Journal: The EMBO Journal

Article Title: An actin remodeling role for Arabidopsis processing bodies revealed by their proximity interactome

doi: 10.15252/embj.2022111885

Figure Lengend Snippet: The DCP1‐TurboID‐HF construct efficiently rescues the adult dcp1‐3 mutant phenotype. Phenotypes (3‐week‐old) of adult plants expressing 35Spro:DCP1‐TurboID‐HF and RPS5apro:HF‐mScarlet‐DCP1 in the dcp1‐3 mutant background. In a semi‐controlled greenhouse setting (temperature control 22°C), the dcp1‐3 mutant showed a smaller adult stature. Lower: complementation quantification (rosette diameter), N , biological replicates = 4, n = ( pooled data of 3 biological replicates ) 5–7 rosettes, error bars are (mean + SD) and RT‐qPCR analyses for the quantification of DCP1 expression levels. PP2A and Actin7 were used for normalization (1‐week‐old) seedlings ( N , biological replicates = 2, n ( pooled data of 3 biological replicates ) = 3, error bars are mean + SD). When DCP1 was driven by the RPS5a promoter (stem cell‐specific promoter), we observed partial complementation, suggesting that DCP1 is important also in non‐meristematic cells. Immunoblot analyses of lines expressing sGFP‐TurboID‐HF or DCP1‐TurboID ‐HF upon NS or HS conditions (same as used for APEAL). The immunoblots also show the accumulation of auto‐biotinylated sGFP‐TurboID‐HF or DCP1‐TurboID‐HF in a time course of biotin administration (as detected with streptavidin‐HRP, which captures biotinylated proteins). 50 μM Biotin was delivered in leaves by syringe infiltration and diffusion. Note that HS did not increase the biotinylation efficiency of DCP1‐TurboID‐HF: at 24 h, compare samples “2” with “4” in the “DynaIP”. HF, 6xhis‐3xFLAG. The red arrowhead indicates the position of DCP1‐TurboID‐HF and green for sGFP‐TurboID‐HF ( N , biological replicates = 2). We used the same scheme for biotin application as determined in (Arora et al , ). The 2 h NS/HS corresponds to the timing after the administration of biotin for 24 h ( t = 0 corresponds to 24 h biotin administration in NS conditions). The red asterisk indicates non‐specific band. α‐FLAG was used for the detection of DCP1‐TurboID‐HF and sGFP‐TurboID‐HF (similar size ~ 130 kDa). α‐Tubulin and ponceau staining were used for loading control. Representative confocal micrographs showing that the localization of DCP1 (α‐FLAG, 5‐day‐old seedlings, root cap cells) to PBs is retained in lines expressing DCP1‐TurboID‐HF . As a control, the GFP‐TurboID‐HF line was used that does not show localization to PBs. Right: number of DCP1‐positive foci in the corresponding lines expressing DCP1 ( N , biological replicates = 1, n = 32–60 cells). Data information: In (A and B), P values were determined by ordinary one‐way ANOVA. Upper and lower lines in the violin plots when visible, represent the first and third quantiles, respectively, horizontal lines mark the median and whiskers mark the highest and lowest values. Source data are available online for this figure.

Article Snippet: Monoclonal Mouse α‐Red Fluorescent Protein (RFP) (clone RF5R) , Agrisera , AS15 3028.

Techniques: Construct, Mutagenesis, Expressing, Control, Quantitative RT-PCR, Western Blot, Diffusion-based Assay, Staining

