The Arabidopsis thaliana immunophilin ROF1 directly interacts with PI(3)P and PI(3,5)P2 and affects germination under osmotic stress

Abstract

A direct interaction of the Arabidopsis thaliana immunophilin ROF1 with phosphatidylinositol-3-phosphate and phosphatidylinositol-3,5-bisphosphate was identified using a phosphatidylinositol-phosphate affinity chromatography of cell suspension extracts, combined with a mass spectrometry (nano LC ESI-MS/MS) analysis. The first FK506 binding domain was shown sufficient to bind to both phosphatidylinositol-phosphate stereoisomers. GFP-tagged ROF1 under the control of a 35S promoter was localised in the cytoplasm and the cell periphery of Nicotiana tabacum leaf explants. Immunofluorescence microscopy of Arabidopsis thaliana root tips verified its cytoplasmic localization and membrane association and showed ROF1 localization in the elongation zone which was expanded to the meristematic zone in plants grown on high salt media. Endogenous ROF1 was shown to accumulate in response to high salt treatment in Arabidopsis thaliana young leaves as well as in seedlings germinated on high salt media (0.15 and 0.2 M NaCl) at both an mRNA and protein level. Plants over-expressing ROF1, (WSROF1OE), exhibited enhanced germination under salinity stress which was significantly reduced in the rof1(-) knock out mutants and abolished in the double mutants of ROF1 and of its interacting homologue ROF2 (WSrof1(-)/2(-)). Our results show that ROF1 plays an important role in the osmotic/salt stress responses of germinating Arabidopsis thaliana seedlings and suggest its involvement in salinity stress responses through a phosphatidylinositol-phosphate related protein quality control pathway.

Figure 1. Identification of a band enriched in a PI(3,5)P₂ affinity chromatography of A. thaliana cell extracts, run on an SDS-PAGE and silver stained.

(A) 0.1S (0.1M NaCl elution of the S ion-exchange chromatography; see materials and methods) fraction of A. thaliana extract incubated with the PI(3)P and PI(3,5)P₂ columns. (B) 0.2S fraction incubated with PI(3)P, PI(3,4)P₂ and PI(3,5)P₂ columns. (C) FT collected after sequential incubation with a Q followed by an S column (Q–>S) and applied to a PI(3)P and PI(3,5)P₂ affinity chromatography. (D) Incubation of a 0.2S fraction -obtained from extracts of 0 M (C) or 0.4 M NaCl treated, culture cells - with PI(3)P and PI(3,5)P₂ columns_. Arrows indicate an enriched band in the PI(3,5)P₂ incubation (A, B and C) as well as its accumulation in both the PI(3)P and PI(3,5)P₂ incubations following salt stress of the A. thaliana cell culture (D).

Figure 2. ROF1 construct description.

(A) Domain organization of ROF1. Black shaded boxes indicate low complexity regions (source: PFAM and UNIPROT). (B) ROF1 and ROF1 truncated mutants used in GST-His fusions for protein overexpression. NROF1: truncated ROF1 missing the first 37 aminoacids of the protein (low complexity region) (starting aminoacids: QGLKKKLL). TKFD: truncated ROF1 missing the low complexity region part of the FKBD1 and the polylysine motif (starting aminoacids: TKFDSSR). 3FKTPR: truncated ROF1 missing the N- terminus sequence containing both the polylysin and the DSSRDR motives (starting aminoacids: PFKFTLG). Also see Figure S1.

Figure 3. Lipid overlay assays in order to characterize ROF1 binding to different PIP stereoisomers.

Dot blotting of different PIP stereoisomers probed with 1 µg/ml of the ROF1 protein (A) and its truncated mutant 1FK (FKBD1) (B) (Figure 2B; Figure S1) and detected with αντι-GST. (−): No competing lipid has been added in the incubation, [PI(3,5)P₂]: 50 µM of the lipid has been added during the incubation; [PI(3)P]: 50 µM of the lipid has been added during the incubation. (C) Dot blotting of different PIP stereoisomers probed with 2 µg/ml of the ROF1 truncated mutants 2FKTPR and FK3TPR (Figure 2B; Figure S1) and detected with αντι-GST. (D) Characterization of the binding site of ROF1 to PI(3)P and PI(3,5)P₂ using the N-terminus specific antibody anti-ROF1. Overexpressed ROF1 was pre-incubated with different concentrations of the antibody, as indicated, prior to its use in a lipid overlay assay and detected with anti-GST.

