Signal‐specific Spatiotemporal Organization of AtRGS1 in Plant Pattern‐triggered Immunity
Bodan Su,A. Wang,Jinxing Lin,Daoxin Xie,Xin Shan
DOI: https://doi.org/10.1111/nph.19658
IF: 9.4
2024-01-01
New Phytologist
Abstract:The plasma membrane (PM) constitutes a regulatory platform for the transduction of external signals into the interior of cells to regulate various physiological processes. The dynamics of plant membrane proteins on the PM, as well as exocytic and endocytic trafficking between the PM and endosomes, are thought to be important determinants that control signaling activation and passivation (Wang et al., 2018; L. Zhang et al., 2019). Under different conditions, plant PM proteins exhibit various spatiotemporal behaviors, including diffusion, aggregation, and translocation on the PM (Gronnier et al., 2017, 2022; Lv et al., 2017; Cui et al., 2018; Xing et al., 2019; Su et al., 2021). Moreover, plant PM proteins often undergo ligand-induced endocytosis through specific pathways (Li et al., 2012; Fan et al., 2015). Both clathrin-mediated endocytosis (CME) and sterol-dependent endocytosis (SDE) cooperatively control the turnover of PM proteins (L. Wang et al., 2015; Mbengue et al., 2016; Xue et al., 2018; X. Zhang et al., 2019; Watkins et al., 2021; Huang et al., 2022). Pattern-triggered immunity (PTI) in plant cells is aroused through the perception of microbial or plant-derived molecular patterns (pathogen-associated molecular patterns (PAMPs) or damage-associated molecular patterns (DAMPs)) by cell surface-localized pattern recognition receptors (DeFalco & Zipfel, 2021). Flg22 and chitin are prototypical PAMPs derived from exogenous bacterial flagellin and fungal cell wall, respectively (Chinchilla et al., 2006; Miya et al., 2007). PAMP-induced peptides (PIPs) and plant elicitor peptides, two notable examples of DAMPs, are released from plant cells upon cellular damage to amplify the immune responses triggered by flg22 (Huffaker et al., 2006; Hou et al., 2014). A wide variety of PM-localized proteins with different characteristics such as receptor-like cytoplasmic kinases, NADPH oxidases, ion channels, and active transporters have been identified to be involved in orchestrating PTI responses (Yu et al., 2017). Arabidopsis thaliana regulator of G protein signaling 1 (AtRGS1), a PM-localized protein contains an RGS box at its C-terminal, contributing to GTP hydrolysis of the Gα subunit (Chen et al., 2003; Chen & Jones, 2004; Urano et al., 2012; Hu et al., 2013; Fu et al., 2014), negatively regulates PAMPs/DAMPs-triggered immune signaling (Bender & Zipfel, 2018; Liang et al., 2018; Zhong et al., 2019). It is still unclear to how AtRGS1 discriminate among patterns that are from different source organisms with different chemical nature. Here, we used various single-molecule techniques, biochemical and physiological methods to elucidate the dynamic heterogeneity and endocytic trafficking of AtRGS1 proteins in response to different PAMPs/DAMPs. We found two separate AtRGS1 pools on the PM and subsequent distinct AtRGS1 endocytosis pathways induced by different PAMPs/DAMPs. Moreover, we showed that the ligand-induced phosphorylation pattern of AtRGS1 contributes to its specific PM organization feature and endocytic model. Finally, we demonstrated that the PAMPs/DAMPs-induced endocytosis of AtRGS1 fits the context of plant immunity responses. Arabidopsis thaliana ecotype Col-0 was used as wild-type (WT) control. The gpa1-4/agb1-2 double mutant, AtRGS1-YFP/rgs1-2 (AtRGS1-YFP), AtRGS1Y112A-YFP/rgs1-2 (AtRGS1Y112A-YFP), AtRGS1ΔCt-YFP/rgs1-2 (AtRGS1ΔCt-YFP), and AtRGS13SA-YFP/Col-0 (AtRGS13SA-YFP) transgenic plants were kindly provided by Prof. Alan M. Jones. The AtRGS1-YFP/AtCLC-mCherry and AtRGS1-YFP/AtFlot1-mCherry plants were generated by genetic cross of AtRGS1-YFP with AtCLC-mCherry and AtFlot1-mCherry, respectively. The GPA1-mCherry construct was transformed into Agrobacterium strain GV3101 and then introduced into AtRGS1-YFP/rgs1-2 plants to generate stable transformants, which co-expressed AtRGS1 and AtGPA1. The AtRGS1-YFP/fls2, AtRGS1-YFP/rlk7, and AtRGS1-YFP/cerk1 plants were generated by genetic cross of AtRGS1-YFP with fls2, rlk7, and cerk1 respectively. For variable-angle total internal reflection fluorescence microscope (VA-TIRFM) and laser scanning confocal microscopy (LSCM) assays, seeds were surface-sterilized in 90% ethanol for 5 min, 70% ethanol for 10 min, and washed five times with sterile water. Seedlings were grown in ¼ liquid Murashige and Skoog (MS) medium (PhytoTech, Lenexa, KS, USA) without