Adenosine monophosphate deaminase modulates BIN2 activity through hydrogen peroxide-induced oligomerization

Abstract

The Arabidopsis thaliana GSK3-like kinase, BRASSINOSTEROID-INSENSITIVE2 (BIN2) is a key negative regulator of brassinosteroid (BR) signaling and a hub for crosstalk with other signaling pathways. However, the mechanisms controlling BIN2 activity are not well understood. Here we performed a forward genetic screen for resistance to the plant-specific GSK3 inhibitor bikinin and discovered that a mutation in the ADENOSINE MONOPHOSPHATE DEAMINASE (AMPD)/EMBRYONIC FACTOR1 (FAC1) gene reduces the sensitivity of Arabidopsis seedlings to both bikinin and BRs. Further analyses revealed that AMPD modulates BIN2 activity by regulating its oligomerization in a hydrogen peroxide (H2O2)-dependent manner. Exogenous H2O2 induced the formation of BIN2 oligomers with a decreased kinase activity and an increased sensitivity to bikinin. By contrast, AMPD activity inhibition reduced the cytosolic reactive oxygen species (ROS) levels and the amount of BIN2 oligomers, correlating with the decreased sensitivity of Arabidopsis plants to bikinin and BRs. Furthermore, we showed that BIN2 phosphorylates AMPD to possibly alter its function. Our results uncover the existence of an H2O2 homeostasis-mediated regulation loop between AMPD and BIN2 that fine-tunes the BIN2 kinase activity to control plant growth and development.

Figure 1

bres1 and bres2 mutants are insensitive to bikinin and brassinolide, but hypersensitive to brassinazole. A and C, Arabidopsis seedlings of bres1, bres2, and the parental line PIN2p:PIN2-GFP/Col-0 were germinated and grown for 5 d on agar medium supplemented with 50 μM bikinin (BIK), 100 nM brassinolide (BL), or DMSO (mock) under long-day conditions (16 h light/8 h dark cycle) (A) and on agar medium supplemented with the same chemicals as in (A) and 1 μM brassinazole (BRZ) in the dark (C). B and D, Quantification of the hypocotyl length of genotypes presented in (A) and (C). Scatter dot plots show all the individual points with the means and standard errors. P-values compared to PIN2p:PIN2-GFP/Col-0 plants. Two-way ANOVA with Dunnett’s multiple comparisons test was used, ***P < 0.001. n ≥ 40 seedlings from three independent experiments. E, Phenotypes of the bres1 and bres2 mutants and the control, PIN2p:PIN2-GFP/Col-0, grown in soil for 4 weeks. Scale bars, 1 cm (A) and (C), 2 cm (E)

Figure 2

H₂O₂ dependence of the fac1-3 sensitivity to bikinin. A, Arabidopsis seedlings of fac1-3 and PIN2p:PIN2-GFP/Col-0 germinated and grown for 5 d on agar medium supplemented with 1 mM and 2 mM H₂O₂ in the presence of 50 µM bikinin (BIK) or DMSO (mock). B, Quantification of the hypocotyl length of seedlings in (A). P-values compared to PIN2p:PIN2-GFP/Col-0. Two-way ANOVA with Dunnett’s multiple comparisons test was used, ***P < 0.001, ns, not significant. n ≥ 40 seedlings from three independent experiments. C and E, Confocal images of root tips of 5-d-old seedlings of PIN2p:PIN2-GFP/Col-0, fac1-3, and two independent transgenic lines AMPDp:gAMPD/fac1-3 stained with the H₂O₂ probe H₂O₂-BES-Ac (C) and the ROS probe 2′,7′-dichlorodihydrofluorescein diacetate (H₂DCFDA) (E). The frames indicate the region used for quantification. D and F, Quantification of the fluorescent intensities in the root tips of the seedlings in (C) and (E). n, at least 15 seedlings from three independent experiments. P-values compared to PIN2p:PIN2-GFP/Col-0. One-way ANOVA with Dunnett’s post hoc test was used, ***P < 0.001. ns, not significant. B, D, and F, Scatter dot plots show all the individual points with the means and standard errors. Scale bars, 1 cm (A), 100 µm (C and E).

Figure 3

Exogenous ATP restored the fac1-3 sensitivity to bikinin and BRs. A, Arabidopsis seedlings of fac1-3 and PIN2p:PIN2-GFP/Col-0 germinated and grown for 5 d on agar medium supplemented with 1 mM ATP or H₂O (mock for ATP) and in the presence of 50 µM bikinin (BIK), 100 nM brassinolide (BL), or DMSO (mock for BIK and BL). B, Quantification of the hypocotyl length of seedlings in (A). P-values compared to PIN2p:PIN2-GFP/Col-0 plants. Two-way ANOVA with Dunnett's multiple comparisons test was used, ***P < 0.001; ns, not significant. n ≥ 40 seedlings from three independent experiments. C, Immunoblot analysis of BES phosphorylation of seedlings in (A) with a specific anti-BES1 antibody. pBES1, phosphorylated BES1; Ø, unspecific bands used as a control. D, Quantification of the ratio between dephosphorylated BES1 and total BES1. Bar chart shows the means and standard errors. P-values compared to PIN2-GFP. Two-way ANOVA with Dunnett's multiple comparisons test was used, *P < 0.05; ***P < 0.001; ns, not significant. n, three independent experiments. E, Confocal images of root tips of 5-d-old PIN2p:PIN2-GFP/Col-0 and fac1-3 seedlings stained with the H₂O₂ probe H₂O₂-BES-Ac. The frames indicate the region used for quantification. F, Quantification of the fluorescence intensity in the root tips in (D). n, at least 15 seedlings from three independent experiments. P-values compared to PIN2p:PIN2-GFP/Col-0 (Student’s t-test). ns, not significant. B and F, Scatter dot plots show all the individual points with the means and standard errors. Scale bars, 1 cm (A) and 100 µm (D).

