Abstract
Nucleotide-binding domain and leucine-rich repeat (NLR) proteins with pathogen sensor activities have evolved to initiate immune signaling by activating helper NLRs. However, the mechanisms underpinning helper NLR activation by sensor NLRs remain poorly understood. Although coiled-coil (CC) type sensor NLRs such as the Potato virus X disease resistance protein Rx have been shown to activate the oligomerization of their downstream helpers NRC2 and NRC4, the domains involved in sensor-helper signaling are not known. Here, we show that the nucleotide binding (NB) domain within the NB-ARC of the Potato virus X disease resistance protein Rx is necessary and sufficient for oligomerization and immune signaling of downstream helper NLRs. In addition, the NB domains of the disease resistance proteins Gpa2 (cyst nematode resistance), Rpi-amr1, Rpi-amr3 (oomycete resistance) and Sw-5b (virus resistance) are also sufficient to activate their respective downstream NRC helpers. Moreover, the NB domain of Rx and its helper NRC2 form a minimal functional unit that can be transferred from solanaceous plants (lamiids) to the Campanulid species lettuce (Lactuca sativa). Our results challenge the prevailing paradigm that NLR proteins exclusively signal via their N-terminal domains and reveal a signaling activity for the NB domain of NRC-dependent sensor NLRs. We propose a model in which helper NLRs monitor the status of the NB domain of their upstream sensors.
Introduction
NLRs (nucleotide binding and leucine-rich repeat) are intracellular innate immune receptors of eukaryotes and prokaryotes (Chou et al, 2023; Contreras et al, 2023a). In plants, they directly or indirectly sense pathogen virulence proteins, termed effectors, and mediate robust immune signaling and disease resistance (Contreras et al., 2023a; Duxbury et al, 2021). Plant NLRs exhibit a conserved domain architecture, consisting of an N-terminal domain, a central NB-ARC (nucleotide-binding domain shared by APAF1, R gene product and CED-4) and a C-terminal leucine-rich repeat (LRR) region (Kourelis et al, 2021). The NB-ARC can be subdivided into a nucleotide-binding (NB) domain, a helical domain 1 (HD1) and a winged-helix domain (WHD) (Chou et al., 2023; Förderer & Kourelis, 2023). Based on their N-terminal domain features, angiosperm NLRs can be broadly categorized into Toll/Interleukin-1 receptor (TIR)-type, coiled coil (CC)-type, G10 subclade coiled coil (CCG10)-type and RPW8 coiled coil (CCR)-type, which follow the NB-ARC-based NLR phylogeny (Kourelis et al., 2021). Some NLRs can function as individual units, termed singletons, mediating both pathogen perception and immune signaling (Adachi et al, 2019b). However, coevolution with pathogens has led to functionally specialized NLRs, where pathogen perception and immune signaling become uncoupled. In these cases, one NLR acts as a pathogen sensor and relies on a downstream helper NLR to mediate immune signaling and disease resistance. Sensors and helpers can function as genetically linked pairs or in higher order configurations that can include genetically unlinked receptor networks (Adachi & Kamoun, 2022; Wu et al, 2018). However, in contrast to singleton NLRs (Contreras et al., 2023a), the activation mechanisms of paired and networked NLR are less understood. In particular, the molecular mechanisms underpinning CC-NLR sensor-helper communication are not known.
The current paradigm for NLR activation is that effector perception leads to oligomerization and induced-proximity of the N-terminal domains (Contreras et al., 2023a; Duxbury et al., 2021). For CC-NLRs such as Arabidopsis ZAR1 and wheat Sr35, effector recognition triggers conformational changes in inactive NLR monomers which lead to the assembly of a pentameric resistosome complex (Förderer et al, 2022; Wang et al, 2019; Zhao et al, 2022). In the resistosome, the N-terminal α1-helices of the CC domains come together to form a funnel-like structure that inserts into the plasma membrane, presumably to perturb membrane integrity and act as a calcium channel (Bi et al, 2021; Wang et al., 2019). The α1-helices of ZAR1, Sr35 and around 20% of all angiosperm CC-NLRs are defined by the MADA motif, which is crucial for cell death induction (Adachi et al, 2019a). In some cases, the CC domain or the α1-helix on their own are sufficient to initiate cell death induction, presumably via the assembly of a resistosome-like structure that retains the capacity to de-stabilize the plasma membrane and mediate calcium influx (Adachi et al., 2019a; Bentham et al, 2018; Bi et al., 2021; Förderer et al., 2022). The CC domain is, therefore, considered as the executor domain of downstream signaling in singleton or helper CC-NLRs. In contrast, sensor NLRs have lost this activity throughout evolution due to degeneration of the MADA motif and even acquisition of novel N-terminal domains upstream of the CC domain (Adachi et al., 2019b; Contreras et al., 2023a). How sensor-helper sub-functionalization shaped the activities of paired and networked CC-NLR domains is not well understood. In particular, the domains of sensor NLRs that mediate signal transduction and communication with downstream helper NLRs are unknown.