sucrose in a 6-well Clear Multiple Well Plate (Corning, NY, USA) for 4–6 d in the dark after 2-d vernalization. For reactive oxygen species (ROS) burst, mitogen-activated protein kinases (MAPK) activity, and root length assay, seeds were surface-sterilized in 20% (v/v) bleach with a sharp shake for 8 min and then washed five times with sterile water. After vernalization at 4°C for 2 d, seeds were plated on ½MS medium and transferred into a growth room at 22°C for growth and subsequent experiments. Nicotiana benthamiana (N. benthamiana) plants were grown in a growth room at 22°C with a 16 h : 8 h, light : dark photoperiod. The AtRGS1-GFP and AtGPA1-mCherry plant expression vectors were generated previously (Li et al., 2019). The AtFlot1-mCherry and AtCLC-mCherry plant expression vectors were kindly provided by Prof. Jinxing Lin (L. Wang et al., 2015). Agrobacterium tumefaciens strains carrying specific plasmids were mixed with equal OD value and injected into 7-wk-old N. benthamiana leaves at 22°C for 24 h in the dark and transferred into light condition for 12–24 h before imaging. Pictures were taken after 10, 20, and 30 min of treatment with various ligands (¼MS, 10 μg ml−1 chitin, 1 μM flg22, or 100 nM PIP1) for further analysis. The concentration of chitin, flg22, PIP1, and various inhibitors used in our study were determined by referring to previous researches (Wang et al., 2013; Hao et al., 2014; Hou et al., 2014; Keinath et al., 2015; Erwig et al., 2017; Liang et al., 2018; Xue et al., 2018; Wu et al., 2020; Watkins et al., 2021). For PAMP/DAMP treatments, cytological analysis uses the same or a relatively lower concentration as defense responses analysis, while root growth inhibition analysis uses the same or a relatively higher concentration as defense responses analysis. For LSCM and VA-TIRFM assays, inhibitors and PAMPs/DAMPs were used at the following concentrations: 10 mM Methyl-beta-cyclodextrin (MβCD, 200 mM stock in distilled deionized water; Frontier Scientific, Newark, DE, USA); 50 μM Tyrphostin A23 (Tyr A23, 50 mM stock in DMSO; Sigma-Aldrich); 200 nM K252a (200 μM stock in distilled deionized water; Sigma-Aldrich); 10 μg ml−1 chitin (10 mg ml−1 stock in distilled deionized water; purified from crab shells; Sigma-Aldrich); 1 μM flg22 (10 mM stock in distilled deionized water); and 100 nM PIP1 (100 μM stock in distilled deionized water). For MAPK activation and ROS burst analysis, PAMPs/DAMPs were used at the following concentrations: 1 μM flg22, 1 μM PIP1, and 10 μg ml−1 chitin. For root growth inhibition analysis, PAMPs/DAMPs were used at the following concentrations: 10 μM flg22, 1 μM PIP1, and 10 μg ml−1 chitin. The epidermal cells of the hypocotyl of AtRGS1-YFP, AtRGS1Y112A-YFP, AtRGS1ΔCt-YFP, and AtRGS13SA-YFP transgenic plants were mounted on a specific coverslip and imaged under an Olympus VA-TIRFM using a ×100 oil-immersion lens (NA, 1.45; Olivo-Marin, 2002). VA-TIRFM contained a back-illuminated EMCCD camera (Andor Technology, Belfast, UK), a diode laser, and a BA510IF detector as described previously (X. Wang et al., 2015). Yellow fluorescent protein (YFP) was excited with 488 nm laser line while the emitted fluorescence was collected by a 525/550-nm bandpass filter. For dynamic parameter assay of single particles of AtRGS1-YFP or AtRGS13SA-YFP, image series were time-averaged by 100 ms exposure times for 100 frames. MatLab software was used to analyze SPT according to the method described by Jaqaman et al. (2008). For dwell time assay, 20 s of time-lapse images were gained with 200-ms acquisition speed (Su et al., 2023). The diffusion coefficients, motion range, particle path, particle velocity, and dwell time were measured following the previously described method (Lv et al., 2017; Su et al., 2021). Olympus FV1200 confocal microscopy with a ×63 water-immersion objective was used in this study. YFP fluorescent protein was excited at 488 nm wavelengths with an argon ion laser while detected at 505–525 nm for fluorescence emissions. mCherry fluorescent protein was excited at 561 nm wavelengths with an argon ion laser while detected at 560–640 nm for fluorescence emissions. The Olympus Fv1200 software package, Adobe Illustrator Cs5, and ImageJ software were used for image analysis. For internalization assays of AtRGS1-YFP, AtRGS1Y112A-YFP, AtRGS1ΔCt-YFP, and AtRGS13SA-YFP induced by different PAMPs/DAMPs, image acquisitions were gained on the hypocotyl epidermis c. 3 mm with a 0.5 μm interval of Z-stack series. The measurements of endocytosis ratio were described by Urano et