Figure 4

H₂O_2-induced BIN2 oligomerization. A, Coomassie Brilliant Blue (CBB)-stained non-reducing SDS-PAGE gel analysis of the oligomeric state of HIS-SUMO-BIN2 protein (20 µM) in the absence or presence of 20 mM DTT and after treatment with 5 mM H₂O₂ for 30 min. B, Size-exclusion chromatography analysis of the oligomeric state of the HIS-SUMO-BIN2 protein (20 µM) after treatment with 5 mM H₂O₂ for 30 min. C, CBB-stained SDS-PAGE gel analysis of the oligomeric state of the mutated BIN2 protein HIS-SUMO-BIN^25CS (20 µM) in the absence or presence of 20 mM DTT and after treatment with 5 mM H₂O₂ for 30 min. D, CBB-stained nonreducing SDS-PAGE gel analysis of the oligomeric state of HIS-SUMO-BIN2^C59S, HIS-SUMO-BIN2^C95S/C99S, and HIS-SUMO-BIN2^C162S after treatment with 5 mM H₂O₂ for 30 min. E, In vitro kinase assay with the HIS-SUMO-BIN2 monomer and oligomer and maltose-binding protein (MBP)-BES1 as a substrate in the absence or presence of 10 mM bikinin (BIK). The normalized band intensities are shown. Autoradiography (top) and CBB staining (bottom). Similar results were obtained in two independent experiments. F, Immunoblot analysis of BIN2-GFP and BIN2^5CS-GFP. Five-day-old seedlings of BIN2p:BIN2-GFP/bin2-3 and BIN2p:BIN2^C5S-GFP/bin2-3 were treated with 100 nM deaminoformycin (DF) or H₂O (MOCK) for 24 h at 21 °C or heated at 42 °C for 1 h in liquid medium with 24 h MG132 co-treatment or pre-treatment. G, Quantification of the ratio between BIN2 oligomer and total BIN2. Bar chart shows the means, standard errors, and all individual points. P-values were compared to MOCK with one-way ANOVA with Dunnett's post hoc test, **P < 0.01, ***P < 0.001. n, three independent experiments. H, Representative images of two independent Arabidopsis transgenic seedlings, each expressing either BIN2p:BIN2-GFP/bin2-3 or BIN2p:BIN2^C5S-GFP/bin2-3 and Col-0 grown for 12 d on agar medium (top) or 25 d in soil (bottom). Scale bars, 1 cm.

Figure 5

BIN2 regulates H₂O₂ homeostasis probably through AMPD. A, AMPD-GFP coimmunoprecipitated with HA-tagged AtSKs in N. benthamiana. AMPD-GFP was immunoprecipitated with anti-GFP Trap beads. AMPD-GFP and AtSKs-HA were detected with an anti-GFP and anti-HA antibody, respectively. IP, immunoprecipitation; IB, immunoblot. B, In vitro kinase assay for HIS-SUMO-BIN2 with HIS-SUMO-AMPD^32-839 and with HIS-SUMO as negative control. Autoradiography (top) and CBB staining (bottom). C, Confocal images of root tips of 5-d-old PIN2p:PIN2-GFP/Col-0 and fac1-3 plants stained with H₂O₂-BES-Ac (30 min) after 24 h of treatment with 50 µM bikinin (BIK), 10 nM brassinolide (BL), and DMSO (mock). D, Quantification of fluorescent intensities in the root tips of the seedlings in (C). Scatter dot plots show all the individual points with the means and standard errors. Significant differences were determined with two-way ANOVA with Tukey’s multiple comparisons test and labeled with different letters (P < 0.05). n, at least 15 seedlings from three independent experiment. AtSK, A. thaliana Shaggy/GSK3-like kinase. Scale bar, 50 µm (C).

Figure 6

Model for AMPD-dependent BIN2 regulation. A, AMPD facilitates the production of ROS and hydrogen peroxide (H₂O₂), which induces the oligomerization of BIN2. Compared to the BIN2 monomer, the oligomer is less active but more sensitive to bikinin. Moreover, BIN2 phosphorylates AMPD and probably regulates its function. The exogenous H₂O₂ and ATP could increase the endogenous H2O2 levels. B, When the AMPD function is impaired by mutation or by its chemical inhibitor deaminoformycin (DF), the amount of H₂O₂ is reduced, leading to the monomerization of BIN2 possibly by an unknown thioredoxin. Compared to the oligomeric protein, the BIN2 monomer is more active but less sensitive to bikinin, probably causing the bikinin and partially the BR insensitivity in the ampd mutant or DF-treated wild-type plants. The size of the circles for H₂O₂ and of the arrows indicates their levels and strength of promotion or inhibition strength, respectively. P2K1, PURINOCEPTOR P2K1.