In the Solanaceae, the NLRs required for cell death (NRC) network is composed of multiple sensor CC-NLRs and cell-surface receptors which genetically require downstream helper CC-NLRs, known as NRCs (Kourelis et al, 2022; Wu et al, 2017). This immune receptor network is of great agronomical importance, mediating immunity to diverse plant pathogenic oomycetes, fungi, nematodes, viruses, bacteria, and insects (Derevnina et al, 2021; Kourelis et al., 2022; Wu et al., 2017). All NRC-dependent sensors fall into an expanded phylogenetic clade that includes many well-known disease resistance proteins while the NRC helpers form a tight and well-supported sister clade (Wu et al., 2017). The NRC-dependent sensor NLRs themselves cluster into two distinct phylogenetic groups, the Rx-type and the Solanaceous domain (SD)-type clade (Contreras et al., 2023a). Interestingly, not all NRC-dependent sensors can activate all helpers, reflecting a degree of specificity that has probably resulted from sensor-effector and sensor-helper co-evolution (Adachi et al., 2019b; Contreras et al., 2023a). For instance, the disease resistance sensor NLRs Rx, Sw-5b, Gpa2, Bs2 and Rpi-amr3 can signal interchangeably through NRC2, NRC3 or NRC4, whereas Rpi-amr1e signals through NRC2 and NRC3 but not NRC4 and Rpi-blb2 and Mi-1.2 signal through NRC4 but not NRC2 or NRC3 (Ahn et al, 2023; Contreras et al, 2023c; Derevnina et al., 2021; Lin et al, 2022; Witek et al, 2021; Wu et al., 2017). This complex configuration of many-to-one and one-to-many functional sensor–helper connections likely contributes to increased robustness and evolvability of the NRC immune network (Adachi et al., 2019b; Contreras et al., 2023a; Wu et al., 2018).
Phylogenomic analyses revealed that the NRC superclade emerged from a pair of genetically linked NLRs prior to the split between the Asterid and Caryophyllales lineages over 100 million years ago (mya) (Wu et al., 2017). Two recent studies on NRC network diversity and macroevolution across Asterid plants mapped out massive expansions of family-specific NRC sensors and helpers in the Lamiid lineage within asterids, which include the Solanacaeae, in contrast to the more broadly conserved NRC0 helpers and their genetically linked sensors (Goh et al, 2023; Sakai et al, 2023). These studies also highlighted an overall lack of sensor-helper interchangeability across taxa (Goh et al., 2023; Sakai et al., 2023). How sensor-helper compatibility and specificity have evolved is not understood, given that the molecular determinants for sensor-helper specificity are not known.
In previous studies, we proposed an activation-and-release biochemical model for sensor-helper activation in the NRC network: Pathogen-activation of NRC-dependent sensor NLRs leads to homo-oligomerization of their NRC helpers into resistosome complexes which accumulate at the plasma membrane, separate from the sensors that activated them (Contreras et al., 2023c). Specifically, the sensor NLR proteins Rx (virus resistance), Bs2 (bacterial resistance) and Rpi-amr3 (oomycete resistance) can trigger oligomerization of NRC2 and NRC4 (Ahn et al., 2023; Contreras et al, 2023b; Contreras et al., 2023c). In contrast, Rpi-amr1e (oomycete resistance) can oligomerize NRC2 but not NRC4 and Rpi-blb2 (oomycete resistance) can oligomerize NRC4 but not NRC2 (Ahn et al., 2023; Contreras et al., 2023c). In addition, this activation mechanism appears to be conserved in the tomato (S. lycopersicum) NRC0 helper and its genetically linked sensor (Sakai et al., 2023). This indicates that the activation-and-release mechanism is likely conserved across the NRC superclade. However, the nature of the signal relayed by sensor NLRs to initiate NRC oligomerization is unknown.