al. (2012). Transgenic plants were always kept in darkness to avoid light-induced internalization. Quantifications of the spatial clustering index (SCI) were performed as previously described (Gronnier et al., 2017, 2022). Briefly, the SCI was calculated by dividing the mean of the 5% highest values by the mean of 5% lowest fluorescence intensities values of the 10-μm-long line plot on the maximum projection of VA-TIRFM images. Three line plots were randomly recorded per surface plane of the PM cell and at least 15 cells per experiment were analyzed under different conditions. Details of the Förster resonance energy transfer and fluorescence lifetime imaging microscopy (FLIM-FRET) methods and FRET efficiency (%) analysis were described previously (Lv et al., 2017; Su et al., 2021). Briefly, the picosecond pulsed diode laser was performed for fluorescence excitation with 488 nm laser line. The emitted light was detected by a Micro Photon Devices (MPD) Single Photon Avalanche Diode (SPAD) detector with a 520/535 nm bandpass filter. Single-pixel fluorescence lifetimes were averaged across a representative region of interest (ROI). The SymphoTime 64 software (PicoQuant, Berlin, Germany) and ImageJ were used to analyze the data. PPI analysis was performed by protein proximity analyzer software analysis (Wu et al., 2010; Zinchuk et al., 2011). Briefly, hypocotyl cells of 4-d-old transgenic seedlings expressing AtRGS1-YFP and AtFlot1-mCherry/AtCLC-mCherry or AtRGS1-YFP and AtGPA1-mCherry were taken by LSCM to choose the ROI in the YFP and mCherry images. Then they were opened in protein proximity analyzer software to analyze the cross-colocalization values. Leaves of 3-wk-old seedlings grown under short-day conditions were punched to circular with 6 mm in diameter. Then they were sliced into c. 1.5 mm strips and preincubated overnight in 200 μl sterile distilled water to eliminate wound-induced ROS in Optiplate™-96-well Microplate (Jin et al., 2022). Luminescence detection buffer containing 20 μM luminol (Sigma-Aldrich) and 10 μg ml−1 horseradish peroxidase (Sigma-Aldrich) with 1 μM flg22, 1 μM PIP1, 10 μg ml−1 chitin or water as a control was added to replace preincubated water for detection. Luminescence was monitored continuously for 25 min with a multimode plate reader (EnSpire®; PerkinElmer). Three individual leaves from different seedlings were used in each treatment. Nine-day-old seedlings on ½MS plates were treated with ½MS liquid medium containing 1 μM flg22, 1 μM PIP1, or 10 μg ml−1 chitin in a 12-well plate for 30 min. After treatment, seedlings were harvested and ground into fine powder in liquid nitrogen, then suspended in extraction buffer (50 mM Tris–HCl, pH 7.5, 150 mM NaCl, 10% (v/v) glycerol, 0.1% (v/v) Tween 20, 20 mM β-mercaptoethanol, and 1× protease inhibitor cocktail (Roche)). The homogenate was centrifuged twice at 12 000 g and supernatants were collected after centrifugation. The samples were incubated with SDS loading buffer for 8 min at 100°C and loaded onto 12% SDS-PAGE gels. Activated MAPKs were measured by immunoblotting with an anti-Phospho-p44/42 MAPK antibody (1 : 2000; Cell Signaling Technology, Danvers, MA, USA), and the actin protein was used as a loading control which was detected using an anti-actin antibody (1 : 2000; CWBIO, Jiangsu, China). For flg22 or PIP1 treatment, seeds were plated on ½MS medium with or without 10 μM flg22 and 1 μM PIP1 and grew for 9 d. For chitin treatment, 2-d-old seedlings grown on ½MS medium were transferred to ½MS medium with 10 μg ml−1 chitin and continued to grow for 7 d. Root lengths were measured by ImageJ as the pictures of seedlings were taken. The dynamic character of PM is a key determinant of the fate of membrane proteins. To characterize the kinetics of individual AtRGS1-YFP particles in vivo, a transgenic Arabidopsis line expressing AtRGS1-YFP in the rgs1-2 mutant background was used for VA-TIRFM analysis. The fluorescence signals of AtRGS1-YFP exhibited a well-dispersed diffraction-limited pattern on the PM (Fig. 1a). Then, we performed 10 s recording in 0.5 s intervals and found a dynamic heterogeneity model of AtRGS1 (Fig. 1b). Some AtRGS1-YFP particles with high fluorescence intensity stayed at certain positions for a few frames and then disappeared, whereas other spots first appeared with low fluorescence intensity, and then their fluorescence gradually increased (Fig. 1c,d). As AtRGS1 is known to function in PAMPs/DAMPs-triggered immunity signaling, we speculated whether the spatiotemporal feature of AtRGS1 on the PM is regulated by different