Feedback inhibition of starch degradation in Arabidopsis leaves mediated by trehalose 6-phosphate

Abstract

Many plants accumulate substantial starch reserves in their leaves during the day and remobilize them at night to provide carbon and energy for maintenance and growth. In this paper, we explore the role of a sugar-signaling metabolite, trehalose-6-phosphate (Tre6P), in regulating the accumulation and turnover of transitory starch in Arabidopsis (Arabidopsis thaliana) leaves. Ethanol-induced overexpression of trehalose-phosphate synthase during the day increased Tre6P levels up to 11-fold. There was a transient increase in the rate of starch accumulation in the middle of the day, but this was not linked to reductive activation of ADP-glucose pyrophosphorylase. A 2- to 3-fold increase in Tre6P during the night led to significant inhibition of starch degradation. Maltose and maltotriose did not accumulate, suggesting that Tre6P affects an early step in the pathway of starch degradation in the chloroplasts. Starch granules isolated from induced plants had a higher orthophosphate content than granules from noninduced control plants, consistent either with disruption of the phosphorylation-dephosphorylation cycle that is essential for efficient starch breakdown or with inhibition of starch hydrolysis by β-amylase. Nonaqueous fractionation of leaves showed that Tre6P is predominantly located in the cytosol, with estimated in vivo Tre6P concentrations of 4 to 7 µm in the cytosol, 0.2 to 0.5 µm in the chloroplasts, and 0.05 µm in the vacuole. It is proposed that Tre6P is a component in a signaling pathway that mediates the feedback regulation of starch breakdown by sucrose, potentially linking starch turnover to demand for sucrose by growing sink organs at night.

Figure 1.

Effect of TPS overexpression on metabolites and the redox status of AGPase in Arabidopsis rosettes during the day. Ethanol-inducible TPS line 29.2 (TPS) and plants expressing the ethanol-binding transcription factor (AlcR) were grown in soil with a 12-h photoperiod. Four-week-old plants were sprayed with water or 2% (v/v) ethanol at the beginning of the day and harvested at 1- or 2-h intervals after spraying for the determination of Tre6P (A), starch (B), Suc (C), the starch-Suc ratio (D), ADPG (E), and the redox status of AGPase (F). Gray symbols at 0 h represent unsprayed AlcR (gray triangles) or TPS (gray squares) plants. Data are means ± sd (n = 3). Significant differences (one-way ANOVA, Holm-Sidak test) between the ethanol-sprayed TPS line and the three controls, AlcR (sprayed with water or ethanol) and TPS (water), are indicated by asterisks: *P < 0.05 and ***P < 0.001. FW, Fresh weight; ZT, Zeitgeber time.

Figure 2.

Induced changes in the Tre6P content of Arabidopsis rosettes at night. Ethanol-inducible TPS plants (TPS29.2 and TPS31.3) were grown in soil with a 12-h photoperiod. Wild-type plants (Col-0) and plants expressing the AlcR ethanol-binding transcription factor (AlcR) were grown as controls. Four-week-old plants were sprayed with water (white bars) or 2% (v/v) ethanol (black bars) at the beginning of the night. Pools of 10 rosettes were harvested 12 h later at the EN for the determination of Tre6P (A), trehalose (B), Suc (C), Glc (D), Fru (E), starch (F), maltose (G), and maltotriose (H). Data are means ± sd (n = 3). Significant difference between ethanol- and water-treated plants from the same genotype are indicated by asterisks (Student’s t test): *P < 0.05, **P < 0.01, and ***P < 0.001. FW, Fresh weight.

Figure 3.

Inhibition of starch degradation at night by induced high levels of Tre6P. Ethanol-inducible TPS29.2 plants were grown in soil with a 12-h photoperiod. Four-week-old plants were sprayed with water (white circles) or 2% (v/v) ethanol (black circles) at the ED (A–D) or in the middle of the day (E–H). Pools of five rosettes were harvested at the ED and at 2- or 4-h intervals through the night for the determination of Tre6P (A and E), starch (B and F), Suc (C and G), and maltose (D and H). Data are means ± sd (n = 3). Significant differences between the water- and ethanol-treated plants at the same time point are indicated by asterisks (Student’s t test): *P < 0.05, **P < 0.01, and ***P < 0.001. FW, Fresh weight.

Figure 4.

Phosphate content and phosphorylation of starch granules isolated from induced and noninduced TPS29.2 plants at night. Inducible TPS29.2 plants were sprayed with water (white circles) or 2% (v/v) ethanol (black circles) 6 h before the ED. A and B, Rosettes were harvested at the ED and at 2 to 12 h into the night for the determination of Tre6P and the isolation of starch granules. C to F, The phosphate content of the starch granules was determined after enzymatic and acid hydrolysis. C and D, Pi content (C₆ only). E and F, Total phosphate content (C₆ + C₃). G and H, In vitro phosphorylation of the granules by recombinant potato GWD and Arabidopsis PWD was determined by the incorporation of ³³P from [β-³³P]ATP. Tre6P data are means ± sd (n = 3 or 4). Significant differences between ethanol- and water-treated plants are indicated by asterisks (Student’s t test): *P < 0.05, **P < 0.01, and ***P < 0.001. Starch granule data are single measurements from two independent experiments: experiment 1 (A, C, E, and G) and experiment 2 (B, D, F, and H). FW, Fresh weight.

Figure 5.

Effect of an early dusk on metabolite levels in Arabidopsis leaves. Wild-type Arabidopsis Col-0 plants were grown in soil with a 12-h photoperiod for 3 weeks. Pools of five rosettes were harvested at various intervals through two sequential diurnal cycles, 12 h of light/12 h of dark (control; 1) and 8 h of light/16 h of dark (early dusk treatment; 2), for the measurement of starch (A), maltose (B), Suc (C), and Tre6P (D). Data are means ± sd (n = 4). FW, Fresh weight; ZT, Zeitgeber time.

Figure 6.

Integrated model of the control of starch breakdown by Tre6P and the circadian clock. The maximum permissible rate of starch degradation is set by the circadian clock to ensure that starch reserves are not exhausted before the expected dawn. If Suc export is restricted by low demand from sink organs, Suc accumulates in the leaves and Tre6P rises, leading to the inhibition of an early step in the pathway of starch degradation upstream of maltose production. As the core components of the clock operate in the nucleus, the pathway whereby it regulates starch breakdown presumably involves the transport of an unidentified signal into the chloroplast. Tre6P probably also inhibits starch breakdown via an intermediary that is formed in the cytosol and transmitted to the chloroplast.