The NRC-dependent sensor NLR Rx mediates extreme immunity to Potato virus X (PVX) by recognizing its coat protein (CP) via an unknown mechanism (Bendahmane et al, 1999; Bendahmane et al, 1995; Tameling & Baulcombe, 2007). In a study published over 20 years ago, Moffett and colleagues showed that Rx can perceive PVX CP and cause hypersensitive cell death when expressed in trans as two separate halves in the model plant Nicotiana benthamiana (Moffett et al, 2002). This work resulted in a a highly influential mechanistic model, which postulated that multiple intramolecular interactions maintain Rx in an inactive state. Upon effector perception, these intramolecular interactions are relieved via conformational changes within the NLR protein, resulting in dissociation of the Rx halves and initiation of immune signaling (Moffett et al., 2002). In a follow up study, Rairdan and colleagues went further, demonstrating that a ∼150 amino acid region of Rx corresponding to the NB domain (RxNB) is sufficient to trigger the hypersensitive cell death when expressed in N. benthamiana (Rairdan et al, 2008). They proposed that upon activation, intramolecular rearrangements in the Rx protein release the NB domain from negative regulation by the CC and LRR domains to enable signaling and cell death induction. However, these findings have remained puzzling considering that the prevailing models of CC-NLR activation assign the singaling activity to the very N-terminal CC and not the NB domain as discussed above. Moreover, the work by Moffett, Rairdan and colleagues predates the discovery of Rx dependence on the NRC2, NRC3 or NRC4 helpers to cause the hypersensitive cell death and confer immunity and resistance to Potato virus X (Derevnina et al., 2021; Wu et al., 2017). Whether the deconstructed and minimal Rx domains signal via the NRC helpers is unknown.
In this study, we revisited the work of Moffett, Rairdan and colleagues on Rx in the context of the NRC immune receptor network. We show that the Rx halves expressed in trans as well as RxNB-eGFP can activate downstream helper NLR oligomerization and immune signaling. In addition, the NB domain truncations of the Rx-type sensors Gpa2, Rpi-amr1e and Rpi-amr3 as well as the SD-type sensor Sw-5b can also activate NRC-dependent cell death and exhibit the same downstream helper-specificities of their full-length counterparts. Finally, we show that full-length PVX CP-activated Rx and RxNB-eGFP both trigger cell death when co-expressed with NRC2 in the unrelated Campanulid lettuce, suggesting that RxNB and NRCs are likely a minimal two-component system that can be transferred across plant taxa. Our findings reveal a novel signaling role for the NB domain in NRC-dependent sensor NLRs, serving as a minimal signal for helper activation. We suggest a model where sensor NLRs undergo conditional NB-ARC domain rearrangements upon effector perception, exposing the NB domain to activate their downstream NRC helpers and triggering oligomerization into resistosomes.
Results
The hypersensitive response caused by expression of Rx halves in trans with the coat protein (CP) of Potato virus X (PVX) is dependent on NRC helpers
To test whether the Rx halves expressed in trans activate cell death via NRC-dependent pathways, we expressed RxCCNBARC and RxLRR or full-length Rx with or without PVX CP in leaves of WT, nrc2/3, nrc4a/b or nrc2/3/4 KO N. benthamiana (Figure 1A). The constitutively active NbZAR1D481V variant was used as a control for NRC-independent cell death (Harant et al, 2022). Like full-length Rx, cell death mediated by RxCCNBARC and RxLRR expressed in trans was only abolished in the nrc2/3/4 KO N. benthamiana background but not in the nrc2/3 or nrc4a/b KO backgrounds, suggesting that it is also NRC2/3/4-dependent (Figure 1B). We also carried out complementation assays in nrc2/3/4 KO N. benthamiana lines to further confirm the NRC-dependency of cell death mediated by RxCCNBARC and RxLRR. We co-expressed the two Rx halves with PVX CP in the nrc2/3/4 background and complemented with NbNRC2, NbNRC3 and NbNRC4. We included tomato (S. lycopersicum) SlNRC0, an NRC that full length Rx is unable to signal through, as an independent negative control for complementation (Sakai et al., 2023). Like full-length Rx, the cell death in response to PVX CP mediated by the RxCCNBARC and RxLRR halves expressed in trans was restored upon complementation with NbNRC2, NbNRC3 and NbNRC4, but not SlNRC0 (Figure 1C).
(A) Schematic representation of Rx halves and NRC constructs used. Domain boundaries for Rx halves are indicated above. Approximate position of MADA, p-loop and MHD motifs are indicated above. Presence of an N-terminal MADA motif is indicated in dark red. (B-C) Representative leaves of different N. benthamiana KO lines agroinfiltrated to express constructs shown and photographed 5-7 days after infiltration. (B) Cell death mediated by RxCCNBARC and RxLRR complemented in trans is only abolished in nrc2/3/4 KO N. benthamiana plants. Red dotted circle highlights absence of hypersensitive cell death in nrc2/3/4 KO background. Wild-type Rx was included for comparison. NbZAR1D481V was included as a control for NRC-independent cell death. (C) Cell death mediated by PVX CP-activated RxCCNBARC and RxLRR is complemented by NbNRC2, NbNRC3 and NbNRC4 in leaves of nrc2/3/4 KO N. benthamiana lines when activated by co-expression of PVX CP. Free eGFP was used as a negative control for C-terminally eGFP-tagged PVX CP. SlNRC0 was used as a negative control as a helper NRC that does not get activated by Rx. Experiments were repeated 3 times with similar results.