PAMPs/DAMPs. After VA-TIRFM observation, the lateral stability of AtRGS1 particles was analyzed using MatLab and Origin8. Compared with that at steady state (0.82 ± 0.007 μm), the center values of the AtRGS1 motion range were significantly increased to 1.04 ± 0.003 μm, 1.14 ± 0.004 μm, and 1.07 ± 0.002 μm with chitin, flg22, and PIP1 treatment, respectively (Supporting Information Fig. S1). However, the diffusion coefficient of AtRGS1-YFP particles triggered by chitin, flg22, and PIP1 was markedly lower than that under normal condition (0.10 μm2 s−1, 0.06 μm2 s−1, and 0.06 μm2 s−1 vs 0.14 μm2 s−1), respectively (Figs 1e, S2a). In summary, the majority of AtRGS1 molecules were restricted in confinement zones at steady state and could be activated by different PAMP/DAMP treatments. We further detected the path of AtRGS1-YFP particles on the PM with different PAMP/DAMP stimulations using VA-TIRFM. In comparison with that at steady state (4.20 μm), the center values of the AtRGS1 particle path decreased to 3.41 μm upon chitin treatment but increased to 5.88 and 5.38 μm upon flg22 and PIP1 treatment, respectively (Fig. S3). In addition, the velocity of AtRGS1-YFP spots after treatment with chitin decreased from 2.03 μm s−1 to 1.74 μm s−1 but increased to 3.06 μm s−1 and 2.89 μm s−1 upon flg22 and PIP1 treatment, respectively (Figs 1f, S2b). Thus, AtRGS1 might form two distinct pools at the PM, depending on different PAMP/DAMP stimuli. AtRGS1 particles move farther at a relatively high speed when triggered with flg22 or PIP1, whereas treatment with chitin leads to AtRGS1 movement at a relatively low speed. Moreover, the fluorescence punctum of AtRGS1-YFP were evidently enlarged by chitin treatment compared with ligand deficiency but not by flg22 or PIP1 stimulation (Fig. 1g). To confirm that chitin treatment induces AtRGS1 aggregation, we quantitated the average size of the AtRGS1-YFP punctum during 10 s time-lapse imaging with PAMPs/DAMPs incubation. After stimulation with chitin for 30 min, the surface area of the AtRGS1-YFP punctum was significantly larger than that upon flg22 or PIP1 treatment (0.55 μm2 vs 0.48 μm2 or 0.51 μm2; Fig. S4). To further characterize the impact of different PAMPs/DAMPs on AtRGS1 organization, we analyzed the fluorescence intensity of a 10 μm line in the AtRGS1-YFP single-particle images using ImageJ software (Fig. S5a–d). The plot lines of AtRGS1-YFP upon flg22 or PIP1 treatment were smoother than those with chitin treatment (Fig. S5e). We then used the SCI, which represents protein lateral organization, to quantitatively analyze these observations (Gronnier et al., 2017). The SCI of AtRGS1-YFP for chitin treatment (5.08) was significantly higher than those under other conditions (1.42, 1.45, and 1.40 for control, flg22, and PIP1 treatments, respectively; Fig. 1h), indicating that chitin treatment promotes the clustering of AtRGS1-YFP on the PM. Taken together, two populations of differently organized AtRGS1 protein clusters provide further evidence to support the idea that different PAMPs/DAMPs activate AtRGS1 in two distinct ways. To elucidate whether the observed mobility of AtRGS1 is signaling-dependent, we constructed Arabidopsis lines expressing AtRGS1-YFP in cerk1, fls2, and rlk7 mutant backgrounds (loss of functional receptor for chitin, flg22, and PIP1, respectively), and further analyzed the spatiotemporal organization of AtRGS1-YFP on the PM. In cognate receptor mutant seedlings, the diffusion coefficient and particle velocity of AtRGS1-YFP were comparable with or without chitin, flg22, and PIP1 treatments respectively (Figs 1e,f, S2c–h). Moreover, chitin treatment did not significantly alter AtRGS1-YFP aggregation on the PM in cerk1 background (Fig. 1h). These results demonstrate that the spatiotemporal dynamics of AtRGS1 on the PM is indeed signaling-dependent. Dwell time of PM proteins is a key indicator associated with signal transmission for environmental stimuli (Flores-Otero et al., 2014). The VA-TIRFM observation showed that the ratio of dark points to red points (the relative dark points indicated shorter dwell time and the relative red points indicated longer dwell time) obviously increased when treated with different PAMPs/DAMPs (Fig. 2a). We then reconstructed a time-lapse series of VA-TIRFM images for 10 s into 3D formation and performed a single molecular (SM) kymograph analysis. In response to different PAMP/DAMP treatments, the trace of AtRGS1-YFP particles presented shorter timelines in comparison with that under steady state (Fig. 2b), suggesting a significant