Figure 4. Recognition of ROF1 and ROF2 using anti-ROF1.

5-day old A. thaliana wild type, ROF1 knock-out (rof1⁻) of both Columbia and Wassilevskija background and ROF1/2 double mutants (WSrof1⁻/2⁻) were incubated at 37°C for up to 4 hrs. anti-ROF1 specifically detects the ROF1 protein in plant extracts and to a less extent a second upper band (ROF2) following heat treatment only. Columbia ROF1 knock-out belongs to the WDsLox line.

Figure 5. Gene expression and protein accumulation study for ROF1.

(A) Real time PCR of wild type A. thaliana plants treated for different time points and under different salt concentrations (0 M, 0.2M and 1M NaCl). Results are expressed as fold increase relative to the control set at 1. Control represents mock treatment with H₂O for the respective time points to the NaCl treatment. (B) Real time PCR of control (C) wild type 4-day old A. thaliana seedlings germinated on MS media and of seedlings germinated and grown on media containing NaCl or mannitol. Results are expressed as fold increase relative to the control set at 1. Control represents plants grown on MS media. Results are average of three independent experiments performed in triplicates. (C) Western blotting of wild type A. thaliana young leaves obtained from plants which have been treated for different time points and at different salt concentrations (0 M, 0.2M and 1M NaCl). Control represents mock treatment with H₂O for the respective time points to the NaCl treatment. (D) Western blotting of control (C) wild type 4 day old A. thaliana seedlings germinated on MS media and seedlings germinated and grown on MS media containing different NaCl concentrations. (E) Western blotting of WS and WSROF1OE seeds 3 hrs after water imbibition.

Figure 6. ROF1 localisation using tobacco plants transiently transformed and expressing GFP tagged ROF1 and GFP only.

(A) Control (z-stack image projection), green: ROF1-GFP, red: chloroplast autofluorescence. (B) GFP-tagged ROF1 localization and bright field imaging (single confocal section) (C) GFP localization (single confocal section). N indicates nuclei. Scale bars: 10 µm.

Figure 7. Localization of ROF1 in 4% PFA fixed A. thaliana root tips of 4 day old plants using αντι-ROF1.

(A) confocal image (z-stack projection) of untreated (control) plants; red: ROF1, blue: DAPI. (B) fluorescence microscope image of plants treated with 0.4 M NaCl for 2 mins. (C) confocal image (z-stack projection) of plants grown in 0.2 M NaCl; red: ROF1, blue: DAPI. Scale bars: 50 µm.

Figure 8. Germination efficiency of ROF1 and ROF2 mutants under osmotic/salinity stress.

(A) Germination rate of the Wassilevskija background plants WS, WSrof1⁻, WSrof2⁻, WSrof1⁻/2⁻, WSROF1CM and WSROF1OE on MS media containing 0.2 M NaCl. (B) Germination rate of the wild type Wassilevskija background plants WS, WSrof1⁻, WSrof2⁻, WSrof1⁻/2⁻, and WSROF1OE on MS media containing 0.4 M mannitol. Three independent characterized lines (in each experiment) were used for each genotype. The results show the average of three independent experiments performed for each treatment. Values are means, and bars are SDs. (C) Seeds germinated on MS media containing 0.2 M NaCl or 0.4 M mannitol (day 6, 22°C).

Figure 9. Effect of PI3K inhibitors on seed germination.

Seed germination efficiency of WSROF1OE and WSrof1⁻/2⁻ at different time points under control (A) or salinity stress (0.1 M NaCl) (B) in the presence or absence of 33 µM wortmannin or 60 µM LY294002. The results show the average of four independent experiments performed for each treatment. Values are means and bars are SDs.