Phosphatidylinositol 4-kinase activation is an early response to salicylic acid in Arabidopsis suspension cells

Abstract

Salicylic acid (SA) has a central role in defense against pathogen attack. In addition, its role in such diverse processes as germination, flowering, senescence, and thermotolerance acquisition has been documented. However, little is known about the early signaling events triggered by SA. Using Arabidopsis (Arabidopsis thaliana) suspension cells as a model, it was possible to show by in vivo metabolic phospholipid labeling with (33)P(i) that SA addition induced a rapid and early (in few minutes) decrease in a pool of phosphatidylinositol (PI). This decrease paralleled an increase in PI 4-phosphate and PI 4,5-bisphosphate. These changes could be inhibited by two different inhibitors of type III PI 4-kinases, phenylarsine oxide and 30 microm wortmannin; no inhibitory effect was seen with 1 microm wortmannin, a concentration inhibiting PI 3-kinases but not PI 4-kinases. We therefore undertook a study of the effects of wortmannin on SA-responsive transcriptomes. Using the Complete Arabidopsis Transcriptome MicroArray chip, we could identify 774 genes differentially expressed upon SA treatment. Strikingly, among these genes, the response to SA of 112 of them was inhibited by 30 microm wortmannin, but not by 1 microm wortmannin.

Figure 1.

Expression of marker genes of the SA-responsive pathway in Arabidopsis cell suspensions. Time course of early marker gene GST6 and late marker gene PR1 expression was detected by RT-PCR after SA treatment of cells. The number of PCR cycles was optimized for each primer pair. 40S ribosomal protein S24 (RPS24A) was used as a housekeeping gene. GST6, Glutathione S-transferase6.

Figure 2.

Time course of metabolic phospholipid labeling. ³³P_i was added to the cell suspension and lipids were extracted at the times indicated. Phosphomonoester (A) and phosphodiester (B) phospholipids show different labeling patterns. The experiment is representative of three independent repetitions. Results are expressed as counts incorporated into each lipid class. r.a., Right axis scale. PIP and PI(4,5)-P₂ were separated in the alkaline solvent system; other lipids were separated in the acid solvent system.

Figure 3.

Labeled phosphoinositides in SA-treated Arabidopsis cells. A, Medium time course of phosphoinositide-labeling rate upon SA treatment. B, Changes in labeled phosphoinositide levels appear during the first minutes of SA treatment. C, Dose dependence of the observed changes. Cells were treated with the corresponding concentration of SA for 8 min before lipid extraction. r.a., Right axis scale. A to C, Cells were labeled with ³³P_i 5 min before lipid extraction. PIP and PI(4,5)-P₂ were separated in the alkaline solvent system; PI was separated in the acid solvent system. Values for water-treated cells were taken as controls (100%).

Figure 4.

SA-induced PIP is PI(4)-P. Cells were treated with SA or with water 8 min before lipid extraction and labeled by ³³P_i 5 min before lipid extraction. Lipids were extracted and separated in parallel using an alkaline solvent system (A) and a borate solvent system (B). The expected position of PI(3)-P is also indicated (R_f = 0.45).

Figure 5.

Inhibitory effect of wortmannin and PAO on the decrease in labeled PI caused by SA. Cells were incubated 45 min with wortmannin (A) or 15 min with PAO (B) prior to SA application and then treated with SA for 8 min before lipid extraction. Cells were labeled by ³³P_i 5 min before lipid extraction. Inhibitors were given in dimethylsulfoxide (final concentration 0.5% [v/v]). Controls (100%) represent water-treated cells. Results are averages of three independent repetitions.

Figure 6.

Functional categories of SA-induced and SA-repressed genes after 4 h of SA treatment in Arabidopsis suspension cells. Genes with unambiguous probe-to-gene assignment that were either up-regulated or down-regulated according to the microarray analysis were classified into functional categories using the MIPS functional catalog interface (Ruepp et al., 2004).

Figure 7.

Determination of SA-regulated genes downstream of the PI 4-kinase pathway. Among the 3,270 genes that showed differential gene expression in one of the five tested comparisons, 129 showed transcript level changes both in SA versus water and in SAW30 versus SAW1 conditions. Additionally, for these genes, the transcript levels did not change in the same direction in W30 versus W1 as in SAW30 versus W1. Hence, these genes are potentially SA regulated via the PI 4-kinase pathway. Of these genes, 108 were up-regulated and 21 were down-regulated.