The ubiquitin E3 ligase SR1 modulates the submergence response by degrading phosphorylated WRKY33 in Arabidopsis

Abstract

Oxygen deprivation caused by flooding activates acclimation responses to stress and restricts plant growth. After experiencing flooding stress, plants must restore normal growth; however, which genes are dynamically and precisely controlled by flooding stress remains largely unknown. Here, we show that the Arabidopsis thaliana ubiquitin E3 ligase SUBMERGENCE RESISTANT1 (SR1) regulates the stability of the transcription factor WRKY33 to modulate the submergence response. SR1 physically interacts with WRKY33 in vivo and in vitro and controls its ubiquitination and proteasomal degradation. Both the sr1 mutant and WRKY33 overexpressors exhibited enhanced submergence tolerance and enhanced expression of hypoxia-responsive genes. Genetic experiments showed that WRKY33 functions downstream of SR1 during the submergence response. Submergence induced the phosphorylation of WRKY33, which enhanced the activation of RAP2.2, a positive regulator of hypoxia-response genes. Phosphorylated WRKY33 and RAP2.2 were degraded by SR1 and the N-degron pathway during reoxygenation, respectively. Taken together, our findings reveal that the on-and-off module SR1-WRKY33-RAP2.2 is connected to the well-known N-degron pathway to regulate acclimation to submergence in Arabidopsis. These two different but related modulation cascades precisely balance submergence acclimation with normal plant growth.

Figure 1

SR1 negatively regulates the DS response in Arabidopsis. (A) Phenotypic analysis of Col, sr1, pSR1:SR1/sr1, and SR1OE plants treated with DS for 60 or 72 h, followed by 5 days of recovery. (B) Survival rates of Col, sr1, pSR1:SR1/sr1, and SR1OE plants treated with DS for 60 h, followed by 5 days of recovery. (C) DWs of Col, sr1, pSR1:SR1/sr1, and SR1OE plants treated with DS for 60 h, followed by 5 days of recovery and drying for 2 days. (D) MDA content of Col, sr1, pSR1:SR1/sr1, and SR1OE plants before submergence (Air) and after 2 days of DS (Sub) and subsequent recovery for 24 or 72 h. FW: fresh weight. (E) ROS accumulation detected by DAB staining in Col, sr1, pSR1:SR1/sr1, and SR1OE plants after treatment with or without DS for 2 or 20 h. Bar = 0.5 mm. (F–G) Total RNA was extracted from Col, sr1, and SR1OE plants treated by DS for the time periods indicated. ADH1 and PDC1 transcript levels were detected in Col, sr1, and SR1OE plants by qRT-PCR analysis. Data are average values ±SD (Standard Deviation) (n = 3) of three biological replicates (separate experiments). **P < 0.01 and *P < 0.05 indicate significant differences from Col.

Figure 2

Expression profile of SR1 and its ubiquitin E3 ligase activity. (A) qPCR analysis showing that SR1 is repressed by DS but induced by reoxygenation compared with DA or LA treatment controls. Total RNA was extracted from Col and treated by submergence or reoxygenation (Re) after submergence for the time periods indicated. Three independent biological replicates were analyzed, and similar results were obtained. Data are average values ±SD (n = 3) of three biological replicates. **P < 0.01 and *P < 0.05 indicate significant differences from the control. (B) Subcellular localization analysis of SR1. SR1-GFP and GFP (control) constructs were transformed into Arabidopsis protoplasts. GFP fluorescence was detected under a laser-scanning confocal microscope. DAPI was used as a nuclear marker. At least 10 cells were observed, and they all showed similar expression patterns. Bars = 10 μm. (C) Schematic diagram showing the key domains (RING-type zinc finger and C2H2-like zinc finger) of SR1 protein. (D, E) Assays of in vitro self-ubiquitination of SR1. “+” and “-” denote the presence or absence of the components of each reaction mixture. The molecular weight of GST-SR1N is ∼62 kDa. Protein ubiquitination bands generated by GST-SR1N are indicated on the right, and protein molecular mass markers are labeled on the left. Anti:HIS (D) and anti:GST (E) antibodies were used for immunoblot analysis. The band at 72 kDa (E) is an unspecific band.

