{"paper_id":"b5a877b2-322e-4686-9560-ee9b49dc9d93","body_text":"Single-cell-resolved calcium and organelle dynamics in resistosome-mediated \ncell death \n \n \nYi-Feng Chen1, Kuan-Yu Lin1, Ching-Yi Huang1, Liang-Yu Hou1, Enoch Lok Him Yuen2, \nWei-Che Sun1, Bing-Jen Chiang1, Chin-Wen Chang1, Hung-Yu Wang1, Tolga O. Bozkurt2, \nChih-Hang Wu1* \n \n \n1 Institute of Plant and Microbial Biology, Academia Sinica, Taipei 115201, Taiwan. \n2 Department of Life Sciences, Imperial College London, London SW7 2AZ, UK. \n*Corresponding author: wuchh@gate.sinica.edu.tw \n \n \nAbstract \nPlant nucleotide-binding domain leucine-rich repeat-containing (NLR) proteins act as \nintracellular immune receptors that assemble into resistosomes to execute immune \nresponses. However, the subcellular processes during cell death following resistosome \nactivation remain unclear. Here, we visualized the changes in calcium signaling and \norganelle behavior after activation of the NRC4 (NLR-required for cell death 4) resistosome. \nWe found that NRC4 membrane enrichment coincided with calcium influx. This is followed \nby sequential mitochondria and plastid disruption, endoplasmic reticulum fragmentation and \ncytoskeleton depolymerization. Subsequent loss of plasma membrane integrity, nuclear \nshrinkage, and vacuolar collapse mark the terminal stage of cell death. Our findings reveal a \nspatiotemporally-resolved cascade of subcellular events downstream of resistosome \nactivation, providing new mechanistic insight into the execution phase of plant immune cell \ndeath.  \n \nKeywords: nucleotide-binding domain leucine-rich repeat-containing (NLR), NLR-required \nfor cell death (NRC), hypersensitive cell death, calcium, organelles \n \n1 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted July 1, 2025. ; https://doi.org/10.1101/2025.06.27.662017doi: bioRxiv preprint \n\n \nIntroduction \n \nNucleotide-binding domain leucine-rich repeat receptors (NLRs) play critical roles in the \ninnate immunity of both plants and animals (Duxbury et al. 2021). In animals, activated NLRs \nform inflammasomes that guide Gasdermin oligomers to the plasma membrane, resulting in \npyroptotic cell death and the elimination of pathogens. In plants, activated NLRs form \noligomeric structures known as resistosomes (Duxbury et al. 2021; Chai et al. 2023). These \nresistosomes can function as  enzymes that generate small secondary messenger \nmolecules or as calcium channels targeting the plasma membrane, ultimately triggering \nhypersensitive cell death to restrict pathogen invasion (Chai et al. 2023).  \n \nThe NLR required for cell death (NRC) family comprises multiple helper NLRs with partial \nfunctional redundancy, acting downstream of several sensor NLRs (Wu et al. 2017). For \nexample, the Nicotiana benthamiana helper NLR NRC4 works with both the Potato virus X \n(PVX) resistance protein Rx and the late blight resistance protein Rpi-blb2, whereas NRC2 \nfunctions with Rx but not Rpi-blb2. Upon activation, NRCs form high molecular weight \nresistosome complexes and localize at the plasma membrane as puncta (Duggan et al. \n2021; Ahn et al. 2023; Contreras et al. 2023). Recent cryogenic electron microscopy \n(cryo-EM) studies showed that NRC2 and NRC4 exist as autoinhibited dimers at the resting \nstate (Ma et al. 2024; Selvaraj et al. 2024). Upon activation by sensor NLRs or autoactivation \nmutations, NRCs oligomerize into hexameric resistosome complexes (Liu et al. 2024; \nMadhuprakash et al. 2024). Activated NRCs likely function as calcium channels, triggering a \ncalcium influx into the cytosol and initiating downstream immune responses (Liu et al. 2024). \nHowever, how resistosome activation and calcium signaling drive subcellular reorganization \nand culminate in cell death remains poorly understood. \n \nA major obstacle in dissecting these events has been the rapid onset of hypersensitive cell \ndeath following resistosome activation. Consequently, most studies have relied on \ndeath-deficient NLR variants or immune-compromised mutant backgrounds, limiting the \nability to monitor dynamic cellular changes during the execution phase. To overcome this, we \nemployed a recently developed copper-inducible transient expression system in N. \nbenthamiana (Chiang et al. 2024), allowing us to temporally control resistosome formation \nand perform live-cell imaging to uncover the subcellular events that accompany \nNRC4-mediated cell death.  \n \n \n \n2 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted July 1, 2025. ; https://doi.org/10.1101/2025.06.27.662017doi: bioRxiv preprint \n\n \nResults \n \nNRC4 forms the resistosome complex and triggers cell death within three hours of \ncopper-induced effector expression \n \nTo resolve the timing of cell death under the copper-inducible system, we established a leaf \ndisc-based cell death assay using autofluorescence as a readout. We transiently expressed \nRpi-blb2 along with either inducible AVRblb2 from Phytophthora infestans or inducible CP \nfrom PVX in N. benthamiana leaves. Two days post-agroinfiltration, we infiltrated copper \nsolution into the leaves and punched leaf discs for real-time fluorescence measurement in a \nplate reader (Fig. S1A). Autofluorescence in samples expressing copper-inducible AVRblb2 \nrose immediately and saturated ~3 h post-copper infiltration (hpci), indicating that Rpi-blb2 \ntriggers cell death efficiently within this window (Fig. 1A). By contrast, samples expressing \nnegative control, the copper-inducible CP, did not show any considerable increase in \nfluorescence. The AVRblb2-induced autofluorescence signal was abolished in the nrc2/3/4 \ntriple knockout (nrc) background but was restored by expression of wild-type NRC4 (Fig. \n1B), whereas previously characterized cell death deficient NRC4 L9E, NRC4 K190R mutants or \nNRC2 failed to complement the phenotype (Fig. 1B). These results are consistent with the \nvisible cell death phenotype observed at 2 days post copper infiltration (dpci) (Fig. S1B and \nS1C). \n \nTo assess cell viability under a confocal microscope, we transiently expressed a blue \nfluorescent protein (mTagBFP2) fused to nuclear localization signal (NLS) and performed \npropidium iodide (PI) staining. Time-course experiments showed that the percentage of \nepidermal cells with PI-positive nuclei (indicating dead cells) increased over time, while the \npercentage of cells displaying nuclear BFP signal (indicating live cells) decreased \ncorrespondingly (Fig. 1C and 1D). To determine whether NRC4 high molecular weight \nresistosome complexes form within this timeframe, we conducted time-course native-PAGE \nassays using the NRC4 L9A/V10A/L14A variant (hereafter referred to as NRC4 3A) (Wang et al. \n2025). NRC4 3A resistosome high molecular weight complexes became detectable as early \nas 0.5 hours post-induction and continued to accumulate at 3 and 6 hours (Fig. 1E). \nAdditionally, we observed that activated NRC4 3A-GFP formed puncta on the plasma \nmembrane (Fig. 1F). A time-course analysis revealed that the proportion of cells exhibiting \nNRC43A puncta increased over time, reaching approximately 80% by 3 hours post-AVRblb2 \ninduction (Fig. 1G). Collectively, these findings demonstrate that Rpi-blb2/AVRblb2-triggered, \nNRC4-dependent cell death can be efficiently and reproducibly activated using the \ncopper-inducible system.  \n3 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted July 1, 2025. ; https://doi.org/10.1101/2025.06.27.662017doi: bioRxiv preprint \n\n \n \nFig. 1. Copper-inducible effector expression triggers rapid hypersensitive cell death \nand NRC4 resistosome formation in N. benthamiana.  \nA. Quantification of cell death in a leaf disc-based assay upon copper-inducible effector expression. Leaves were \nco-infiltrated with 35S::Rpi-blb2 and a copper-inducible effector construct (AVRblb2  or CP). Autofluorescence \nwas measured using a plate reader following copper infiltration. Fluorescence intensities were normalized to \nmock-treated controls. Solid lines represent mean values; shaded areas indicate standard error (n = 34–36 discs \nfrom 3 independent experiments).  \nB.Complementation assay in nrc2/3/4 triple knockout (nrc) N. benthamiana plants. NRC4, NRC4 mutants (L9E, \nK190R), or NRC2 were expressed along with Rpi-blb2 and inducible AVRblb2. Autofluorescence was measured \nas in (A). \nC. Representative confocal images showing a live cell (top) with nuclear BFP (BFP-NLS) and a dead cell (bottom) \nstained with propidium iodide (PI). Scale bar = 30 μm.  \nD. Time-course quantification of live (BFP-NLS-positive) and dead (PI-positive) cells following copper treatment. \nBox plots show the percentage of total cells that are either live or dead at each time point. Data represent three \nindependent biological replicates (n = 18 images). \nE. Time-course analysis of NRC4 3A-HF resistosome assembly. Blue native PAGE (BN-PAGE) shows the \naccumulation of high molecular weight NRC4 3A-HF complexes after AVRblb2/RFP-Rpi-blb2 activation at the \nindicated hours post copper infiltration (hpci). SDS-PAGE was used to assess total NRC4 3A-HF protein as a \nloading control.  \nF. Confocal Z-stack maximum projection showing subcellular localization of NRC4 3A-GFP at 0 and 3 hpci. Gray: \nNRC43A-GFP; magenta: plastid autofluorescence. Leaves were co-expressed with 35S::NRC43A-GFP, \n35S::Rpi-blb2, and CBS4::AVRblb2 in the nrc background. Images were taken 2 days post infiltration. Scale bar = \n20 μm.  \nG. Time-course quantification of NRC4 3A-GFP puncta formation. Box plots show the percentage of cells with \nvisible NRC4 3A-GFP puncta at each time point post copper infiltration. Data represent three independent \nbiological replicates (n=16-19 images).  \n \n \n4 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted July 1, 2025. ; https://doi.org/10.1101/2025.06.27.662017doi: bioRxiv preprint \n\n \nActivated wild-type NRC4 forms puncta and filamentous structures prior to cell \ncollapse \n \nNext, we used the inducible system to investigate the subcellular dynamics of NRC4 \nactivation using time-lapse imaging. First, we compared the dynamics of NRC4 3A and NRC4 \nupon activation by Rpi-blb2 and AVRblb2. We transiently expressed NRC4-GFP variants, \nuntagged Rpi-blb2, and inducible AVRblb2 in N. benthamiana leaves. Two days  after \nagroinfiltration, we infiltrated copper solution into the leaves to induce effector expression. \nLeaf discs were then immediately prepared for confocal microscopy to observe NRC4 \ndynamics (Fig. S1A). Since NRC4 forms puncta at the plasma membrane, we adjusted the \nfocal plane to the cell periphery for time-lapse imaging. At the early stage of time-lapse \nimaging, NRC4 3A-GFP displayed highly dynamic behavior reflecting its diffused cytosolic \ndistribution at the resting state. Around 30 minutes to an hour (timing varied between cells) \npost induction, small punctate structures began to appear. These puncta were immobile and \ngradually increased in intensity over time, likely reflecting the clustering of resistosome \ncomplexes at the plasma membrane (Fig. 2A, 2C, S2A, S3A; Movie S1). Experiments using \ncytosolic mRFP revealed that the cytosol remains dynamic following the formation of \nNRC43A-GFP puncta (Movie S2), with reduced GFP/mRFP co-localization coefficient value \npost-NRC4 puncta formation (Fig. S3B). Fluorescence recovery after photobleaching \n(FRAP) assays confirmed that these NRC43A puncta are not mobile (Fig. S3C-D). \n \nWe then conducted time-lapse imaging with NRC4-GFP. Similar to NRC4 3A, NRC4 exhibits \ndynamic cytosolic movement initially. However, unlike NRC4 3A, NRC4-GFP abruptly formed \npuncta along with filament-like structures (Fig. 2B, 2D, S2B; Movie S3). These puncta and \nfilaments remained immobile until the cell collapsed (Fig. 2D; Movie S3). Cytoplasmic \nstreaming ceased at the onset of puncta/filament formation (Movie S4), coinciding with a \nsharp drop in the co-localization coefficient value between NRC4-GFP and cytosolic mRFP \n(Fig. S2E). Within 15–30 minutes following the appearance of these structures, the plasma \nmembrane detached from the cell wall, overall fluorescence weakened, and the NRC4–GFP \nsignal disappeared (Fig. S2B, 2E; Movie S3). \n  \n \n \n5 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted July 1, 2025. ; https://doi.org/10.1101/2025.06.27.662017doi: bioRxiv preprint \n\n \nFig. 2. NRC4 forms resistosome puncta and filamentous structures prior to cell \ncollapse. \nA. and B. Time-lapse confocal images showing subcellular localization dynamics of (A) NRC4 3A-GFP and (B) \nNRC4-GFP following copper-induced expression of AVRblb2. N. benthamiana nrc plants were co-infiltrated with \n35S::Rpi-blb2, CBS4::AVRblb2, and 35S::CUP2-p65. At 2 days post infiltration, copper solution was applied to \ninduce effector expression, followed by confocal imaging at the cell periphery. White lines in the final panels \nindicate regions of interest (ROIs) used for intensity profiling in (C) and (D). Scale bar = 5 μm. (See Movies S1 \nand S3.)  \nC. and D. Time-course fluorescence intensity profiles of (C) NRC4 3A-GFP and (D) NRC4-GFP along the ROIs \nmarked in (A) and (B), respectively. Each line represents fluorescence intensity across the ROI at the indicated \ntime point. A pseudocolor gradient reflects the progression of time.  \nE. Bright-field time-lapse images showing morphological changes during NRC4-GFP–induced cell collapse. Solid \nand dotted lines outline the upper and lower cell boundaries, respectively. White arrowheads indicate vacuolar \nshrinkage. Scale bar = 10 μm. (See Movie S3.) \n \n \n \n \n6 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted July 1, 2025. ; https://doi.org/10.1101/2025.06.27.662017doi: bioRxiv preprint \n\n \nCytoplasmic calcium influx coincides with NRC4 membrane enrichment \nPrevious studies have shown that NRC4 activation induces calcium signaling in leaves (Liu \net al. 2024), but the kinetics of this influx at single-cell resolution have remained unclear. To \naddress this, we generated a stable transgenic N. benthamiana line expressing the Ca²⁺  \nreporter GCaMP6 and performed copper-inducible cell-death assays described above, \nimaging calcium dynamics in individual cells (Fig. S1A) (Chen et al. 2013). Roughly 30–60 \nmin after copper induction (varying between cells), GCaMP6 fluorescence rose sharply, \nindicating a rapid increase in cytosolic and nuclear Ca²⁺  (Fig. 3A-B, S4A, S5A; Movie S5). \nThe calcium signal peaked within 3-5 minutes of influx onset and returned to baseline over \nthe next 3-5 minutes (Fig. 3A-B, S5A, and Fig. S5B). Then, rapid cell collapse occurred, \nmarked by the detachment of the plasma membrane from the cell wall at approximately 15 \nminutes or longer after the calcium peak (Fig. 3A-B; Movie S5). Thus, we conclude that \nNRC4 activation elicits a transient Ca²⁺  influx that precedes the onset of cell death. To \nconfirm that this influx requires NRC4, we repeated the assay in nrc2/3/4 knockout GCaMP6 \nplants. Wild-type NRC4 restored both the transient calcium influx and cell death, whereas \nthe NRC4 L9E mutant did not (Fig. S5C), demonstrating that the observed responses are \ndependent on functional NRC4. \nTo investigate the timing of calcium influx relative to NRC4 puncta formation, we co-imaged \nNRC4–GFP with the red-shifted calcium reporter RCaMP1h (Akerboom et al. 2013). To our \nsurprise, visible NRC4 puncta appeared only after the calcium signal had already peaked \n(Fig. 3C, S4B; Movie S6). This prompted us to hypothesize that NRC4 resistosome are \nactivated before puncta become detectable. To further investigate this, we quantified RCaMP \nand NRC4 fluorescence at the plasma membrane and found that the initial enrichment of \nNRC4 at the membrane coincided with the onset of calcium influx (Fig. 3D, S4B and Fig. \nS5D). NRC4 then reached a peak at the membrane and subsequently coalesced into puncta \nas the calcium signal declined. These data suggest that functional resistosomes assemble \nduring the initial membrane‐ enrichment phase and that the later visible larger puncta \nrepresent post-activation clusters of multiple complexes rather than individual active \nresistosomes. \n \n7 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted July 1, 2025. ; https://doi.org/10.1101/2025.06.27.662017doi: bioRxiv preprint \n\n \n \nFigure 3. Activation of NRC4 resistosomes induces calcium influx into the cytoplasm. \nA. Time-lapse images showing cytosolic calcium dynamics during hypersensitive cell death. Leaves of \nGCaMP6-expressing N. benthamiana  were infiltrated with 35S::Rpi-blb2 , CBS4::AVRblb2, and 35S::CUP2-p65. \nTwo days post infiltration, copper solution was applied to induce effector expression, followed by confocal \nimaging. Top: pseudocolored GCaMP6 fluorescence representing calcium intensity; bottom: corresponding \nbright-field images. Solid and dotted lines outline the right and the left cell boundaries. Images were extracted at \n7-minute intervals from the time series in (B). Scale bar = 5\n \nμm. (See Movie S5.) \nB. Quantification of GCaMP6 fluorescence intensity and plasma membrane shrinkage during hypersensitive cell \ndeath. GCaMP6 intensity was normalized to the mean intensity of 50 baseline frames recorded before calcium \nsignal initiation. Membrane–cell wall distance was measured using the TrackMate plugin in ImageJ. Bottom: \nkymograph of the GCaMP6 signal from the time-lapse series in (A).  \nC.  Coordinated dynamics of cytosolic calcium (RCaMP1h) and plasma membrane-localized NRC4-GFP during \nhypersensitive cell death. N. benthamiana  leaves were infiltrated with 35S::Rpi-blb2, CBS4::AVRblb2, \n35S::CUP2-p65, 35S::RCaMP1h, and 35S::NRC4-GFP. White lines outline the cell boundaries. Top: \npseudocolored RCaMP1h signal; bottom: pseudocolored NRC4-GFP signal. Open arrowheads mark initial NRC4 \nmembrane enrichment; filled arrowheads indicate puncta formation. Images were extracted at 2.8-minute \nintervals from the time series. Scale bar = 5 μm. (See Movie S6.)  \n8 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted July 1, 2025. ; https://doi.org/10.1101/2025.06.27.662017doi: bioRxiv preprint \n\n \nD. Quantification of RCaMP1h and NRC4-GFP fluorescence intensities during cell death progression. Traces \nwere normalized to the mean intensity of 50 baseline frames recorded before calcium signal initiation. The inset \nshows a zoomed-in time window around calcium signal onset and decline. Open and filled arrowheads denote \nNRC4 membrane enrichment and puncta formation, respectively. Bottom: kymographs of RCaMP1h and \nNRC4-GFP signals from the time-lapse series in (C).  \n  \nNRC4 resistosome activation halts vesicle, mitochondrial, and Golgi dynamics \n \nBecause cytoplasmic streaming stops soon after NRC4 activation but before cell collapse, \nwe asked whether other organelles show a comparable arrest. Since the formation of NRC4 \nresistosome cluster puncta is the most distinctive feature during the process, we used it as a \nreference point (time 0) to assess the timing of various subcellular events. We first examined \nARA7-labeled endosomes and found that the movement of endosomes slowed markedly  \naround 3 min before puncta became visible and had stopped entirely by that time, roughly \nmatching the onset and peak of the transient calcium influx (Fig. 4A–B, and Fig. S6A; Movie \nS7) (Scheuring et al. 2011). Likewise, GmMan49-marked Golgi stacks lost motility ~2–3 min \nbefore puncta formation (Fig. S7A,B; Movie S8) (Nelson et al. 2007).  \n \nTime-lapse imaging using the mitochondrial matrix marker proATPsyn-RFP showed that \nmitochondria were highly motile and alternated between spherical and rod-like shapes (Lee \net al. 2012). Roughly 2–3 min before NRC4 puncta appeared, mitochondrial movement \nstopped and the mitochondrial population became exclusively spherical, as reflected by a \nmarked increase in roundness (Fig. 4C–E, and Fig. S6B; Movie S9). To examine plastid \ndynamics, we expressed the plastid stroma marker proRubisco-RFP and used chlorophyll \nautofluorescence to visualize thylakoids (Nelson et al. 2007). Similar to other organelles, \nplastids moved dynamically before NRC4 activation. Notably, around two minutes before the \nappearance of NRC4 puncta, plastids began to swell, followed by a rapid burst that released \nthe stromal contents into the cytosol while the thylakoid signal remained intact (Fig. 4F-H, \nS6C; Movie S10). Overall, these observations indicate that NRC4 resistosome activation \ntriggers dramatic changes in organelles, including cessation of movement, altered \nmorphology, and loss of integrity. These events likely begin near the time of calcium influx, \nas inferred from Fig. 3, and continue during the appearance of NRC4 puncta, preceding cell \ncollapse. \n9 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted July 1, 2025. ; https://doi.org/10.1101/2025.06.27.662017doi: bioRxiv preprint \n\n \n \nFigure 4. NRC4 resistosome activation halts the movement of vesicles, mitochondria, \nand Golgi bodies, and causes plastid disruption. \nA. Time-lapse confocal images showing the dynamics of mOrange2-ARA7–labeled endosomes (top) and \nNRC4-GFP (bottom) during hypersensitive cell death. N. benthamiana  leaves were co-infiltrated with \n35S::Rpi-blb2, CBS4::AVRblb2, 35S::CUP2-p65, 35S::mOrange2-ARA7, and 35S::NRC4-GFP. Two days post \ninfiltration, copper was applied to induce effector expression. Images were acquired at 3-minute intervals. Time 0 \ncorresponds to the onset of visible NRC4 puncta. Scale bar = 2.5\n \nμm. (See Movie S7.)  \nB. Quantification of endosome trafficking velocity at different time windows relative to NRC4 puncta formation. \nVelocities were calculated across 50-frame windows. Each dot represents an individual endosome; violin plots \nshow distribution. Different letters denote statistically significant differences (one-way ANOVA with Tukey’s HSD, \np < 0.05).  \nC. Kymographs showing the temporal dynamics of mitochondria (proATPsyn-RFP, top) and NRC4-GFP (bottom) \nduring hypersensitive cell death.  \nD. Time-lapse images of mitochondrial morphology during NRC4 activation. Top: proATPsyn-RFP signal from the \ncell periphery; bottom: NRC4-GFP signal from the same region. Time 0 marks NRC4 puncta formation. White \narrowhead indicates a representative NRC4 punctum. Scale bar = 2\n \nμm. (See Movie S9.)  \nE. Time-course quantification of mitochondrial roundness. Each point represents one mitochondrion. Color scale \nindicates data density (purple: low; yellow: high). The dashed line shows a generalized linear model (GLM) fit; \nshading represents the 95% confidence interval of the predicted mean. (n = 40-63 mitochondria.) \nF. Kymographs showing the temporal dynamics of plastids (autofluorescence, top), proRubisco-RFP–labeled \nstroma (middle), and NRC4-GFP (bottom) during hypersensitive cell death.  \n10 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted July 1, 2025. ; https://doi.org/10.1101/2025.06.27.662017doi: bioRxiv preprint \n\n \nG. Time-lapse images showing changes of plastid morphology and integrity during NRC4 activation. Top: merged \nimage of plastid autofluorescence and NRC4-GFP; bottom: proRubisco-RFP signal. Time 0 marks visible NRC4 \npuncta formation. White arrowhead indicates an NRC4 punctum. Scale bar = 5\n \nμm. (See Movie S10.)  \nH. Quantification of plastid roundness and cytosolic proRubisco-RFP signal intensity over time. Only plastids that \nremained visible throughout the time series were analyzed. Yellow dots: individual plastid measurements; black \nline: GLM trend of predicted mean with 95% confidence interval (gray shading). Magenta line (right Y-axis): mean \ncytosolic proRubisco-RFP signal intensity. (n = 6-10 plastids.)  \n \n \nNRC4 resistosome activation depolymerizes the actin cytoskeleton \n \nWe next tracked cytoskeletal dynamics during NRC4-mediated cell death. Live imaging with \nthe LifeAct–mOrange2 marker revealed highly motile actin filaments under resting \nconditions. Their movement stopped around 3–4 min before NRC4 puncta became visible, \nand fluorescence then grew progressively weaker and more diffuse (Fig. 5A, S8A, and \nS9A–B; Movie S11). Quantitative anisotropy analysis of LifeAct-mOrange2-labeled actin \nfilaments revealed a sharp drop in filament alignment at 0.5-1.0 minutes before the \nappearance of NRC4 puncta, indicative of rapid actin depolymerization (Fig. 5B). Similarly, \nwe analyzed microtubule dynamics during NRC4-mediated cell death using the \nmOrange2-MAP4 marker. Anisotropy analysis revealed that microtubule filaments \nprogressively lost alignment during cell death, with disassembly initiating 0.5-1.0 minutes \nbefore the appearance of NRC4 puncta (Fig. 5C-D, S8B, and S9C–D; Movie S12). \nIntriguingly, NRC4 filaments arose in regions that  previously had been occupied by \nmicrotubules, although any functional relationship between microtubules and NRC4 \nresistosomes remains so far unclear (Movie S13). Because the cytoskeleton underpins the \nmotility and positioning of multiple organelles, its disassembly could explain the abrupt arrest \nof  organelle movement and may help remodel cellular architecture for the rapid execution of \nthe immune response  (Perico and Sparkes 2018). \n \n \n \n11 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted July 1, 2025. ; https://doi.org/10.1101/2025.06.27.662017doi: bioRxiv preprint \n\n \n \nFigure 5. Cytoskeletons undergo depolymerization upon NRC4 resistosome \nactivation.  \nA. Time-lapse images showing actin dynamics labeled by LifeAct-mOrange2 (top) and NRC4-GFP (bottom) \nduring hypersensitive cell death. N. benthamiana  leaves were infiltrated with 35S::Rpi-blb2, CBS4::AVRblb2, \n35S::CUP2-p65, 35S::LifeAct-mOrange2, and 35S::NRC4-GFP. Copper solution was applied 2 days post \ninfiltration, followed by confocal imaging. Images were subsetted from the time series at 1.5-minute intervals. \nTime 0 corresponds to the first appearance of NRC4 puncta. Scale bar = 10\n \nμm. (See Movie S11.)  \nB. Quantification of actin filament anisotropy over time. Filament coverage was measured within the cortical \nregion and expressed as anisotropy values indicating directional alignment. A rapid decline in actin anisotropy \nwas observed upon NRC4 puncta formation.  \nC. Time-lapse images showing microtubule dynamics labeled by mOrange2-MAP4-MBD (top) and NRC4-GFP \n(bottom) during hypersensitive cell death. Plants were co-infiltrated with 35S::Rpi-blb2 , CBS4::AVRblb2, \n35S::CUP2-p65, 35S::mOrange2-MAP4-MBD, and 35S::NRC4-GFP. Copper was applied at 2 dpi, and images \nwere subsetted from the time series at 1.5-minute intervals. Time 0 corresponds to visible NRC4 puncta \nformation. Scale bar = 10\n \nμm. (See Movie S12.)  \nD. Quantification of microtubule anisotropy over time. Filamentous regions were segmented within the cortical \nplane, and anisotropy was computed as a measure of microtubule organization. Microtubule alignment declined \nsharply following NRC4 puncta formation.  \n \nActivation of NRC4 resistosome leads to loss of ER, plasma membrane, and tonoplast \nintegrity \nSince an intact cytoskeleton helps maintain endoplasmic-reticulum (ER) morphology, we \ntracked ER dynamics during NRC4-mediated cell death using the ER marker mCherry-HDEL \n(Wang and Hussey 2015; Pain et al. 2023). Initially, the ER exhibited dynamic movement \nduring the early phases of imaging after copper treatment. About four minutes before NRC4 \npuncta appeared, this movement stopped, although the reticulate network remained intact. \nStrikingly, once NRC4 puncta became visible, the ER network fragmented into vesicle-like \n12 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted July 1, 2025. ; https://doi.org/10.1101/2025.06.27.662017doi: bioRxiv preprint \n\n \nstructures (Fig. 6A and S10A; Movie S14). Persistence mapping confirmed that the network \nwas stable during the ten minutes preceding puncta formation (Fig. S11A; Movie S14), but \ndisintegrated during the following ten minutes(Fig. S11AB; Movie S14). Quantitative analysis \nshowed that ER mesh counts fell from around 15-20 per 400 μm 2 to zero at the moment \nNRC4 puncta appeared, confirming the swift transition from a tubular network to dispersed \nvesicles-like forms (Fig. 6B). \n \nTo visualise plasma-membrane behaviour during NRC4-mediated cell death, we used \nSlSOBIR1-mCherry and the lipophilic dye FM4-64 (Peng et al. 2015). As soon as \nNRC4-GFP puncta became visible, SlSOBIR1-mCherry reorganised into discrete, \nprotein-rich microdomains; around 6–7 min later these microdomains resolved into a \nreticulate pattern, implying a profound change in membrane organization (Fig. 6C; Movie \nS15). In contrast, the bulk lipid distribution labelled by FM4-64 showed little change over the \nsame period (Fig. 6D; Movie S16). Quantification confirmed that SlSOBIR1 coverage at the \nplasma membrane declined steadily after NRC4 activation, whereas FM4-64 coverage \nremained stable until the late phase when overt cell collapse became apparent (Fig. 6E; \nMovie S16). These findings indicate that NRC4 resistosome activation alters the protein \nlandscape and biophysical properties of the plasma membrane well before the lipid bilayer \nitself is compromised.    \n \nTo investigate tonoplast dynamics, we used AtTPK1 as a marker for the tonoplast membrane \n(Kasaras and Kunze 2017). The AtTPK1-mOrange2 signal remained uniform until NRC4 \npuncta appeared, after which fluorescence intensity and membrane coverage dropped \nsharply, signalling rapid loss of tonoplast integrity (Fig. 6F, S10B–C, and S11D; Movies S17, \nS18). Roughly two minutes before puncta formation, transvacuolar strands, whose stability \ntypically depends on the cytoskeleton, began to bulge into bleb-like swellings; these strands \nthen fragmented into vesicle-like bodies and gradually vanished (Fig. 6G; Movie S19). These \nobservations suggest that NRC4-mediated hypersensitive cell death involves coordinated \nremodeling and destabilization of multiple membrane systems, including the tonoplast. \n \n \n13 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted July 1, 2025. ; https://doi.org/10.1101/2025.06.27.662017doi: bioRxiv preprint \n\n \n \nFigure 6. NRC4 resistosome activation disrupts ER, plasma membrane, and tonoplast \nintegrity. \nA. Time-lapse images showing fragmentation of the endoplasmic reticulum (ER) during hypersensitive cell death. \nN. benthamiana  leaves were co-infiltrated with 35S::Rpi-blb2, CBS4::AVRblb2, \n35S::CUP2-p65,35S::mCherry-HDEL, and 35S::NRC4-GFP. Two days post infiltration, copper solution was \napplied, followed by confocal imaging. Top: mCherry-HDEL; bottom: NRC4-GFP. Images were extracted from the \ntime series at 4.2-minute intervals. White arrowhead marks an NRC4 punctum at the cell periphery. Scale bar = \n10\n \nμm. (See Movie S14.)  \nB. Quantification of ER mesh structures over time. Meshes were manually segmented in cortical regions from \nthree cells, where a mesh was defined as an area enclosed by ER tubules. Time 0 corresponds to the \nappearance of NRC4 puncta at the cell periphery.  \nC. and D. Time-lapse images showing changes in PM protein and lipid distribution during hypersensitive cell \ndeath. Leaves were co-infiltrated as in (A), and additionally with SlSOBIR1-mCherry for PM protein labeling (C), \nor stained with FM4-64 for lipid labeling (D). NRC4-GFP was co-expressed in both setups. Images were \nextracted from the time series at 3.4-minute (C) and 1.9-minute (D) intervals. White arrowheads indicate NRC4 \npuncta formation. Scale bars = 10\n \nμm. (See Movies S15, S16.)  \nE. Quantification of plasma membrane coverage by SlSOBIR1-mCherry and FM4-64 over time. Signal coverage \nis expressed as the percentage of membrane area labeled relative to the initial frame. Time 0 marks the onset of \nvisible NRC4 puncta.  \nF. Time-lapse images showing tonoplast dynamics using AtTPK1-mOrange2 as a tonoplast marker. Plants were \nco-infiltrated with 35S::Rpi-blb2, CBS4::AVRblb2, 35S::CUP2-p65, AtTPK1-mOrange2, and 35S::NRC4-GFP. \nRepresentative images were extracted at a 1.9-minute interval. Top: TPK1-mOrange2 signal; bottom: NRC4-GFP \nsignal. White arrowhead indicates an NRC4 punctum. Scale bar = 5\n \nμm. (See Movie S17.)  \n14 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted July 1, 2025. ; https://doi.org/10.1101/2025.06.27.662017doi: bioRxiv preprint \n\n \nG. Maximum intensity projection showing merged signals of NRC4-GFP and AtTPK1-mOrange2 during disruption \nof trans-vacuolar strands. Time 0 marks the point of visible NRC4 puncta formation. White arrowhead indicates \nan NRC4 punctum at the cell edge. Scale bar = 5\n \nμm. (See Movie S19.)  \n \nNRC4 resistosome activation triggers nuclear shrinkage and nucleoplasm release \nbefore cell collapse \n \nNuclear shrinkage is a hallmark of cell death (Mur et al. 2008). To track the dynamics of the \nnucleus during NRC4-mediated cell death, we expressed NRC4-GFP along with BFP-NLS \nand performed PI staining. Time-lapse experiments revealed that nuclear shrinkage started \nat the same time or shortly before the appearance of NRC4 puncta (Fig. 7A-B, S7A-B; Movie \nS20, S21). As contraction continued, BFP-NLS leaked into the cytosol, indicating loss of \nnuclear-envelope integrity  (Fig. 7A; Movie S20). Initially, the released BFP signal was \nrestricted to the cytosol, indicating that the tonoplast was still intact at that stage. The BFP \nsignal then abruptly dispersed into the center of the cell, likely reflecting the rupture of the \ntonoplast membrane, and then quickly diminished thereafter.(Fig. 7A-B, S12, and S13A–C; \nMovie S20, S21). PI entered the nucleus at the moment NRC4 puncta formed and steadily \nincreased in intensity, saturating ~20–25 min later  (Fig. 7A-B, S13C; Movie S20, S21). \nTogether, these data place nuclear collapse and tonoplast rupture downstream of NRC4 \nactivation but upstream of the final cellular collapse. \n \nBecause nucleoplasm leakage implied loss of nuclear envelope integrity, we monitored the \nenvelope directly with the outer-nuclear-membrane marker AtPNET2-mOrange2 (Tang et al. \n2022). Prior to the appearance of NRC4 puncta, AtPNET2-mOrange2 sharply outlined the \nnuclear rim (Fig. 7C). Approximately 4–5 minutes after NRC4 puncta formation, the \nAtPNET2-mOrange2 signal began to diffuse into the cytosol, indicating nuclear envelope \nbreakdown (Fig. 7C; Movie S22). In the final collapse phase—typically 20–30 min after \npuncta emergence, though the timing varied across individual cells—the tonoplast and \nplasma membrane, detached from the cell wall and retracted toward the cell center (Fig. 7D, \nand Fig.S14; Movie S23).  15 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted July 1, 2025. ; https://doi.org/10.1101/2025.06.27.662017doi: bioRxiv preprint \n\n \n \nFigure 7. NRC4 resistosome activation induces nuclear condensation and \nnucleoplasm release, followed by cell collapse. \nA. Dynamic changes in nuclear integrity, morphology and NRC4 localization during hypersensitive cell death. N. \nbenthamiana leaves were agro-infiltrated with 35S::Rpi-blb2, CBS4::AVRblb2, 35S::CUP2-p65 , 35S::BFP-NLS \nand 35S::NRC4-GFP. At 2 days post-agroinfiltration, copper solution and propidium iodide were applied, followed \nby confocal imaging. The montage was generated from time-lapse images with 4.6-minute intervals. Top and \nmiddle panels show the signals of PI staining and BFP-NLS fluorescence, respectively; the bottom panel shows \nNRC4-GFP fluorescence signal cropped from the cell edge. Time zero was set to be relative to NRC4 puncta \nformation, with white arrowheads indicating a visible NRC4 punctum. Scale bar = 10\n \nμm. (Movie S20) \nB. Temporal analysis of subcellular events occurring between NRC4 puncta formation and saturated PI entry. \nEach dot represents the time point for a subcellular event occurring in a cell. Time points for nuclear \ncondensation, nuclear envelope (NE) breakdown, and tonoplast breakdown, were determined from the \ntime-lapse series of n = 23 cells. For PI entry, n = 16 cells were analyzed, with 3 outliers excluded from the plot. \nThe x-axis shows the time relative to NRC4 puncta formation (set as time zero).  \nC. Time-lapse montage showing the dynamics of the nuclear envelope label with AtPNET2 during cell death. N. \nbenthamiana leaves were agro-infiltrated with 35S::Rpi-blb2, CBS4::AVRblb2, 35S::CUP2-p65, \n35S::AtPNET2-mOrange2 and 35S::NRC4-GFP. At 2 days post-agroinfiltration, copper solution was applied, \nfollowed by confocal imaging. The montage was generated from time-lapse 3D image stacks captured at \n4.8-minute intervals, showing merged signals from three fluorescence channels. Scale bar = 10\n \nμm. (Movie S22) \nD. Time-lapse montage showing tonoplast dynamics during hypersensitive cell death. Leaves were \nagro-infiltrated with 35S::Rpi-blb2, CBS4::AVRblb2, 35S::CUP2-p65, 35S::TPK1-mOrange2 and \n35S::NRC4-GFP. At 2 days post-agroinfiltration, copper solution was applied, followed by confocal imaging. The \nmontage was generated from time-lapse images with 6-minute intervals. Scale bar = 20\n \nμm. (Movie S23) \n \n16 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted July 1, 2025. ; https://doi.org/10.1101/2025.06.27.662017doi: bioRxiv preprint \n\n \nDiscussion \n \nUsing a copper-inducible expression system (Chiang et al. 2024), we conducted \nhigh-resolution time-lapse imaging to capture the sequence of subcellular events that \nunderlie resistosome-mediated hypersensitive cell death, a process that has been very \nchallenging to visualize in the past (Fig. S15). Our analyses revealed that NRC4 resistosome \nactivation triggers a rapid influx of calcium into the cell, coinciding with NRC4 enrichment at \nthe plasma membrane. This spike initiates a cascade of profound cellular changes: organelle \nmotility ceases, morphology is disrupted, and membrane integrity deteriorates. Both actin \nfilaments and microtubules depolymerise rapidly, and membrane systems remodel on a \nbroad scale: plasma-membrane proteins redistribute, the ER and transvacuolar structures \nfragment into vesicle-like bodies, and the tonoplast loses integrity. Nuclear events begin  \nwith shrinkage, followed by nucleoplasm leakage and rupture of the nuclear envelope. These \nprocesses culminate in catastrophic tonoplast failure and complete cell collapse, marked by \nthe plasma membrane peeling away from the cell wall.  \n \nSince calcium influx can trigger actin and microtubule depolymerization (Cai et al. 2015; \nMadina et al. 2019), we propose that cytoskeleton breakdown is a coordinated component of \nthe death execution program, facilitating the architectural dismantling associated with \nimmune-mediated cell collapse. However, additional, as yet unidentified, signaling cascades \nare likely to contribute to the physiological transitions and final stages of cell death. Together, \nour findings provide a spatiotemporally resolved framework for understanding the \ncomprehensive subcellular reorganization driven by resistosome activation, shedding light \non how immune receptors orchestrate hypersensitive cell death. \n \nA key next step is to determine how the calcium influx triggered by resistosome activation is \ndecoded within the cell. It remains unclear whether calcium alone is sufficient to orchestrate \nthe diverse subcellular processes observed during resistosome-mediated cell death, or \nwhether additional secondary messengers are required. Furthermore, it is not known \nwhether certain organelles play more central roles in mediating cell death, or whether the \nobserved changes are merely downstream consequences of the death process. Addressing \nthese questions will be crucial for understanding how immune signals are integrated at the \ncellular level to coordinate cell death and defense.  \n \nRecent transcriptomic studies have begun to identify genetic components associated with \nimmune-triggered cell death (Salguero-Linares et al. 2022; Burke et al. 2023). Although \nbased on  different experimental systems, it would be valuable to investigate whether these \n17 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted July 1, 2025. ; https://doi.org/10.1101/2025.06.27.662017doi: bioRxiv preprint \n\n \ncandidate genes influence the organelle dynamics described here. Additionally, enzymes \nsuch as metacaspases and autophagy-related proteins, both previously implicated in plant \nhypersensitive cell death  (Hofius et al. 2009, 2017; Coll et al. 2014), may act at specific \nstages such as cytoskeletal disassembly or membrane rupture, or may operate more broadly \nacross death cascade.  \n \nThis study focuses on the NRC4 resistosome, a hexameric complex representing one of \nseveral structural classes of resistosomes in plants  (Huang et al. 2025a). Other forms, such \nas pentameric resistosomes, likely function in a similar manner, while TNL-derived tetrameric \nresistosomes act as NADase enzymes to induce cell death through distinct mechanisms \n(Huang et al. 2025a). Whether these structurally diverse complexes, as well as downstream \npartners such as NRG1 and ADR1, converge on common or distinct cell death programs \nremains an open question (Yu et al. 2024; Huang et al. 2025b). Similarly, the differences \nbetween the cell death processes observed here to the vacuolar-type cell death mediated by \nproteins like PML5, or membrane-disrupting cell death triggered by MLKL-like proteins, \nrequire further investigation (Shen et al. 2024; Sunil et al. 2024). Beyond NLRs, cell surface \nimmune receptors such as Cf-4 and ELR also induce cell death upon recognition of the \ncorresponding ligands (Thomas et al. 1997; Du et al. 2015). Although NRCs, in particular \nNRC3, have been shown to contribute to cell death triggered by cell-surface immune \nreceptors, the extent of overlap in downstream signaling, including calcium dynamics and \norganelle remodeling, between PRR- and NLR-mediated pathways remains unclear \n(Kourelis et al. 2022).  \n \nFinally, programmed cell death during plant development shares morphological features with \nimmune-related cell death, but the mechanistic parallels remain underexplored (Wang et al. \n2023). Beyond the plant kingdom, animal cell death programs such as apoptosis and \npyroptosis exhibit distinct but occasionally analogous features (Coll et al. 2011; Maekawa et \nal. 2023). How much convergence exists in subcellular dynamics between plant and animal \nimmune cell death programs remains to be addressed (Mur et al. 2008). Our \nspatiotemporally resolved analyses of calcium signaling and organelle dynamics offer a \nvaluable reference framework for further investigations into  the mechanisms of plant cell \ndeath across immune, evolutionary, and developmental contexts, and also inform broader \nunderstanding of cell death processes beyond the plant kingdom. \n \n \nAcknowledgments:  \n \n18 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted July 1, 2025. ; https://doi.org/10.1101/2025.06.27.662017doi: bioRxiv preprint \n\n \nWe thank Mark Youles (SynBio, The Sainsbury Laboratory, UK) for sharing plasmids for \nmolecular cloning, Dr. Tien-Shin Yu (Institute of Plant and Microbial Biology, Academia \nSinica), and Dr. Yen-Ping Hsueh (Institute of Molecular Biology, Academia Sinica) for sharing \nmarkers for cell biology studies. We thank Mei-Jane Fang and Ji-Ying Huang in the Cell \nBiology Core Lab (Institute of Plant and Microbial Biology, Academia Sinica, Taiwan) and \nShu-Chen Shen in the Advanced Optical Microscope Core Facility (Agricultural \nBiotechnology Research Center, Academia Sinica, Taiwan) for help with confocal imaging. \nWe thank Lin-Yun Kuang in the Transgenic Plant Laboratory (Academia Sinica) for \ngenerating transgenic N. benthamiana lines.  \n \n \nFunding:  \n \nNational Science and Technology Council (NSTC) grant NSTC-113-2628-B-001-004 (CHW) \nInstitute of Plant and Microbial Biology, Academia Sinica, intramural fund (CHW) \nBiotechnology and Biological Sciences Research Council (BBSRC) BB/X016382/1 (TOB) \nNSTC-Royal Society bilateral exchange grant NSTC-113-2927-I-001-514, \nIEC\\NSFC\\233289 (CHW, TOB) \n \n \nAuthor contributions:  \nConceptualization: YFC, KYL, CYH, CHW \nMethodology: YFC, KYL, CYH, HYW \nInvestigation: YFC, KYL, CYH, LYH, WCS, BJC, CWC \nVisualization: YFC, KYL \nFunding acquisition: CHW, TOB \nProject administration: CHW \nSupervision: CHW, TOB \nWriting – original draft: YFC, KYL, LYH, BJC, ELHY, CHW \nWriting – review & editing: YFC, KYL, LYH, CYH, WCS, BJC, ELHY, TOB, CHW \n \nCompeting interests: TOB receives funding from the industry on NLR biology, and is a \nco-founder of Resurrect Bio Ltd. The remaining authors have no conflicts of interest to \ndeclare. \n \nData and materials availability: All data are available in the main text or the supplementary \nmaterials. \n19 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted July 1, 2025. ; https://doi.org/10.1101/2025.06.27.662017doi: bioRxiv preprint \n\n \n \nReferences \nAhn H, Lin X, Olave‐Achury AC, Derevnina L, Contreras MP, Kourelis J, Wu C, \nKamoun S, and Jones JDG. Effector‐dependent activation and oligomerization of \nplant NRC class helper NLRs by sensor NLR immune receptors Rpi‐amr3 and \nRpi‐amr1. The EMBO Journal. 2023:42(5):e111484. \nhttps://doi.org/10.15252/embj.2022111484 \nAkerboom J, Carreras Calderón N, Tian L, Wabnig S, Prigge M, Tolö J, Gordus A, \nOrger MB, Severi KE, Macklin JJ, et al. Genetically encoded calcium indicators for \nmulti-color neural activity imaging and combination with optogenetics. Front Mol \nNeurosci. 2013:6:2. https://doi.org/10.3389/fnmol.2013.00002 \nBurke R, McCabe A, Sonawane NR, Rathod MH, Whelan CV, McCabe PF, and Kacprzyk \nJ. Arabidopsis cell suspension culture and RNA sequencing reveal regulatory \nnetworks underlying plant-programmed cell death. The Plant Journal. \n2023:115(6):1465–1485. https://doi.org/10.1111/tpj.16407 \nCai G, Parrotta L, and Cresti M. Organelle trafficking, the cytoskeleton, and pollen tube \ngrowth. Journal of Integrative Plant Biology. 2015:57(1):63–78. \nhttps://doi.org/10.1111/jipb.12289 \nChai J, Song W, and Parker JE. New Biochemical Principles for NLR Immunity in Plants. \nMPMI. 2023:36(8):468–475. https://doi.org/10.1094/MPMI-05-23-0073-HH \nChen T-W, Wardill TJ, Sun Y, Pulver SR, Renninger SL, Baohan A, Schreiter ER, Kerr \nRA, Orger MB, Jayaraman V, et al. Ultrasensitive fluorescent proteins for imaging \nneuronal activity. Nature. 2013:499(7458):295–300. \nhttps://doi.org/10.1038/nature12354 \nChiang B-J, Lin K-Y, Chen Y-F, Huang C-Y, Goh F-J, Huang L-T, Chen L-H, and Wu C-H. \nDevelopment of a tightly regulated copper-inducible transient gene expression \nsystem in Nicotiana benthamiana incorporating a suicide exon and Cre recombinase. \nNew Phytol. 2024:244(1):318–331. https://doi.org/10.1111/nph.20021 \nColl NS, Epple P, and Dangl JL. Programmed cell death in the plant immune system. Cell \nDeath Differ. 2011:18(8):1247–1256. https://doi.org/10.1038/cdd.2011.37 \nColl NS, Smidler A, Puigvert M, Popa C, Valls M, and Dangl JL. The plant metacaspase \nAtMC1 in pathogen-triggered programmed cell death and aging: functional linkage \nwith autophagy. Cell Death Differ. 2014:21(9):1399–1408. \nhttps://doi.org/10.1038/cdd.2014.50 \nContreras MP, Pai H, Tumtas Y, Duggan C, Yuen ELH, Cruces AV, Kourelis J, Ahn H, \nLee K, Wu C, et al. Sensor NLR immune proteins activate oligomerization of their \nNRC helpers in response to plant pathogens. The EMBO Journal. \n2023:42(5):e111519. https://doi.org/10.15252/embj.2022111519 \nDu J, Verzaux E, Chaparro-Garcia A, Bijsterbosch G, Keizer LCP, Zhou J, Liebrand \nTWH, Xie C, Govers F, Robatzek S, et al. Elicitin recognition confers enhanced \nresistance to Phytophthora infestans in potato. Nat Plants. 2015:1(4):15034. \nhttps://doi.org/10.1038/nplants.2015.34 \nDuggan C, Moratto E, Savage Z, Hamilton E, Adachi H, Wu C-H, Leary AY, Tumtas Y, \n20 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted July 1, 2025. ; https://doi.org/10.1101/2025.06.27.662017doi: bioRxiv preprint \n\n \nRothery SM, Maqbool A, et al. Dynamic localization of a helper NLR at the \nplant–pathogen interface underpins pathogen recognition. Proceedings of the \nNational Academy of Sciences. 2021:118(34):e2104997118. \nhttps://doi.org/10.1073/pnas.2104997118 \nDuxbury Z, Wu C, and Ding P. A Comparative Overview of the Intracellular Guardians of \nPlants and Animals: NLRs in Innate Immunity and Beyond. Annual Review of Plant \nBiology. 2021:72(Volume 72, 2021):155–184. \nhttps://doi.org/10.1146/annurev-arplant-080620-104948 \nHofius D, Li L, Hafrén A, and Coll NS. Autophagy as an emerging arena for \nplant–pathogen interactions. Current Opinion in Plant Biology. 2017:38:117–123. \nhttps://doi.org/10.1016/j.pbi.2017.04.017 \nHofius D, Schultz-Larsen T, Joensen J, Tsitsigiannis DI, Petersen NHT, Mattsson O, \nJørgensen LB, Jones JDG, Mundy J, and Petersen M. Autophagic components \ncontribute to hypersensitive cell death in Arabidopsis. Cell. 2009:137(4):773–783. \nhttps://doi.org/10.1016/j.cell.2009.02.036 \nHuang S, Li E, Jia F, Han Z, and Chai J. Assembly and functional mechanisms of plant \nNLR resistosomes. Current Opinion in Structural Biology. 