Journal: The EMBO Journal

Article Title: An actin remodeling role for Arabidopsis processing bodies revealed by their proximity interactome

doi: 10.15252/embj.2022111885

Figure Lengend Snippet: Bimolecular fluorescence complementation (BiFC) assay of the indicated proteins. Left: the cartoon depicts the BiFC concept and YFP reconstitution caused by protein–protein interaction in vivo . When two proteins interact, the cYFP and nYFP halves are brought in proximity and produce a fluorescent signal. For protein selection, we classified the associations with DCP1, in both AP and PDL steps, according to their relative enrichment (log 2 FC). We selected proteins presenting moderate relative enrichment in the AP step, while having high predictability in the PDL step (log 2 FC > 0 in both steps). As an additional filter, we selected proteins rich in IDRs (half of the proteins from the list have high prion‐like domain (PRD) scores and high Finum [aa]/total [aa] ratios as defined through the PLAAC algorithm; Source File 8), and as such proteins would be likely to localize to condensates. We tested the association between PBs (using DCP1 as one representative component) and selected five proteins from these bins using BiFC and colocalization assays (Source File 8). These results suggested that the APEAL approach may help identify PB components or DCP1 interactors. These interactions should be further studied in Arabidopsis . BiFC efficiency was estimated from the reconstituted YFP raw signal intensity and YFP‐positive puncta per cell in maximum projections ( N , biological replicates = 3, n = 4–12 cells). Lower left: representative confocal micrographs showing that YFP signal is reconstituted at cellular puncta that most likely correspond to PBs. XRN3 represents negative control (see also B), as it localizes in the nucleus. Right: number of cellular puncta per total cell volume (in maximum projection images; N , biological replicates = 3, n ( pooled data of 3 biological replicates ) = 20 cells, error bars are mean + SD). Scale bars: 10 μm. Colocalization of selected proteins with DCP1‐CFP‐positive puncta ( 35Spro:DCP1‐CFP transgene). The coding sequences of the corresponding “interactors” (PPIs; direct or indirect, defined in APEAL) were driven by the 35Spro and cloned in frame with mCherry at their 5′ end. Two‐color colocalizations were estimated by Pearson's correlation coefficients (PCC) using ultra‐fast super‐resolution microscopy combined with image deconvolution (~ 120 nm axial resolution). Numbers in “merge” indicate colocalization frequency between DCP1‐CFP and the corresponding interacting protein ( N , biological replicates = 2, n = 5 cells). Yellow arrowheads indicate colocalization and red arrowheads lack of colocalization. Lower right: PCC of pixel intensities between DCP1‐CFP and the corresponding putative interacting protein ( N , biological replicates = 3, n = 4–12 cells, error bars are mean + SD). We confirmed the ECT domain‐containing proteins, MLN51, FLXL1, EIN2, VAP27‐1, uncharacterized AT1G33050 (hypothetical protein), and AT5G53330/AT2G26020 (ubiquitin‐associated/translation elongation factors EF1B) as novel PB components. By applying a pre‐selection criterion of enrichment (log 2 FC > 0.5) for the selection of prey, we significantly increased the probability of identifying successful binary interactions between PBs (i.e., cYFP‐DCP1) and the identified proteins. All preys were confirmed as PB components, while PDL had higher interaction predictive power than the AP step irrespective of whether proteins were enriched in NS or HS. XRN3 was used as a threshold control (log 2 FC = −0.58, PDL/NS conditions). The heatmap shows the enrichment of these proteins in the different conditions; the scale at right shows log 2 FC. Data information: In (A and B), P values were determined by ordinary one‐way ANOVA (differences were all calculated compared to XRN3). Source data are available online for this figure.

Article Snippet: Monoclonal Mouse α‐Red Fluorescent Protein (RFP) (clone RF5R) , Agrisera , AS15 3028.

Techniques: Bimolecular Fluorescence Complementation Assay, In Vivo, Selection, Negative Control, Clone Assay, Super-Resolution Microscopy, Ubiquitin Proteomics, Control

Journal: The EMBO Journal

Article Title: An actin remodeling role for Arabidopsis processing bodies revealed by their proximity interactome

doi: 10.15252/embj.2022111885

Figure Lengend Snippet:

Article Snippet: Monoclonal Mouse α‐Red Fluorescent Protein (RFP) (clone RF5R) , Agrisera , AS15 3028.

Techniques: Subcloning, CRISPR, Mutagenesis, Recombinant, Sequencing, Cloning, Magnetic Beads, In Situ, Affinity Purification, Fluorescence, Plasmid Preparation, Staining, Western Blot, Software, Microscopy, Gel Extraction, Mass Spectrometry