Figure 3

SR1 interacts with WRKY33 both in vivo and in vitro. (A) Schematic diagram showing the various constructs used in the Y2H analysis. Different numbers indicate the full-length or truncated SR1 and WRKY33 proteins. (B) Y2H analysis of the interaction between SR1 and WRKY33. The full-length SR1 protein and its C-terminal and N-terminal regions were each fused with the GAL4-transcription AD. The full-length WRKY33 protein and its C-terminal and N-terminal regions were each fused with the GAL4-DNA BD. Transformants were plated on synthetic dropout (SD) medium without leucine or tryptophan (-LW) and transferred to SD medium without leucine, tryptophan, histidine, or alanine (-LWHA) to detect interactions. Here, 10⁻¹, 10⁻², and 10⁻³ indicate dilution concentrations of 10, 100, and 1000 times, respectively. (C) BiFC assay of the interaction between SR1 and WRKY33. WRKY33-nVENUS and SR1-cGFP or WRKY33-nVENUS and cCFP or SR1-cGFP and nVENUS constructs were co-transformed into Arabidopsis protoplasts to detect the interaction between SR1 and WRKY33 in vivo. YFP fluorescence was detected under a laser-scanning confocal microscope. At least 10 cells were observed, and similar results were obtained. Bars = 10 μm. (D) Co-IP to examine the interaction between SR1 and WRKY33. Proteins were isolated from N. benthamiana leaves expressing 35S:MYC-SR1N and 35S:FLAG-WRKY33 for 3 days. Anti:FLAG beads were used for the IP experiment. Anti:MYC and anti:FLAG antibodies were used for immunoblot analysis. (E) In vitro pull-down assay to examine the interaction between SR1 and WRKY33. Purified proteins (GST-SR1N, MBP-WRKY33, and MBP) were used. MBP affinity beads were used for the pull-down assay. Anti:MBP and anti:GST antibodies.

Figure 4

SR1 Promotes the degradation of WRKY33. (A) Nuclear protein levels of SR1 examined using 35S:FLAG-SR1 plants subjected to dark submergence treatment and during the reoxygenation process for the time periods indicated. (B) Nuclear protein levels of WRKY33 examined using 35S:FLAG-WRKY33 plants subjected to dark submergence treatment and during the reoxygenation process for the time periods indicated. (C) Nuclear protein levels of WRKY33 examined in 35S:FLAG-WRKY33 and 35S:FLAG-WRKY33 sr1 plants. (D) Effects of MG132, a chemical inhibitor of the 26S proteasome, on the stability of WRKY33. WRKY33 protein levels were examined in 35S:FLAG-WRKY33 and 35S:FLAG-WRKY33 sr1 plants supplied with or without 50 μM MG132 for 24 h. (E) WRKY33 showed a decrease in turnover rate in sr1 compared to Col plants. Two-week-old 35S:FLAG-WRKY33 and 35S:FLAG-WRKY33 sr1 plants were treated with 100 mM CHX for the time periods indicated. Nuclear proteins were isolated and analyzed by immunoblotting. (F) Nuclear proteins were isolated from N. benthamiana leaves after expressing 35S:FLAG-WRKY33 alone, or 35S:FLAG-WRKY33 and 35S:MYC-SR1N together for 3 days following 20 h dark submergence treatment or control treatment (dark air was used as the control(-)). Anti:FLAG and anti:MYC antibodies were used for detection. (A–F) Nuclear proteins were extracted from rosette leaves of 3-week-old transgenic plants and histone H3 was used as the internal control. The molecular weight of FLAG-SR1 is 84 kDa, FLAG-WRKY33 is 58 kDa and MYC-SR1N is 52 kDa. (G) Levels of WRKY33 ubiquitination detected in vivo by an IP experiment. N. benthamiana leaves after expressing 35S:FLAG-WRKY33 and HA empty vector or 35S:FLAG-WRKY33 and 35S:HA-ubi for 3 days were used for the IP experiment. Anti:HA and anti:FLAG antibodies were used for immunoblot analysis. The empty HA vector was used as a negative control. Protein molecular mass markers are labeled on the right. (H) Levels of WRKY33 ubiquitination detected in 35S:FLAG-WRKY33 and 35S:FLAG-WRKY33 sr1 plants. Nuclear proteins were isolated from rosette leaves of 3-week-old 35S:FLAG-WRKY33 and 35S:FLAG-WRKY33 sr1 plants followed by a FLAG-IP experiment. The levels of FLAG-WRKY33 ubiquitination were examined using an anti:ubi antibody. (I) Nuclear proteins were isolated from N. benthamiana leaves after expressing 35S:FLAG-WRKY33 without treatment, or expressing 35S:FLAG-WRKY33 and 35S:MYC-SR1N together and carrying out a 20 h dark submergence treatment or control treatment (dark air was used as the control(-)), and used in a FLAG-IP experiment. The levels of FLAG-WRKY33 ubiquitination were examined using an anti:ubi antibody.