decrease in AtRGS1 dwell time upon ligand stimulation. To substantiate these observations, we applied MatLab software to quantify the dwell time of AtRGS1 molecules stimulated with or without PAMPs/DAMPs. In the absence of ligands, the dwell time of the AtRGS1-YFP particles was 2.17 ± 0.002 s. After incubation with chitin, flg22, and PIP1, the dwell time of AtRGS1 was significantly lower than under control condition (1.45 ± 0.001 s, 1.95 ± 0.006 s, and 1.66 ± 0.009 s, respectively; Figs 2c, S6). Thus, we concluded that different PAMP/DAMP treatments might cause the disappearance of AtRGS1 from the PM. To explore why AtRGS1 proteins disappear from the PM, we calculated the birth and death ratio of AtRGS1 particles under different conditions by analyzing 10 s time-lapse VA-TIRFM images using MatLab. Under control condition, the birth ratio was c. 34.2%, and the death ratio was 34.4%, which maintained dynamic equilibrium (Fig. S7). In response to chitin, flg22, or PIP1 treatments, the birth and death ratios of AtRGS1 also exhibited a dynamic balance (the birth ratio was 29.5% and the death ratio was 29.9% for chitin, 36.0% and 36.4% for flg22, 42.5% and 42.9% for PIP1; Fig. S7), indicating that PAMPs/DAMPs-triggered disappearance of AtRGS1 particles might be caused by the endocytosis events rather than protein aggregation on the PM. Then, we sought to determine whether PAMPs/DAMPs induce the internalization of AtRGS1. Confocal images showed that the endocytosis of AtRGS1-YFP exhibited ligand dependence (Fig. S8a). Compared with the resting-state baseline (10.8%), the internalization rate (the ratio of internal YFP intensity and total YFP intensity in a single cell) of AtRGS1 was significantly increased to 59.7%, 38.8%, and 39.1% after chitin, flg22, and PIP1 treatments, respectively (Fig. S8b). Collectively, these results demonstrate that various PAMPs/DAMPs induce AtRGS1 endocytosis. The endocytosis pathways of PM proteins in plant cells are categorized into CME and clathrin-independent endocytosis (CIE), in which SDE is the most thoroughly studied CIE pathway. To investigate whether two distinct pools of AtRGS1 induced by different PAMPs/DAMPs on the PM are individually associated with different endocytosis pathways, we used LSCM to examine PAMPs/DAMPs-triggered AtRGS1 internalization with the pretreatment of endocytosis inhibitors. Methyl-beta-cyclodextrin, which is used to block the SDE pathway by sterol depletion (Ivanov, 2008), especially affected chitin-mediated AtRGS1 endocytosis, whereas perturbing the CME pathway by competitive binding adaptor protein 2 (AP2) with Tyr A23 (Banbury et al., 2003) particularly disturbed both flg22- and PIP1-induced AtRGS1 internalization (Fig. S9). In the absence of inhibitors, the internalization ratios of AtRGS1 were 52.4% for chitin, 42.1% for flg22, and 46.2% for PIP1 incubation (Fig. 2d). After MβCD pretreatment, chitin-induced AtRGS1 internalization was reduced to approximately half (Fig. 2d). Conversely, MβCD pretreatment had no significant effect on flg22- and PIP1-induced AtRGS1 internalization (Fig. 2d). Moreover, Tyr A23 inhibited flg22- and PIP1-activated AtRGS1 endocytosis by 60.9% and 61.4%, respectively, but slightly influenced chitin-triggered AtRGS1 internalization (Fig. 2d). Therefore, these findings reveal that flg22 and PIP1 induce AtRGS1 endocytosis mainly through the CME pathway, while chitin induces AtRGS1 internalization primarily via the SDE pathway. The YXXΦ motif of the PM protein is required for the CME endocytosis pathway due to its mediated interaction with the AP2 complex (Collins et al., 2002; Yoshinari et al., 2019). To confirm two distinct endocytosis modes of AtRGS1 in response to different PAMP/DAMP treatments, we detected the internalization of AtRGS1Y112A, a YXXΦ motif mutant in which the 112th tyrosine (Y) was changed to alanine (A) (Fig. S10). Upon chitin treatment, the internalization ratio of AtRGS1Y112A-YFP was similar to that of AtRGS1-YFP (43.7% vs 48.8%; Fig. 2e). However, the endocytosis of AtRGS1Y112A-YFP was reduced to basal levels when treated with flg22 or PIP1 (17.9% for flg22 and 21.1% for PIP1; Fig. 2e). These results further support the idea that different PAMPs/DAMPs activate two distinct endocytosis pathways of AtRGS1. To corroborate these findings, we then investigated the relationship between AtRGS1 and marker proteins of the CME and SDE pathways using a dual-color transgenic line co-expressing AtRGS1-YFP and AtCLC-mCherry (a marker for CME) or AtFlot1-mCherry (a marker for SDE; Fig. S11). The PPI is a useful method for