A reactive oxygen species Ca2+ signalling pathway identified from a chemical screen for modifiers of sugar-activated circadian gene expression

Abstract

Sugars are essential metabolites for energy and anabolism that can also act as signals to regulate plant physiology and development. Experimental tools to disrupt major sugar signalling pathways are limited. We performed a chemical screen for modifiers of activation of circadian gene expression by sugars to discover pharmacological tools to investigate and manipulate plant sugar signalling. Using a library of commercially available bioactive compounds, we identified 75 confident hits that modified the response of a circadian luciferase reporter to sucrose in dark-adapted Arabidopsis thaliana seedlings. We validated the transcriptional effect on a subset of the hits and measured their effects on a range of sugar-dependent phenotypes for 13 of these chemicals. Chemicals were identified that appear to influence known and unknown sugar signalling pathways. Pentamidine isethionate was identified as a modifier of a sugar-activated Ca²⁺ signal that acts as a calmodulin inhibitor downstream of superoxide in a metabolic signalling pathway affecting circadian rhythms, primary metabolism and plant growth. Our data provide a resource of new experimental tools to manipulate plant sugar signalling and identify novel components of these pathways.

Keywords: Arabidopsis thaliana; LOPAC (Library of Pharmacologically Active Compounds); calcium; calmodulin; circadian clock; reactive oxygen species (ROS); sugar signalling; superoxide.

Fig. 1

A sugar response assay in Arabidopsis seedlings. (a) Luciferase luminescence in dark‐adapted CCR2p:LUC seedlings treated with the indicated concentration of sucrose or mannitol (means ± SEM, n = 6). (b) Sugar content in dark‐adapted seedlings treated with 30 mM sucrose or transferred to the light with or without 3‐(3,4‐dichlorophenyl)‐1,1‐dimethylurea (DCMU) (means ± SD, n = 4; *, P < 0.05 compared to Light, Bonferroni‐corrected t‐test). (c) Luciferase luminescence in dark‐adapted CCR2p:LUC seedlings treated with 30 mM sugars (means ± SEM, n = 8). (d, e) Fold change of peak luciferase reporter luminescence in dark‐adapted wild‐type (Col‐0) or mutant seedlings treated with mannitol (blue) or sucrose (red) compared to luminescence before treatment (means ± SD, n = 4). (f) CCR2 transcript level, normalized to UBQ10, in dark‐adapted wild‐type (Col‐0) or agg1 agg2 agg3 mutant seedlings 12 h after treatment with mannitol (blue) or sucrose (red) (means ± SD, n = 3). (d–f) No significant differences were detected between genotypes (P > 0.05, Bonferroni‐corrected t‐tests).

Fig. 2

Dose–response of the transcriptional response to sucrose for 15 chemicals. Inhibition of peak luciferase luminescence in dark‐adapted CCR2p:LUC Arabidopsis seedlings after treatment with 30 mM sucrose in the presence of the indicated concentration of a chemical compared to DMSO‐treated controls (means ± SD, n = 6–12).

Fig. 3

Validation of transcriptional effect of LOPAC chemicals. (a) Luciferase luminescence in 35Sp:LUC Arabidopsis seedlings, 16 h after transfer to media containing the minimum effective concentration of each chemical (means ± SD, n = 8; *, P > 0.05, Bonferroni‐corrected t‐test). (b) CCR2 transcript level, relative to UBQ10, in dark‐adapted Arabidopsis Col‐0 seedlings 12 h after treatment with 30 mM mannitol (blue), 30 mM sucrose (red) or 30 mM sucrose in the presence of the minimum effective concentration of the chemical (white) (means ± SD, n = 4; *, P > 0.05, Bonferroni‐corrected t‐test).

Fig. 4

Summary of effects of LOPAC chemicals on sugar‐related growth phenotypes. (a) Normalized effects of the minimum effective concentration of 13 LOPAC chemicals in Arabidopsis compared to DMSO‐treated controls on germination of dormant seeds on 30 mM mannitol (1) or sucrose (6), biomass of 7‐d‐old seedlings on ½MS (2), hypocotyl and root length of 7‐d‐old dark‐grown seedlings on 30 mM mannitol (3, 4) or sucrose (7, 8) and anthocyanin content in 9‐d‐old seedlings grown for 2 d on 90 mM mannitol (5) or sucrose (9). Charts are arranged in rank order of the sum of normalized effects. Complete data are shown in Supporting Information Figs S4 and S5. (b) k‐means clustering of growth phenotypes represented in (a). Colours in (a) match clusters in (b).