Figure 5

WRKY33 functions downstream of SR1 to positively regulate the dark submergence response in Arabidopsis. (A) Phenotypic analysis of Col, sr1, wrky33, and sr1 wrky33 plants treated with dark submergence for 60 h, followed by 5 days of recovery. (B) Survival rates of Col, sr1, wrky33, and sr1 wrky33 plants treated with dark submergence for 60 h, followed by 5 days of recovery. (C) DWs of Col, sr1, wrky33, and sr1 wrky33 plants treated with dark submergence for 60 h, followed by 5 days of recovery and drying for 2 days. (D) ROS accumulation detected in Col, sr1, wrky33, and sr1 wrky33 plants by DAB staining after treatment with or without dark submergence for 2 or 20 h. Bar = 0.5 mm. Data are average values ±SD (n = 3) of three independent biological replicates. **P < 0.01 indicates significant differences from Col.

Figure 6

SD but Not SA overexpression enhances tolerance to dark submergence. (A) Dark submergence treatment induces the accumulation of phosphorylated WRKY33, while reoxygenation removes it (top panel shows a short exposure, middle panel shows a long exposure). Phosphorylated and non-phosphorylated WRKY33 proteins were separated using a phos-tag gel. Histone H3 was used as the internal nuclear protein loading control. The molecular weight of FLAG-WRKY33 is 58 kDa, H3 is 15 kDa. (B) Phosphorylated and non-phosphorylated WRKY33 proteins quantified using ImageJ software with the value for the first lane set as 1. The middle panel (long exposure) in Figure 6A was used to quantify protein levels. (C) WRKY33 expression levels in SDOE1, SDOE2, SAOE1, and SAOE2 plants are determined by qPCR. Total RNA was extracted from 3-week-old SDOE1, SDOE2, SAOE1, and SAOE2 plants. (D) Phenotypic analysis of Col, SDOE1, SDOE2, SAOE1, and SAOE2 plants after dark submergence treatment for 72 h, followed by 5 days of recovery. (E) Survival rates of Col, SDOE1, SDOE2, SAOE1, and SAOE2 plants determined after dark submergence treatment for 72 h, followed by 5 days of recovery. (F) DWs of Col, SDOE1, SDOE2, SAOE1, and SAOE2 plants measured after dark submergence treatment for 72 h, followed by 5 days of recovery and drying for 2 days. Data are average values ±SD (n = 3) of independent biological replicates. **P < 0.01 indicates significant differences from Col.

Figure 7

SD but not SA induces RAP2.2 expression. (A) qPCR analysis showing RAP2.2 expression levels in Col, SDOE1, SDOE2, SAOE1, and SAOE2 plants. (B, C) ChIP-qPCR analysis showing that the binding ability of FLAG-SD to the RAP2.2 promoter is comparable to that of FLAG-SA in vivo. DNA/protein complexes were isolated from 35S:FLAG-WRKY33SD/SA transgenic plants line2#. Relative enrichment of RAP2.2 promoter was determined by qPCR and calculated against input levels. (D, E) The abilities of MBP-SD and MBP-SA to bind to the promoter of RAP2.2 examined by EMSA. Here 250× and 1000× cold probes were used as competitors. (F) Comparison of the binding ability of MBP-SD and MBP-SA to the RAP2.2 promoter by EMSA. Equal amounts of MBP-SD and MBP-SA proteins were used. (D–F) 12% native gels were used to separate the free or bound DNA-protein complexes. (G) Schematic diagram of effectors (including 35S:FLAG-SR1, 35S:FLAG-SA, 35S:FLAG-SD) and reporter (proRAP2.2:LUC). (H) The reporter proRAP2.2:LUC together with the indicated effectors was co-infiltrated into N. benthamiana leaves and expressed for 3 days. LUC and REN values were then measured. The value for proRAP2.2:LUC was set to 1.0. Data are average values ±SD (n = 3) of independent biological replicates. **P < 0.01 indicates significant differences from the control.