yielding good estimates of protein–protein co-localization, which reduces background signaling by combining the median filtering and advanced thresholding procedures. (Wu et al., 2010; Zinchuk et al., 2011). The mean PPI value between AtRGS1 and AtFlot1 protein increased from 0.49 to 0.60 for 10 min, 0.66 for 20 min, and 0.70 for 30 min with chitin incubation, while it exhibited no significant differences with or without flg22 or PIP1 treatment (Fig. 2f). However, the mean PPI value between AtRGS1-YFP and AtCLC-mCherry did not show a significant change with chitin treatment in 30 min of incubation but increased with flg22 or PIP1 stimulation at different time periods (from 0.44 to 0.62 for 10 min, 0.72 for 20 min, and 0.71 for 30 min of flg22 incubation; from 0.44 to 0.51 for 10 min, 0.69 for 20 min, and 0.74 for 30 min of PIP1 induction; Fig. 2g). These results provide strong evidence for chitin-induced SDE pathway and flg22- or PIP1-triggered CME pathway for AtRGS1 endocytosis. Considering the dynamics of AtRGS1 on the PM, we speculate that different PAMPs/DAMPs induce AtRGS1 endocytosis through two separate CME and SDE pathways from two signal-dependent AtRGS1 origins with distinct size and movement characters on the PM. The phosphorylation of AtRGS1 by various receptor complexes in specific amino acid sites is required for the regulation of different PAMPs/DAMPs-triggered signaling (Tunc-Ozdemir et al., 2016; Liang et al., 2018). Among these, phosphorylation of AtRGS1 is a necessary requisite for flg22-induced endocytosis and G protein activation (Tunc-Ozdemir et al., 2016). To examine the role of phosphorylation in chitin- or PIP1-induced AtRGS1 internalization, we first assessed the percentage of AtRGS1 endocytosis by treatment with serine/threonine protein kinase inhibitor, K252a. After K252a pretreatment, the PAMPs/DAMPs-activated internalization of AtRGS1 was significantly compromised (54.0% vs 21.2% for chitin, 41.3% vs 24.6% for flg22, and 33.1% vs 19.7% PIP1; Figs 2h, S12a). Then, a C-terminal truncation mutant AtRGS1∆Ct lacking the phosphorylation motif was assayed for PAMPs/DAMPs-dependent endocytosis. Unlike AtRGS1-YFP, the fluorescence of AtRGS1∆Ct-YFP mainly remained on the PM with chitin, flg22, or PIP1 treatment (Fig. S12b). The internalization rates of AtRGS1∆Ct-YFP in response to PAMPs/DAMPs were similar to those in resting condition (17.3%, 17.5%, 12.9%, and 10.7% for control, chitin, flg22, and PIP1 treatments, respectively; Fig. 2h). Together, these results suggest that the phosphorylation of amino acids at the C-terminal tail is necessary for PAMPs/DAMPs-induced AtRGS1 endocytosis. Three main serine residues (Ser428, Ser435, and Ser436) in the C-terminus of AtRGS1 are required for d-glucose- and flg22-induced endocytosis of AtRGS1 (Urano et al., 2012; Tunc-Ozdemir et al., 2016). To determine whether these residues also contribute to chitin- or PIP1-activated AtRGS1 internalization, the AtRGS13SA mutant (in which these three phosphorylation sites were mutated to alanine residues) was subjected to endocytosis analysis. As expected, flg22 failed to induce AtRGS13SA-YFP endocytosis (Fig. 2i). Meanwhile, the internalization of AtRGS13SA-YFP remained at the basal level upon treatment with PIP1 (Fig. 2i). However, chitin-induced endocytosis was slightly impaired by the AtRGS13SA mutation, and the addition of MβCD ablated AtRGS13SA-YFP internalization to the basal level (from 43.8% to 20.0%; Fig. 2j). Given that the CME and SDE pathways are specific to flg22/PIP1-induced and chitin-activated internalization of AtRGS1, respectively (Fig. 2d–g), our results suggest that the AtRGS1 trafficking paths might be associated with its phosphorylation patterns; thus, the CME pathway requires phosphorylation at Ser428/435/436 while the SDE pathway depends on other possible phosphorylation sites in the C-terminus of AtRGS1. Considering that different PAMPs/DAMPs induce distinct spatiotemporal organization of AtRGS1 on the PM, we next asked whether the aggregation of AtRGS1 might similarly rely on its phosphorylation patterns. The plot lines of AtRGS13SA-YFP displayed smoother patterns upon flg22 or PIP1 treatment compared with that with chitin treatment (Figs 2k, S13). Moreover, the SCI of AtRGS13SA-YFP showed a significant increase upon chitin treatment (3.43), surpassing those under flg22 (2.33) and PIP1 (2.28) treatments, indicating that chitin-induced organization of AtRGS1 is independent of Ser428/435/436 (Fig. 2l). Previous studies have shown that AtRGS1 endocytosis is critical for its decoupling from the