchronized with increases in their concentrations in the corolla (Oyama-Okubo et al., 2005). Variations in scent emission strength among the subspecies of P. axillaris are also correlated with corolla concentrations (Kondo et al., 2006). Therefore, based on these previous studies, it is concluded that under constant temperature conditions, the levels of floral scent emission are primarily determined by corolla benzenoid concentrations.

By in vivo stable isotope labeling and computer-assisted metabolic flux analysis, all aromatic compounds in Petunia floral scents were found to be biosynthesized from a common precursor, phenylalanine (Boatright et al., 2004), derived from shikimate and chorismate pathway. Fig. 1 depicts the overall pathway to phenylalanine, including the glucose metabolic pathway which branches after glucose-6-phosphate (G6P) into the glycolysis pathway and the pentose phosphate pathway; these later converge at 3-deoxy-arabinoheptulosonic acid-7-phosphate (DAHP) and lead to phenylalanine as indicated. mRNAs encoding DAHP synthase, 5-enolpyruvylshikimate-3-phosphate (EPSP) synthase, chorismate mutase, and phenylalanine ammonia-lyase show rhythmic expression similar to the emission of volatile benzenoids in another strongly scented Petunia, P. hybrida cv. Mitchell (Verdonk et al., 2005). Genes encoding enzymes that catalyse steps in the SAM cycle, including S-adenosyl-l-methionine and benzoic acid/salicylic acid carboxyl methyltransferases were also reported to show similar rhythmic expression (Negre et al., 2003). However, the expression pattern of ODORANT1, a novel R2R3-type MYB transcription factor that regulates expression of the EPSP synthase gene, did not correspond with the increase in volatile benzenoids (Verdonk et al., 2005). Moreover, the expression of many genes encoding methyltransferases that catalyse the synthesis of volatile benzenoids decreased during the night, a finding that is opposite to the emission pattern of volatile benzenoids (Colquhoun et al., 2010). To date, none of the enzymes involved in the biosynthesis of volatile benzenoids have shown significant day–night activity changes (Kolosova et al., 2001, Colquhoun et al., 2010). These findings indicate that major factors other than the expression of biosynthetic genes operate to regulate biosynthesis of volatile scent benzenoids. Here, the possibility was investigated that the activities of some parts of the biosynthetic pathways could be regulated by different substrate concentrations. Changes in the concentration of metabolites from sugars to volatile scent benzenoids were thus analyzed over time to find nocturnal increases in levels of compounds and to identify metabolic steps regulated by substrate concentration in the P. axillaris corolla.

Among wild type P. axillaris plants, there are lines that emit different amounts of scent (Kondo et al., 2006). Thus, to understand the regulation of volatile scent benzenoid biosynthesis in more detail, levels of metabolites were compared between a strongly scented line and a weakly scented line; it was hypothesized that a metabolic interaction between parts of the biosynthetic pathways could be revealed by this comparison. Here, two different P. axillaris lines were selected for study: a strongly scented line, U1AXI, and a weakly scented line, B1320AXI10. Endogenous levels of floral scent compounds in the strongly scented line have diurnal oscillations with maxima at 0000 (midnight) and minima at 1200 (noon) (Oyama-Okubo et al., 2005). In the weakly scented line, the concentration of scent compounds including dihydroconiferyl acetate, a biosynthetically related analog of a scent benzenoid iso-eugenol, was ca. 2% of the strongly scented line (Kondo et al., 2007). Therefore, it was considered that the inhibited step in the weakly scented line occurred earlier than in the pathway for the biosynthesis of scent benzenoids, and that these two lines were suitable materials for probing the regulatory mechanism.

Recently, metabolite profiling (metabolomics) of extracts from numerous species of organisms have been performed by using one or more analytical instruments, such as gas chromatography–mass spectrometry (GC–MS) (Fiehn et al., 2000), capillary electrophoresis–mass spectrometry (CE–MS) (Soga et al., 2003, Soga et al., 2007), liquid chromatography–mass spectrometry (LC–MS) (Yoshida et al., 2007), Fourier transform ion cyclotron resonance mass spectrometry (Aharoni et al., 2002) and nuclear magnetic resonance spectrometry (Reo, 2002). In this regards, the precursors of scent compounds include neutral carbohydrates, phosphorylated saccharides, organic acids, and amino acids. Among these various types of compounds, phosphorylated saccharides are the most difficult to analyse due to their high polarity and negative charges. This problem was resolved by developing a CE–MS system, and for this study, organic acids and amino acids were analyzed by this method. Neutral carbohydrates and scent compounds were also studied by high-performance liquid chromatography (HPLC) and GC–MS, respectively, depending on their chemical properties.