Figure 8

Phosphorylation stimulates WRKY33 turnover and is dependent on SR1. (A) FLAG-IP experiment to examine the interaction between SR1 and SA or SD. Nuclear proteins were isolated from N. benthamiana leaves after expressing 35S:MYC-SR1N and 35S:FLAG-SA or 35S:MYC-SR1N and 35S:FLAG-SD for 3 days. Anti:FLAG antibody conjugated agarose beads were used for the IP experiment. Anti:MYC and anti:FLAG antibodies were used for immunoblot analysis. (B) Effects of MG132 on the stability of the phosphorylated form of WRKY33. Nuclear proteins were isolated from leaves of 3-week-old 35S:FLAG-WRKY33 seedlings after treatment with or without 50 μM MG132 for 24 h and detected using a 7.5% phos-tag gel. Anti:FLAG and anti:H3 antibodies were used for immunoblot analysis. (C) Ubiquitination levels of SD and SA proteins detected in vivo. Anti:FLAG antibody-conjugated agarose beads were used for the IP experiment. Anti:ubi antibody was used to detect the levels of WRKY33 ubiquitination. Anti:FLAG antibody was used to check loading levels. (D) Nuclear proteins were extracted from 3-week-old seedlings of 35S:FLAG-SD and 35S:FLAG-SA in a buffer supporting proteasome activity. Extracts were incubated at room temperature for the time periods indicated. Immunoblot analysis was then performed using anti:FLAG antibody. (E) The levels of phosphorylated WRKY33 protein detected following dark submergence treatment. Nuclear proteins were extracted from 3-week-old seedlings of 35S:FLAG-WRKY33 and 35S:FLAG-WRKY33 sr1 plants after dark submergence treatment for 20 h (dark air was used as the control(-)). Phosphorylated proteins were detected using a 7.5% phos-tag gel. Anti:FLAG and anti:H3 antibodies were used for immunoblot analysis. (F) WRKY33 protein levels were examined after 2 h dark submergence treatment and reoxygenation for 14 h. Nuclear proteins were extracted from 3-week-old rosette leaves of 35S:FLAG-WRKY33 and 35S:FLAG-WRKY33 sr1 plants. Phosphorylated proteins were detected using a 7.5% phos-tag gel. Anti:FLAG and anti:H3 antibodies were used for immunoblot analysis. The molecular weight of FLAG-WRKY33, FLAG-SA or FLAG-SD is 58 kDa, MYC-SR1N is 52 kDa.

Figure 9

Working model for the role of SR1 in regulating the dark submergence response by modulating the stability of WRKY33. Under normoxia, SR1 is constitutively expressed and SR1 is stable. WRKY33 undergoes partial degradation by SR1 to maintain dynamic equilibrium; WRKY33 can bind to the RAP2.2 promoter to maintain its constitutive expression. RAP2.2 and other members of the ERF-VII family are normally localized to the plasma membrane where they interact with the membrane-associated ACBP1 and ACBP2. Upon exposure to hypoxia induced by dark submergence, ERF-VII proteins dissociate from the membrane and are translocated into the nucleus to activate the expression of hypoxia-response genes. WRKY33 is simultaneously phosphorylated, possibly by MPK3/MPK6 (MPK3/6), and translocated into the nucleus to strongly activate RAP2.2 expression. In addition, SR1 is repressed and SR1 protein is degraded by an unknown mechanism, ensuring the stabilization of WRKY33-P. After dark submergence is replaced by reoxygenation, ERF-VII proteins are degraded via the N-degron pathway while WRKY33-P is simultaneously degraded by the rapidly accumulated SR1.