heterotrimeric G protein complex to activate downstream signaling (Urano et al., 2012; Tunc-Ozdemir et al., 2016). To elucidate whether PAMPs/DAMPs-induced AtRGS1 endocytosis functions in its dissociation from the G protein complex, we assessed the hetero-oligomerization of AtRGS1/AtGPA1 with high spatial and temporal accuracy by FLIM-FRET (Fig. 3a). Combining FRET and FLIM makes it possible to analyze protein–protein interactions by determining the donor fluorescence lifetime with high spatiotemporal resolution (Wang et al., 2018). Without ligand treatment, the mean donor fluorescence lifetime (τ) of AtRGS1-GFP expressed alone in N. benthamiana leaves was 2.59 ns (Fig. 3b). As it is co-expressed with AtGPA1-mCherry, the lifetime of AtRGS1-GFP was significantly reduced to 2.37 ns at the resting condition (Fig. 3b), revealing that AtRGS1 proteins directly interact with AtGPA1 in the absence of ligands, similar to previous results (Urano et al., 2012). We further monitored the fluorescence lifetime of AtRGS1-GFP in the presence of AtGPA1-mCherry with different PAMP/DAMP treatments. Upon chitin treatment, the GFP lifetime significantly elevated to 2.53 ns at 10 min, 2.55 ns at 20 min, and 2.56 ns at 30 min (Fig. 3b). Meanwhile, the FRET efficiency (interaction pixels, IPS value) decreased obviously from 7.56% at 0 min to 2.16% at 10 min, 1.98% at 20 min, and 1.23% at 30 min (Fig. S14a), suggesting that the interaction between AtRGS1 and AtGPA1 was interrupted at an early stage in response to chitin. The AtRGS1-GFP lifetime after flg22 stimulation increased progressively to 2.47 ns at 10 min, 2.53 ns at 20 min, and 2.56 ns at 30 min (Fig. 3b), and the IPS values gradually declined from 7.56% to 4.92% at 10 min, 2.85% at 20 min, and 1.63% at 30 min (Fig. S14a). A similar tendency was obtained when PIP1 was applied, indicating that the dissociation of AtRGS1/AtGPA1 potentiated in a time-dependent manner when treated with flg22 or PIP1 (Figs 3b, S14a). Taken together, these data demonstrate that PAMPs/DAMPs could efficiently trigger the dissociation of AtRGS1 from AtRGS1/G protein complex with ligand-based manners. Moreover, we estimated the co-localization of AtRGS1 and AtGPA1 in hypocotyl cells of Arabidopsis transgenic lines co-expressing AtRGS1-YFP and AtGPA1-mCherry by PPI analysis, and found that the PPI values for AtRGS1-YFP and AtGPA1-mCherry were significantly reduced after treatments with chitin, flg22, or PIP1 (Fig. S14b). These data indicate a decreasing level of co-localization between AtRGS1 and AtGPA1 in response to different PAMP/DAMP treatments, consistent with similar results obtained with FLIM-FRET analysis. Considering that PAMPs/DAMPs-induced AtRGS1 endocytosis leads to the physical uncoupling of G protein from AtRGS1, we hypothesized that the internalization of AtRGS1 could activate G protein-mediated plant immunity. To test this possibility, we first quantitated ROS production, a rapid event of plant immune response, in the WT (Col-0), gpa1 agb1 double mutant, and AtRGS1Y112A-YFP plants upon different PAMP/DAMP treatments. As predicted, a statistically significant increase in flg22/PIP1/chitin-induced ROS burst was found in the WT plants, while the peak of ROS production was severely compromised in the gpa1 agb1 double mutant upon flg22, PIP1, and chitin treatments (Fig. 3c). Notably, to a great extent, ROS accumulation in AtRGS1Y112A-YFP plants was marred in response to flg22 or PIP1 treatment (Fig. 3c), suggesting that CME-mediated AtRGS1 endocytosis is required for flg22- and PIP1-induced ROS production. However, chitin application only partially impaired the ROS burst in AtRGS1Y112A-YFP plants (Fig. 3c), indicating that AtRGS1 internalization through the SDE pathway makes a greater contribution to chitin-induced ROS burst. Rapid and transient activation of MAPKs is a hallmark of PTI responses (DeFalco & Zipfel, 2021; Manna et al., 2023). MPK3/6, which can be activated by altering the dynamic status of phosphorylation, is a major component of canonical MAPK cascades. To test the impact of AtRGS1 endocytosis on the induction of MPK3/6 activity, we performed immunoblot analysis using an anti-Phospho-p44/42 MAPK antibody that specifically recognizes the phosphorylated forms of MPK3 and MPK6. Treatment of WT seedlings with chitin, flg22, or PIP1 led to strong activation of MPK3 and MPK6, whereas the phosphorylation of both MAPKs was severely compromised in the absence of functional AtGPA1 and AtAGB1 (Fig. 3d). Interestingly, chitin-induced activation of MPK3/6 was comparable between WT and