2025a:90:102977. \nhttps://doi.org/10.1016/j.sbi.2024.102977 \nHuang S, Wang J, Song R, Jia A, Xiao Y, Sun Y, Wang L, Mahr D, Wu Z, Han Z, et al. \nBalanced plant helper NLR activation by a modified host protein complex. Nature. \n2025b:639(8054):447–455. https://doi.org/10.1038/s41586-024-08521-7 \nKasaras A and Kunze R. Dual-targeting of Arabidopsis DMP1 isoforms to the tonoplast and \nthe plasma membrane. PLOS ONE. 2017:12(4):e0174062. \nhttps://doi.org/10.1371/journal.pone.0174062 \nKourelis J, Contreras MP, Harant A, Pai H, Ludke D, Adachi H, Derevnina L, Wu CH, \nand Kamoun S. The helper NLR immune protein NRC3 mediates the hypersensitive \ncell death caused by the cell-surface receptor Cf-4. PLoS Genet. \n2022:18(9):e1010414. https://doi.org/10.1371/journal.pgen.1010414 \nLee S, Lee DW, Yoo Y-J, Duncan O, Oh YJ, Lee YJ, Lee G, Whelan J, and Hwang I. \nMitochondrial Targeting of the Arabidopsis F1-ATPase γ-Subunit via Multiple \nCompensatory and Synergistic Presequence Motifs. The Plant Cell. \n2012:24(12):5037–5057. https://doi.org/10.1105/tpc.112.105361 \nLiu F, Yang Z, Wang C, You Z, Martin R, Qiao W, Huang J, Jacob P, Dangl JL, Carette \nJE, et al. Activation of the helper NRC4 immune receptor forms a hexameric \nresistosome. Cell. 2024:187(18):4877-4889.e15. \nhttps://doi.org/10.1016/j.cell.2024.07.013 \nMa S, An C, Lawson AW, Cao Y, Sun Y, Tan EYJ, Pan J, Jirschitzka J, Kümmel F, Mukhi \nN, et al. Oligomerization-mediated autoinhibition and cofactor binding of a plant NLR. \nNature. 2024:632(8026):869–876. https://doi.org/10.1038/s41586-024-07668-7 \nMadhuprakash J, Toghani A, Contreras MP, Posbeyikian A, Richardson J, Kourelis J, \nBozkurt TO, Webster MW, and Kamoun S. A disease resistance protein triggers \noligomerization of its NLR helper into a hexameric resistosome to mediate innate \nimmunity. Sci Adv. 2024:10(45):eadr2594. https://doi.org/10.1126/sciadv.adr2594 \n21 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted July 1, 2025. ; https://doi.org/10.1101/2025.06.27.662017doi: bioRxiv preprint \n\n \nMadina MH, Rahman MS, Zheng H, and Germain H. Vacuolar membrane structures and \ntheir roles in plant–pathogen interactions. Plant Mol Biol. 2019:101(4):343–354. \nhttps://doi.org/10.1007/s11103-019-00921-y \nMaekawa T, Kashkar H, and Coll NS. Dying in self-defence: a comparative overview of \nimmunogenic cell death signalling in animals and plants. Cell Death Differ. \n2023:30(2):258–268. https://doi.org/10.1038/s41418-022-01060-6 \nMur LAJ, Kenton P, Lloyd AJ, Ougham H, and Prats E. The hypersensitive response; the \ncentenary is upon us but how much do we know? J Exp Bot. 2008:59(3):501–520. \nhttps://doi.org/10.1093/jxb/erm239 \nNelson BK, Cai X, and Nebenführ A. A multicolored set of in vivo organelle markers for \nco-localization studies in Arabidopsis and other plants. The Plant Journal. \n2007:51(6):1126–1136. https://doi.org/10.1111/j.1365-313X.2007.03212.x \nPain C, Tolmie F, Wojcik S, Wang P, and Kriechbaumer V. intER-ACTINg: The structure \nand dynamics of ER and actin are interlinked. Journal of Microscopy. \n2023:291(1):105–118. https://doi.org/10.1111/jmi.13139 \nPeng K-C, Wang C-W, Wu C-H, Huang C-T, and Liou R-F. Tomato SOBIR1/EVR \nHomologs Are Involved in Elicitin Perception and Plant Defense Against the \nOomycete Pathogen Phytophthora parasitica. MPMI. 2015:28(8):913–926. \nhttps://doi.org/10.1094/MPMI-12-14-0405-R \nPerico C and Sparkes I. Plant organelle dynamics: cytoskeletal control and membrane \ncontact sites. New Phytologist. 2018:220(2):381–394. \nhttps://doi.org/10.1111/nph.15365 \nSalguero-Linares J, Serrano I, Ruiz-Solani N, Salas-Gómez M, Phukan UJ, González \nVM, Bernardo-Faura M, Valls M, Rengel D, and Coll NS. Robust transcriptional \nindicators of immune cell death revealed by spatiotemporal transcriptome analyses. \nMolecular Plant. 2022:15(6):1059–1075. https://doi.org/10.1016/j.molp.2022.04.010 \nScheuring D, Viotti C, Krüger F, Künzl F, Sturm S, Bubeck J, Hillmer S, Frigerio L, \nRobinson DG, Pimpl P, et al. Multivesicular Bodies Mature from the Trans-Golgi \nNetwork/Early Endosome in Arabidopsis[W]. Plant Cell. 2011:23(9):3463–3481. \nhttps://doi.org/10.1105/tpc.111.086918 \nSelvaraj M, Toghani A, Pai H, Sugihara Y, Kourelis J, Yuen ELH, Ibrahim T, Zhao H, Xie \nR, Maqbool A, et al. Activation of plant immunity through conversion of a helper NLR \nhomodimer into a resistosome. PLOS Biology. 2024:22(10):e3002868. \nhttps://doi.org/10.1371/journal.pbio.3002868 \nShen Q, Hasegawa K, Oelerich N, Prakken A, Tersch LW, Wang J, Reichhardt F, Tersch \nA, Choo JC, Timmers T, et al. Cytoplasmic calcium influx mediated by plant MLKLs \nconfers TNL-triggered immunity. Cell Host & Microbe. 2024:32(4):453-465.e6. \nhttps://doi.org/10.1016/j.chom.2024.02.016 \nSunil S, Beeh S, Stöbbe E, Fischer K, Wilhelm F, Meral A, Paris C, Teasdale L, Jiang Z, \nZhang L, et al. Activation of an atypical plant NLR with an N-terminal deletion \ninitiates cell death at the vacuole. EMBO Rep. 2024:25(10):4358–4386. \nhttps://doi.org/10.1038/s44319-024-00240-4 \nTang Y, Dong Q, Wang T, Gong L, and Gu Y. PNET2 is a component of the plant nuclear \n22 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted July 1, 2025. ; https://doi.org/10.1101/2025.06.27.662017doi: bioRxiv preprint \n\n \nlamina and is required for proper genome organization and activity. Developmental \nCell. 2022:57(1):19-31.e6. https://doi.org/10.1016/j.devcel.2021.11.002 \nThomas CM, Jones DA, Parniske M, Harrison K, Balint-Kurti PJ, Hatzixanthis K, and \nJones JD. Characterization of the tomato Cf-4 gene for resistance to Cladosporium \nfulvum identifies sequences that determine recognitional specificity in Cf-4 and Cf-9. \nPlant Cell. 1997:9(12):2209–2224. https://doi.org/10.1105/tpc.9.12.2209 \nWang H-Y, Yuen ELH, Lee K-T, Goh F-J, Bozkurt TO, and Wu C-H. A hydrophobic core in \nthe coiled-coil domain is essential for NRC resistosome function. \n2025:2025.01.21.634219. https://doi.org/10.1101/2025.01.21.634219 \nWang J, Bollier N, Buono RA, Vahldick H, Lin Z, Feng Q, Hudecek R, Jiang Q, Mylle E, \nVan Damme D, et al. A developmentally controlled cellular decompartmentalization \nprocess executes programmed cell death in the Arabidopsis root cap. Plant Cell. \n2023:36(4):941–962. https://doi.org/10.1093/plcell/koad308 \nWang P and Hussey PJ. Interactions between plant endomembrane systems and the actin \ncytoskeleton. Front Plant Sci. 2015:6. https://doi.org/10.3389/fpls.2015.00422 \nWu C-H, Abd-El-Haliem A, Bozkurt TO, Belhaj K, Terauchi R, Vossen JH, and Kamoun \nS. NLR network mediates immunity to diverse plant pathogens. Proceedings of the \nNational Academy of Sciences. 2017:114(30):8113–8118. \nhttps://doi.org/10.1073/pnas.1702041114 \nYu H, Xu W, Chen S, Wu X, Rao W, Liu X, Xu X, Chen J, Nishimura MT, Zhang Y, et al. \nActivation of a helper NLR by plant and bacterial TIR immune signaling. Science. \n2024:386(6728):1413–1420. https://doi.org/10.1126/science.adr3150 \n \n \n23 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted July 1, 2025. ; https://doi.org/10.1101/2025.06.27.662017doi: bioRxiv preprint \n\n \n \nFigure S1. Effector expression driven by the copper inducible system triggers \nhypersensitive cell death in N. benthamiana. \nA. Schematic representation of the experimental workflow. NRC4, Rpi-blb2, and CUP2-p65 were transiently \nexpressed in N. benthamiana  leaves under the control of 35S promoter together with a copper-inducible \nAVRblb2. At two days post agro-infiltration, copper solution was infiltrated into leaves followed by plate reader \nassay, protein analysis and confocal imaging. \nB. The cell death phenotype of leaves expressing AVRblb2 or CP in the presence or absence of copper. Leaves \nof N. benthamiana were agro-infiltrated with 35S::Rpi-blb2, 35S::CUP2-p65, and either CBS4::AVRblb2 or CP. At \n24 hours post-agroinfiltration, the copper solution was infiltrated into leaves. Autofluorescence emitted from dead \ncells were recorded using UVP ChemStudio at 36 h after copper infiltration. Raw fluorescence intensity was \nnormalized to the maximum detectable value to represent normalized cell death levels. Dots with different colors \nrepresent the results from three independent biological replicates (n = 18). \nC. The cell death phenotype of leaves expressing NRC2, NRC4 or NRC4 mutants in the presence or absence of \ncopper. Leaves of N. benthamiana nrc plants were agro-infiltrated with 35S::Rpi-blb2, CBS4::AVRblb2, \n35S::CUP2-p65, and either 35S::NRC2, 35S::NRC4 or other NRC4 mutants. At 24 hours post-agroinfiltration, the \ncopper solution was infiltrated into leaves Autofluorescence emitted from dead cells were recorded using UVP \nChemStudio at 36 h after copper infiltration. Raw fluorescence intensity was normalized to the maximum \ndetectable value to represent normalized cell death levels. Raw fluorescence intensity was normalized to the \nmaximum detectable value. Dots with different colors represent the results from three independent biological \nreplicates (n = 18). \n24 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted July 1, 2025. ; https://doi.org/10.1101/2025.06.27.662017doi: bioRxiv preprint \n\n \n \n25 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted July 1, 2025. ; https://doi.org/10.1101/2025.06.27.662017doi: bioRxiv preprint \n\n \n \nFigure S2. Regions of interest selected for generating montages in Figure 2. \nA. Representative NRC43A-GFP fluorescence image. The white solid box indicates the region of interest (ROI) \nused to generate the time-lapse montages shown in Fig. 2A. Scale bar = 20 μm. \nB. Overlay of NRC4-GFP fluorescence and bright-field image. The solid and dashed boxes denote the regions \ncorresponding to ROI_1 and ROI_2, which were used for time-lapse analysis presented in Fig. 2B and 2E, \nrespectively. Scale bar = 20 μm. \n \n26 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted July 1, 2025. ; https://doi.org/10.1101/2025.06.27.662017doi: bioRxiv preprint \n\n \n27 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted July 1, 2025. ; https://doi.org/10.1101/2025.06.27.662017doi: bioRxiv preprint \n\n \nFigure S3. NRC4 spatially separated from cytosol upon activation. \nA. Kymographs showing the temporal dynamics of NRC4 3A-GFP (top) and NRC4-GFP (bottom) in N. \nbenthamiana nrc plants after copper-induced resistosome activation. Leaves were co-infiltrated with \n35S::Rpi-blb2, CBS4::AVRblb2, 35S::CUP2-p65, and either NRC43A-GFP or NRC4-GFP. Confocal imaging was \nperformed 2 days post infiltration. White lines indicate regions of interest (ROIs) used for kymograph analysis, \ndisplayed on the right.  \nB. Colocalization analysis of NRC4 3A-GFP and cytosolic mRFP. Left: representative images of cells before (top) \nand after (bottom) NRC4 3A-GFP puncta formation. Insets show scatterplots of pixel intensity correlations \n(NRC43A-GFP: yellow, Y-axis; mRFP: magenta, X-axis) with corresponding Pearson’s correlation coefficients. \nRight: time-course of Pearson’s correlation coefficients between NRC43A-GFP and cyt-mRFP signals. Scale bar = \n10\n \nμm. (See Movie S2.)  \nC. Fluorescence recovery after photobleaching (FRAP) analysis of NRC4 3A-GFP puncta. The arrowhead marks \nthe photobleaching event. Fluorescence intensity was measured over time at 1.6-second intervals. \nD. Time-lapse images from the FRAP assay shown in (C), displayed at 24-second intervals. The arrowhead \nindicates the photobleaching timepoint; the white circle denotes the bleached area. Scale bar = 5\n \nμm.  \nE. Colocalization analysis of NRC4-GFP and cytosolic mRFP. Left: representative images before (top) and after \n(bottom) NRC4 activation. Scatterplots show intensity correlation between NRC4-GFP (yellow) and cyt-mRFP \n(magenta), with Pearson’s coefficients indicated. Right: time-course of Pearson’s correlation coefficients during \nNRC4 activation. Scale bar = 10\n \nμm. (See Movie S4.)  \n \n28 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted July 1, 2025. ; https://doi.org/10.1101/2025.06.27.662017doi: bioRxiv preprint \n\n \n \nFigure S4. Regions of interest selected for generating montages and quantifying \nintensity in Figure 3. \nA. and B. Representative confocal images of GCaMP6 (A) and overlaid RCaMP1h/NRC4-GFP (B), \ncorresponding to Movies S5 and S6, respectively. Left panels: original fields of view with white boxes indicating \nthe cell regions used for montages in Fig. 3A. and 3C. Right panels: zoomed-in views with white lines denoting \nthe regions of interest (ROIs) used for quantifying (Q_ROI) fluorescence intensity and generating kymographs \nshown in Fig. 3B. and 3D. \n \n \n29 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted July 1, 2025. ; https://doi.org/10.1101/2025.06.27.662017doi: bioRxiv preprint \n\n \n \nFigure S5. Co-expression of Rpi-blb2 and AVRblb2 triggers NRC4-dependent calcium \ninflux. \nA. Quantification of cytosolic GCaMP6 intensity from a cell independent of cell shown in Fig. 3A. Leaves of \nGCaMP6-expressing N. benthamiana were infiltrated with 35S::Rpi-blb2, CBS4::AVRblb2, 35S::CUP2-p65. Two \ndays post-agroinfiltration, copper solution was applied to induce effector expression, followed by confocal \nimaging. Time-lapse images were acquired at 6.43-second intervals, starting at 20 min after copper treatment. \nGCaMP6 intensity was normalized to the mean intensity of 50 baseline frames recorded before calcium signal \ninitiation. Bottom: kymograph of the GCaMP6 signal from the time-lapse series.  \nB. Alignment of calcium influx from independent cells. Calcium levels, indicated by GCaMP6 intensity, were \nnormalized to the peak intensity of each time series. Data from independent cells were aligned to time 0, \ncorresponding to the peak intensity, as a reference point. Lines represent individual cells. The black line \nrepresents the cell with the median progression timing of all cells analyzed (n = 10 cells).  \nC. Quantification of GCaMP6 fluorescence intensity in nrc2/3/4 knockout GCaMP6  plants in the presence or \nabsence of NRC4 complementation. Leaves of N. benthamiana  were infiltrated with 35S::Rpi-blb2, \n30 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted July 1, 2025. ; https://doi.org/10.1101/2025.06.27.662017doi: bioRxiv preprint \n\n \nCBS4::AVRblb2, 35S::CUP2-p65 and either 35S::NRC4 , 35S::NRC4L9E or without complementation. Two days \npost-agroinfiltration, the copper solution was infiltrated into leaves followed by confocal microscopy analysis. \nTime-lapse images were acquired at 6.43-second intervals after copper treatment. \nD. Quantification of RCaMP1h and NRC4-GFP fluorescence intensity during hypersensitive cell death from a cell \nindependent of cell shown in Fig. 3C. Leaves of N. benthamiana  were infiltrated with 35S::Rpi-blb2, \nCBS4::AVRblb2, 35S::CUP2-p65, 35S::RCaMP1h and 35S::NRC4-GFP. At 2 days post-agroinfiltration, the \ncopper solution was applied to leaves followed by confocal microscopy analysis. Time-lapse images were \nacquired at 12.86-second intervals. RCaMP1h intensity was normalized to the mean intensity of 50 baseline \nframes recorded before calcium signal initiation. The inset at the top left shows a shorter time window spanning \nthe onset and decline of calcium signals. Open arrowheads mark initial NRC4 membrane enrichment; filled \narrowheads indicate puncta formation. Bottom: kymographs of RCaMP1h and NRC4-GFP signals from the \ntime-lapse series.   \n \n \n31 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted July 1, 2025. ; https://doi.org/10.1101/2025.06.27.662017doi: bioRxiv preprint \n\n \n  \n \nFigure S6. Regions of interest selected for generating montages and kymographs in \nFigure 4. \nA. Representative images showing overlaid  NRC4-GFP/mOrange2-ARA7 corresponding with Movie S7. The \nwhite box indicates the region of interest (ROI) used to generate the time-lapse montages and for \nendosome-tracking shown in Fig. 4A and 4B. Scale bar = 10 μm. \nB. and C. Representative images showing NRC4-GFP overlaid with proATPsyn-RFP (B, corresponding to Movie \nS9) or with plastids and proRubisco-RFP (C, corresponding to Movie S10). Left panels in (B) and (C) show the \noriginal fields of view with white boxes marking the ROIs used for montages in Fig. 4C and 4F. Scale bars = \n20\n \nμm. Right panels in (B) and (C) display zoomed-in single-channel views. In (B), panels (i) and (ii) show \nNRC4-GFP (i) and proATPsyn-RFP (ii), with arrowheads indicating an NRC4-GFP punctum. Scale bar = 2\n \nμm. \nIn (C), panels (i)–(iv) show plastid autofluorescence (i), proRubisco-RFP (ii), NRC4-GFP (iii), and the merged \nimage (iv). The white line denotes the ROI used for generating the kymograph (K_ROI) shown in Figure 4G. \nScale bar = 5\n \nμm. \n \n \n \n32 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted July 1, 2025. ; https://doi.org/10.1101/2025.06.27.662017doi: bioRxiv preprint \n\n \n \n \nFigure S7. Cis-Golgi network movement ceases during hypersensitive cell death. \nA. Representative image showing region of interest (ROI) used for kymograph analysis in Fig. S4B. White area \nindicates the ROI selected for the kymograph. (see Movies S8.) \nB. Kymograph analysis of golgi dynamics during hypersensitive cell death. The top, middle and bottom panels \nshow the signals of GmMan49-mCherry, NRC4-GFP and merged channel images, respectively. \nS10.) \n \n \n33 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted July 1, 2025. ; https://doi.org/10.1101/2025.06.27.662017doi: bioRxiv preprint \n\n \n \nFigure S8. Regions of interest selected for generating montages and analyzing \nanisotropy in Figure 5. \nA. and B. Representative images showing LifeAct-mOrange2 (A, left) or mOrange2-MAP4 (B, left), and overlaid \nplastid/NRC4-GFP channels (A and B, right). The white boxes indicate the regions of interest (ROIs) used for \nmontage generation and anisotropy quantification shown in Fig. 5A, 5B (for S8A), and 5C, 5D (for S8B). Scale \nbars = 20 μm. \n \n \n \n34 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted July 1, 2025. ; https://doi.org/10.1101/2025.06.27.662017doi: bioRxiv preprint \n\n \n \nFigure S9. Actin movement stopped, and both cytoskeleton depolymerized during \nhypersensitive cell death. \nA. Merged image displaying the fluorescence signals of LifeAct-mOrange2-labeled actin and NRC4-GFP at the \ncell periphery. The white line marks the ROI used for kymograph analysis of both actin and NRC4 dynamics. \nScale bar = 5\n \nμm. (Movie S11) \nB. Kymographs showing the temporal dynamics of actin and NRC4 during hypersensitive cell death. The top, \nmiddle and bottom panels show the signals of LifeAct-mOrange2, NRC4-GFP, and composite image, \nrespectively. The asterisk denotes the cessation of actin movement. \nC. Merged image displaying the fluorescence signals of mOrange2-MAP4-MBD-labeled microtubule and \nNRC4-GFP at the cell periphery. The white line marks the region of interest used for kymograph analysis of both \nmicrotubule and NRC4 dynamics. Scale bar = 5\n \nμm. (Movie S13) \nD. Kymographs showing the temporal dynamics of microtubule and NRC4 dynamics during hypersensitive cell \ndeath. The top, middle and bottom panels show the signals of mOrange2-MAP4-MBD, NRC4-GFP, and \ncomposite image, respectively. \n \n \n35 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted July 1, 2025. ; https://doi.org/10.1101/2025.06.27.662017doi: bioRxiv preprint \n\n \n \nFigure S10. Regions of interest selected for analyzing endoplasmic reticulum (ER) and \ntonoplast. \nA. Representative images showing overlaid  NRC4-GFP/plastids (left) and mCherry-HDEL (right)  corresponding \nwith Movie S14. The white box indicates the region of interest (ROI) used to generate the time-lapse montages in \nFig. 6A. The white solid line indicates the ROI used for generating kymographs.  \nScale bar = 20 μm. \nB. Representative images showing overlaid  NRC4-GFP/TPK1-mOrange2  corresponding with Movie S17.  The \nwhite solid box indicates the ROI used to generate the time-lapse montages in Fig. 6F.  \nScale bar = 20 μm. \nC. Representative  NRC4-GFP, TPK1-mOrange2 and merged channel images. The white circle indicates the ROI \nused for quantifying coverage of TPK1-mOrange2  in Fig. S11D. Scale bar =  20 μm. \n \n \n \n \n \n \n \n36 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted July 1, 2025. ; https://doi.org/10.1101/2025.06.27.662017doi: bioRxiv preprint \n\n \n \nFigure S11. NRC4 resistosome activation affects endoplasmic reticulum (ER) tubule \nstability and tonoplast dynamic.  \nA. Persistence mapping of ER structure before (left) and after (right) the formation of NRC4 puncta. A 10 minutes \ntime-series of ER dynamics during hypersensitive cell death was segmented and processed to generate a \nsubstack of non-moving signals across time. Persistence maps were generated by a Z-projection of this \nnon-moving substack, where the pixel value indicates signal persistence over time. Scale bar = 20\n \nμm. \nB. Representative image showed the fragmented ER with NRC4 puncta. The white line marks the ROI used for \nkymograph analysis. Scale bar = 10 μm. (Movie S14) \nC. Kymograph analysis of ER dynamics during hypersensitive cell death. The top, middle and bottom panels \nshow the signals of mCherry-HDEL, NRC4-GFP with plastid and merged channel images, respectively. \nD. Representative images of tonoplast membrane dynamics during hypersensitive cell death. The representative \nimages were subset from the time series imaging with 5.1-minute intervals. Left panels: z-projection of \nNRC4-GFP, original and contrast-enhancing TPK1-mOrange2. The white arrowhead indicates an NRC4 punctum \nform on the cell periphery. Right panels: fluorescence coverage analysis of TPK1-mOrange2. Coverage (%) \nrepresents the area of TPK1-mOrange2, divided by the total area of ROI. Time zero is set relative to the onset of \nNRC4 puncta formation. Scale bar = 10 μm. (Movie S18) \n \n \n37 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted July 1, 2025. ; https://doi.org/10.1101/2025.06.27.662017doi: bioRxiv preprint \n\n \n \nFigure S12. Nucleus and tonoplast disintegrate during hypersensitive cell death. \nRepresentative confocal images showing sequential events in a single cell: NRC4 puncta formation, BFP-NLS \nleakage, and tonoplast breakdown at 0.00, 15.85, and 25.29 minutes, respectively. Timepoints correspond to \nMovie S21. White circles indicate the nucleus area (N) in each panel. Top: propidium iodide; middle: BFP-NLS; \nbottom: NRC4-GFP. White arrowhead indicates an NRC4-GFP punctum. Scale bar = 20 μm.  \n38 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted July 1, 2025. ; https://doi.org/10.1101/2025.06.27.662017doi: bioRxiv preprint \n\n \n \nFigure S13. Nuclear shrinkage was followed by increased permeability to propidium \niodide during cell death progression. \nA. Region of interests (ROIs) selected for quantification. White squares denote the areas selected for NRC4-GFP \nand the nuclear region used in Fig. 7A. The dashed line marks the nuclear region analyzed in Fig. S13B and \nS13C. The dash-dot line outlines the cytosolic region with BFP signal used in Fig. S13C. Scale bar = 10\n \nμm. (see \nMovie S20) \nB. and C. Quantification of temporal changes in nuclear area (B) and fluorescence intensity of BFP-NLS and PI \nstaining (C) during hypersensitive cell death. Nuclear region was selected for measuring PI (solid pink) and \nnucBFP (solid cyan) intensities, while the cytosol region was used for cytBFP (dash cyan). Asterisk in C marks \nthe timing of a local tonoplast breakdown leading to decrease of cytosolic BFP intensity.  Data corresponds to the \ncell shown in Figure 7A and Movie S20.  \n \n \n39 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted July 1, 2025. ; https://doi.org/10.1101/2025.06.27.662017doi: bioRxiv preprint \n\n \n \nFigure S14. Plasma membrane shrank during hypersensitive cell death. \nA. and B. Representative confocal images showing  the SlSOBI1 (A) or RFP-Remorin1.3 (B) labeled plasma \nmembrane, NRC4-GFP (middle) and bright field (right). Solid and dotted lines mark the cell boundaries of the \ncentral and neighboring cells, respectively. Arrowheads mark regions where plasma membrane shrinkage is \nobserved. Scale bar = 20 μm.  \n40 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted July 1, 2025. ; https://doi.org/10.1101/2025.06.27.662017doi: bioRxiv preprint \n\n \n \nFigure S15. Dynamics of calcium and organelles in NRC4 resistosome mediated \nhypersensitive cell death.  \nActivation of NRC4 triggers a transient influx of calcium into the cytosol. This calcium spike coincides with a halt \nin endoplasmic reticulum (ER) and actin dynamics, nuclear shrinkage, and swelling of plastids and mitochondria. \nAs cell death progresses, the membrane systems become severely disrupted, with the appearance of plasma \nmembrane (PM) nanodomains and the fragmentation of the ER, nuclear envelope (NE), and plastids. The \ncytoskeleton also undergoes depolymerization. At later stages, the tonoplast disintegrates along with a loss of \nPM integrity. Ultimately, the cell undergoes shrinkage, with both the PM and tonoplast collapsing inward toward \nthe cell center. \n \n \n41 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted July 1, 2025. ; https://doi.org/10.1101/2025.06.27.662017doi: bioRxiv preprint \n\n \nMovie S1. NRC43A-GFP forms puncta upon activation, related to Figure 2A, 2C, S2A and \nS3A. \nMovie S2. NRC4 3A-GFP relocalizes from cytosol to cell periphery upon activation, related to \nFigure S3B. \nMovie S3. NRC4-GFP forms puncta and filamentous structures upon activation, followed by \ncell collapse. Related to Figures 2B, 2D-E, S2B and S3A. \nMovie S4. NRC4-GFP relocalizes from cytosol to cell periphery upon activation. Related to \nFigure S3E. \nMovie S5. A cytosolic calcium surge precedes cell collapse. Related to Figure 3A-B. \nMovie S6. Enrichment of NRC4-GFP at the cell periphery coincides with the cytosolic \ncalcium surge, followed by the formation of NRC4 puncta. Related to Figure 3C-D. \nMovie S7. Endosomes labeled with ARA7-RFP stop trafficking prior to the formation of \nNRC4 puncta. Related to Figures 4A-B, and S6A. \nMovie S8. Cis-Golgi cisternae labeled with GmMan49-RFP stop trafficking prior to NRC4 \npuncta formation. Related to Figures S7A-B. \nMovie S9. Mitochondria labeled with proATPsyn-RFP stop trafficking prior to NRC4 puncta \nformation and transition into spherical, vesicle-like structures. Related to Figures 4C-4E and \nS6B. \nMovie S10. Plastids labeled with proRubisco-RFP stop trafficking prior to NRC4 puncta \nformation, then swell and burst, releasing stromal contents into the cytosol after puncta \nformation. Related to Figures 4F-H, and S6C. \nMovie S11. Actin dynamics labeled with LifeAct-mOrange2 stop prior to NRC4 puncta \nformation, with a marked decrease in signal coinciding with the onset of visible NRC4 \npuncta. Related to Figures 5A-B, S8A, and S9A-B. \nMovie S12. Microtubules labeled with mOrange2-MAP4-MBD undergo depolymerization at \nthe onset of NRC4 puncta formation. Related to Figures 5C-D, S8B, and S9C-D. \nMovie S13. Activated NRC4 forms filamentous structures that align with microtubule \npatterns. Related to Figure S9C-D. \nMovie S14. Endoplasmic reticulum labeled with mCherry-HDEL stops streaming prior to \nNRC4 puncta formation, then fragments into vesicle-like structures upon puncta formation. \nRelated to Figure 6A, S10A, S11B-C. \nMovie S15. Plasma membrane labeled with SlSOBIR1-mCherry redistributes upon NRC4 \npuncta formation, and the signal decreases after cell collapse as the cell periphery moves \nout of the focal plane. Related to Figure 6C. \nMovie S16. Plasma membrane lipids stained with FM4-64 remain stable before and after \nNRC4 puncta formation, with signal decrease after cell collapse as the cell periphery moves \nout of the focal plane. Related to Figure 6D. \n42 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted July 1, 2025. ; https://doi.org/10.1101/2025.06.27.662017doi: bioRxiv preprint \n\n \nMovie S17. Tonoplast labeled by AtTPK1-mOrange2 lost integrity after NRC4 puncta \nformation. Related to Figure 6F, and S10B. \nMovie S18. Reconstructed 3D time-lapse imaging of tonoplast membrane integrity during \nhypersensitive cell death. The tonoplast loses integrity after NRC4 puncta formation. Related \nto Figure S10C, and S11D. \nMovie S19. Reconstructed 3D time-lapse imaging of tonoplast dynamics during \nhypersensitive cell death. Transvacuolar strands undergo blebbing and break down into \nvesicle-like structures during the process. Related to Figure 6G. \nMovie S20. Nucleus labeled with BFP-NLS undergoes movement cessation, nuclear \nshrinkage, and propidium iodide (PI) entry during hypersensitive cell death. Related to \nFigure 7A-B, and S13A-C. \nMovie S21. Representative time-lapse series showing a progression of events during \nhypersensitive cell death: nuclear shrinkage, nuclear envelope (NE) disintegration indicated \nby release of BFP-NLS into the cytosol, tonoplast (TN) disintegration indicated by entry of \nBFP-NLS into the vacuole, and propidium iodide (PI) entry. Related to Figure 7B, and S12. \nMovie S22. Nuclear envelope labeled with AtPNET2 disperses into the cytosol after NRC4 \npuncta formation. Related to Figure 7C. \nMovie S23. Tonoplast labeled with AtTPK1 detaches from the cell boundary at a late stage \nof cell death. Related to Figure 7D. \n \n \n \n43 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted July 1, 2025. ; https://doi.org/10.1101/2025.06.27.662017doi: bioRxiv preprint","source_license":"CC-BY-4.0","license_restricted":false}