Abscisic acid inhibits type 2C protein phosphatases via the PYR/PYL family of START proteins

Abstract

Type 2C protein phosphatases (PP2Cs) are vitally involved in abscisic acid (ABA) signaling. Here, we show that a synthetic growth inhibitor called pyrabactin functions as a selective ABA agonist. Pyrabactin acts through PYRABACTIN RESISTANCE 1 (PYR1), the founding member of a family of START proteins called PYR/PYLs, which are necessary for both pyrabactin and ABA signaling in vivo. We show that ABA binds to PYR1, which in turn binds to and inhibits PP2Cs. We conclude that PYR/PYLs are ABA receptors functioning at the apex of a negative regulatory pathway that controls ABA signaling by inhibiting PP2Cs. Our results illustrate the power of the chemical genetic approach for sidestepping genetic redundancy.

Figure 1. Pyrabactin Is A Seed Selective ABA Agonist

(A) Structures Of Molecules Described In This Study. (B) ATH1 Microarray Comparison of Pyrabactin and ABA Effects on Seeds and Seedlings. The axes plot log₂-transformed values for probe responses to pyrabactin (Y-axis) or ABA (X-axis), relative to control samples. Inset in each scatter plot is the Pearson correlation coefficient for each comparison. Probes selected for analyses were those significantly responsive to either ABA or pyrabactin. Germination responsive transcripts were removed for seed analyses. Detailed methods are provided in SOM.

Figure 2. Pyr/Pyls Are Necessary For ABA Signaling

(A) Pyr1 And Pyl1 - Pyl4 Expression Levels. Plots were made using the EFP browser (23); these heatmaps show normalized ATH1 microarray expression values (multiplied by 100) according to the color scales shown, the color bar is for the guard cell data only. (B) PYR/PYL Genes Act Redundantly In ABA Signaling. Genotypes shown were germinated on media containing 0.9 μ,M (+)-ABA and documented 7 days post-imbibition. (C) PYR/PYL Genes Are Required For Normal ABA-lnduced Gene Expression In Seedlings. Shown are qRT-PCR results for the ABA-responsive gene RD29. L= Ler, C = Col, Q = quadruple mutant. (D) The PYR/PYL Genes Are Required For Normal SnRK2 Kinase Activity. In-gel kinase assays were conducted on extracts made from control or ABA treated plants of the genotypes shown; extracts from two separate quadruple (Q) mutant lines are shown. Red arrow corresponds to SnRK2.2 and 2.3; Blue to SnRK2.6/OST1.

Figure 3. PYR/PYLs Bind To Group A PP2Cs In Response To ABA

(A) Wild Type and mutant PYR1-PP2C interactions. PYR1 and 3 different substitution mutants were constructed as activation domain (AD) fusion proteins and tested for their interactions with BD-HAB1 (tope panel) using the Y2H assay using the compounds shown at top. In the 2 lower panels, BD fusions of HAB1, ABI1, ABI2 or ABI2^G168D were tested for ABA/pyrabactin induced interactions with AD-PYR1. (B) ABA Promotes PYR1 - PP2C Interactions In Planta. Total protein extracts (input) were made from N. benthamiana leaves transformed with the indicated constructs/treatments, immunoprecipitated with anti-HA-agarose and immuno-detected using anti-GFP or ant-HA antibodies. (C) ABA-Orfeome Analysis Of ABI1 Interactions. Shown are subsets of an ABA-orfeome queried with ABI1; auto-activators circled. (D) Reconstitution Of ABA-Responses In Vitro. Pull-down assays using GST-HAB1 and 6×His-PYR1 were conducted using purified recombinant proteins. GST-ABI1 and ABI2 tests were done with crude lysates. 10 μM (+)-ABA was used in (A,D); (B,C) 100 μM ABA (mixed stereoisomers).

Figure 4. PYR1 Is An ABA Binding Protein That Regulates PP2C Activity

(A) PYR/PYL Proteins Determine Selectivities For Responses To Different Ligands. A panel of PYR/PYL genes were constructed as AD-fusions and tested in yeast for interactions with HAB1 and AHG3 in response to (+)-ABA, (−)-ABA, pyrabactin or apyrabactin. Shown are results for 5 PYR/PYLs that interact with HAB1 in response to ABA. (B) ABA-Response Activity Is Distributed Throughout The PYR/PYL Family. Shown is a neighbor-joining tree of the PYR/PYL family. The middle panel summarizes ligand selectivity data derived from Y2H experiments. (+)-ABA responsive PYR/PYLs are colored red, AGIs shown at right. (C) ABA Binds To PYR1 And PYR1^P88S. Shown are sub-regions of HSQC spectra for ¹⁵N-labelled PYR1 and PYR1^P88S in response to increasing amounts of ABA. Arrows indicate amide protons whose chemical environments shift in response to ABA. (D) PYR1 Inhibits PP2C Activity In The Presence Of ABA. Initial reaction velocities of recombinant GST-HAB1 were tested in the presence of PYR1 or PYR1^P88S and differing ABA concentrations using the substrate pNPP. The measured IC₅₀s are 125 nM for PYR1 and 50 μM for PYR1^P88S. (E) Hypothesized Model For PYR/PYL Control Of ABA Signaling. We propose the following model: in the absence of ABA (left), PYR/PYL proteins are not bound to PP2Cs and therefore PP2C activity is high, which prevents phosphorylation and activation of SnRK2s and downstream factors (DFs). In the presence of ABA, PYR/PYLs bind and inhibit PP2Cs. This then allows accumulation of phosphorylated downstream factors and ABA transcriptional responses.