AtRGS1Y112A-YFP seedlings, while the induction of MPK3/6 activity triggered by flg22 or PIP1 was counteracted in AtRGS1Y112A-YFP seedlings (Fig. 3d). These results confirm that the endocytosis of AtRGS1 through ligand-dependent specific routes is required for the activation of the MAPK cascade. Furthermore, we examined the primary root growth (a typical plant defense phenotype) of WT, gpa1 agb1 double mutant, and AtRGS1Y112A-YFP plants under different PAMP/DAMP treatments. The root growth of WT plants was significantly inhibited when grown in ½MS plates supplemented with chitin, flg22, or PIP1 (Figs 3e, S15). However, the attenuation of root growth was obviously weakened in gpa1 agb1 double mutant (Figs 3e, S15). Notably, the AtRGS1Y112A-YFP seedlings displayed a similar tendency to WT after chitin treatment, while they were insensitive to flg22 or PIP1 like gpa1 agb1 double mutant (Figs 3e, S15). Thus, the role of AtRGS1 endocytosis through two distinct pathways fits the context of PAMPs/DAMPs-induced root growth arrest. In conclusion, different PAMPs/DAMPs could induce the endocytosis of AtRGS1 using distinct pathways from separate PM origins, with differently sized AtRGS1 protein clusters moving at different speeds, which could promote the dissociation of AtRGS1 from the AtRGS1/G protein complex and trigger plant immunity responses (Fig. 3f). In the establishment of AtRGS1 spatiotemporal dynamic, ligand-induced phosphorylation pattern is critical for its specific PM organization and endocytic model. We thank Prof. Alan M. Jones for providing seeds of gpa1-4/agb1-2 double mutant, AtRGS1-YFP, AtRGS1Y112A-YFP, AtRGS1ΔCt-YFP, and AtRGS13SA-YFP. We thank Yan-Li Zhang and Bing-Yu Liu (Imaging Core Facility of the Protein Research Center for Technology Development, Tsinghua University) for FLIM-FRET and PPI technical assistance. We thank Li Li for her contribution to the preliminary research of this project. This work was supported by the National Key R&D Program of China (2021YFA1300400 and 2022YFD1400800), the National Natural Science Foundation of China (32250001 and 32388101), and the Postdoctoral Research Foundation of China (2021M701922). BS was supported in part by the Postdoctoral Fellowship of the Tsinghua-Peking Center for Life Sciences. None declared. XS designed the experimental strategy. BS and AW performed the research. BS analyzed the data. BS and XS wrote the manuscript. JL and DX revised the manuscript. All authors read and approved the final manuscript. BS and AW contributed equally to this work. The data that support the findings of this study are available in the Supporting Information of this article. Fig. S1 Motion range of AtRGS1-YFP particles in response to different PAMP/DAMP treatments. Fig. S2 Diffusion coefficient and particle velocity of AtRGS1-YFP particles in response to different PAMP/DAMP treatments in different mutant background. Fig. S3 Particle path of AtRGS1-YFP particles on the PM in response to different PAMP/DAMP treatments. Fig. S4 Relative frequency of surface area for AtRGS1-YFP particles on the PM under different conditions. Fig. S5 The organization of AtRGS1-YFP proteins under different conditions. Fig. S6 Distribution of AtRGS1-YFP dwell time on the PM in response to different PAMP/DAMP treatments. Fig. S7 Birth and death ratio of AtRGS1-YFP particles on the PM in response to different PAMP/DAMP treatments. Fig. S8 Endocytosis of AtRGS1-YFP proteins in response to different PAMP/DAMP treatments. Fig. S9 Typical images of AtRGS1-YFP endocytosis under different endocytosis inhibitors treatments. Fig. S10 Typical images of AtRGS1-YFP and AtRGS1Y112A-YFP endocytosis on the PM in response to different PAMP/DAMP treatments. Fig. S11 The co-localization of AtRGS1 with AtCLC or AtFlot1 in response to different PAMP/DAMP treatments. Fig. S12 Typical images of AtRGS1-YFP and AtRGS1ΔCt-YFP endocytosis in response to different PAMP/DAMP treatments with or without phosphorylation inhibitor K252a. Fig. S13 Organization of AtRGS13SA on the PM in response to different PAMP/DAMP treatments. Fig. S14 The colocalization of AtRGS1 with co-expressed AtGPA1 in response to different PAMP/DAMP treatments. Fig. S15 Root growth of WT, gpa1 agb1 double mutant, and AtRGS1Y112A-YFP plants in response to different PAMP/DAMP treatments. Please note: Wiley is not responsible for the content or functionality of any Supporting Information supplied by the authors. Any queries (other than missing material) should be directed to the New Phytologist Central Office. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.