Fig. 5

Primary metabolite levels in chemical‐treated Arabidopsis seedlings. (a–d) Heatmaps of relative metabolite levels in seedlings treated with DMSO or chemicals at ZT23 and sampled at ZT0, ZT1.5, ZT4, ZT8, ZT10.5, ZT12, ZT22.5 and ZT24 (n = 5; *, P < 0.05 compared to DMSO, t‐test). (e) Relative amino acid levels in DMSO‐ or chemical‐treated seedlings at ZT10.5 (mean ± SD, n = 5; *, P < 0.01 compared to DMSO, t‐test).

Fig. 6

Effect of pentamidine isethionate (PI) and diphenyleneiodinium (DPI) on the circadian period. (a) Normalized luciferase luminescence in CCA1p:LUC and PRR7p:LUC Arabidopsis seedlings in continuous light after transfer to media containing DMSO, 25 μM PI, 5 μM DPI or 25 μM DCMU (means ± SD, n = 4). (b) Circadian period of luminescence rhythms in CCA1p:LUC and PRR7p:LUC Arabidopsis seedlings in continuous light (means ± SD, n = 8; *, P < 0.05 compared to DMSO, Bonferroni‐corrected t‐test).

Fig. 7

Pentamidine isethionate (PI) acts like calmodulin inhibitors to elevate cytosolic Ca²⁺. (a) Cytosolic Ca²⁺ concentration in 35Sp:AEQ Arabidopsis seedlings treated (arrow) with DMSO (control), 10 μM diphenyleneiodinium (DPI) or 25 μM PI (means ± SD, n = 6). (b) Inhibition of peak luciferase luminescence in dark‐adapted CCR2p:LUC Arabidopsis seedlings after treatment with 30 mM sucrose in the presence of the indicated concentration of chemical compared to DMSO‐treated controls (means ± SEM, n = 8). (c) Luciferase luminescence in dark‐adapted CCR2p:LUC Arabidopsis seedlings treated with 30 mM mannitol or 30 mM sucrose in the presence of DMSO (control), 25 μM PI, 150 μM W‐7 or 150 μM trifluoperazine (TFP) (means ± SEM, n = 8). (d) Seedling biomass of 7‐d‐old Arabidopsis Col‐0 seedlings sown on combinations of DMSO, 5 μM PI, 25 μM W‐7 and 25 μM TFP (means ± SD, n = 4 of 12 seedlings; different letters indicate P < 0.01, one‐way ANOVA with Tukey's HSD). (e) Circadian period estimates of luciferase luminescence in PRR7p:LUC, CCR2p:LUC and TOC1p:LUC Arabidopsis seedlings in continuous light treated with DMSO (control), 25 μM PI or 150 μM W‐7 (means ± SD, n = 12; *, P < 0.05 compared to DMSO, Bonferroni‐corrected t‐test). (f) Cytosolic Ca²⁺ concentration in 35Sp:AEQ Arabidopsis seedlings treated with DMSO (control), 100 μM PI or 150 μM W‐7 in the presence of DMSO (control), 1 mM GaCl₃ or 1 mM DNQX (means ± SD, n = 4).

Fig. 8

Pentamidine isethionate (PI) acts downstream of diphenyleneiodinium (DPI). (a) Aequorin luminescence in dark‐adapted 35Sp:AEQ Arabidopsis seedlings treated with 30 mM sucrose or mannitol in the presence of DMSO, 5 μM DPI or 25 μM PI (mean ± SD, n = 6). (b) L‐012 luminescence in dark‐adapted Arabidopsis Col‐0 seedlings treated with 30 mM mannitol or sucrose in the presence of DMSO, 10 μM DPI or 25 μM PI (means ± SEM, n = 6). (c) Images and quantification of nitroblue tetrazolium (NBT) stain for superoxide in dark‐adapted Arabidopsis Col‐0 seedlings 4 h after treatment with 30 mM mannitol or sucrose in the presence of DMSO, 5 μM DPI or 25 μM PI (Tukey boxplots, n = 8; *, P < 0.05 compared to mannitol, Bonferroni‐corrected t‐test).