Rx halves mediate coat protein (CP)-dependent oligomerization of their NRC2 helper
We previously reported that Rx can mediate NRC2 oligomerization and resistosome formation upon PVX CP recognition, without forming part of the activated NRC2 oligomer (Contreras et al., 2023b; Contreras et al., 2023c). To test whether the activation of the Rx halves expressed in trans also leads to oligomerization of NRC2, we leveraged previously established blue-native polyacrylamide gel electrophoresis (BN-PAGE)-based readouts for NRC resistosome formation. We co-expressed the inactive or PVX CP-activated RxCCNBARC and RxLRR halves in leaves of nrc2/3/4 KO N. benthamiana plants, complementing with the previously characterized NbNRC2EEEMADA motif mutant to abolish cell death induction without compromising helper activation (Contreras et al., 2023c). In these assays, RxCCNBARC and RxLRR expressed in trans mediated NRC2 oligomerization in a PVX CP-dependent manner, with both Rx halves being required. Notably, the NRC2 oligomer formed upon activation by RxCCNBARC and RxLRR migrates at the same height as the NRC2 oligomer formed upon activation by full-length Rx. The absence of a shift in size of the NRC2 oligomer in the Rx compared to the RxCCNBARC/RxLRR treatments suggests that the PVX CP-activated Rx halves do not form part of the mature NRC2 oligomer, much like full-length Rx. (Figure 2).
BN-PAGE assays with the inactive and PVX CP-activated RxCCNBARC and RxLRR halves co-expressed with NRC2EEE. C-terminally V5-tagged RxCCNBARC, 6xHA-tagged RxLRR and 4xMyc-tagged NbNRC2EEE were co-expressed with either free GFP or C-terminally GFP-tagged PVX CP. Full-length Rx was included for comparison and as a positive control for NRC2 oligomerization. Protein extracts were run on BN-PAGE and SDS-PAGE assays in parallel and immunoblotted with the appropriate antisera labelled on the bottom right corner of each blot. Red asterisks on the right indicates size of bands corresponding to the activated NRC2 complex. Red dotted lines indicate the molecular weight at which the full-length Rx complex migrates. V5 and HA blots were run on the same gel to allow for precise comparison of molecular weights. Approximate molecular weights (kDa) of the proteins are shown on the left for V5 and HA blots (run on the same gel), and on the right for Myc. SDS-PAGE accompanying BN-PAGE can be found in Figure S1. Experiments were repeated 3 times with similar results.
In previously published BN-PAGE assays, Rx is visualized as a complex of ∼ 400 kDa regardless of its activation state (Contreras & Kamoun, 2022; Contreras et al., 2023c). When probing for the Rx halves expressed in trans in BN-PAGE assays, we observed that RxCCNBARC also migrates as a band of ∼ 400 kDa, although this band migrates comparatively faster than full-length Rx indicating a lower molecular weight (Figure 2). RxLRR, in contrast, is visualized as a high molecular weight smear covering molecular weights from around 400 to 1000 kDa. Like full-length Rx, we did not observe any change in the migration pattern of RxCCNBARC or RxLRR upon activation of the system with PVX CP. Although we can detect signal for RxLRR at a molecular weight that overlaps with the molecular weight of the NRC2 resistosome, this signal is also present in the treatments without PVX CP activation, suggesting that it is unlikely to correspond to a pool of RxLRR that is integrated into the NRC2 oligomer. We also noted that the migration pattern of RxCCNBARC is independent of the presence of RxLRR, and vice-versa (Figure 2). We also did not observe a distinct band for RxLRR co-migrating with RxCCNBARC in BN-PAGE. The lack of co-migration in our BN-PAGE assays suggest that these two domains are not forming stable complexes and likely only associate transiently. Alternatively, a sub-pool of Rx halves that are forming a stable complex prior to activation are not highly abundant and cannot easily be detected in BN-PAGE.
The effector-independent cell death triggered by the 154 amino acid nucleotide binding domain of Rx (RxNB) is dependent on NRC2, NRC3 or NRC4
Previously, Rairdan and colleagues showed that a truncated version of the NRC-dependent sensor Rx encoding only the 154 amino acid NB domain region of the NB-ARC fused to eGFP (RxNB-eGFP) was capable of constitutively triggering cell death in N. benthamiana and N. tabacum (Rairdan et al., 2008). Based on these results, we hypothesized that the NB domain could encode the minimal signal for NRC helper activation. To test this hypothesis, we performed cell death assays with RxNB-eGFP in WT and nrc2/3/4 KO N. benthamiana plants. We included the constitutively active ZAR1D481V and RxD460V variants as controls for NRC-independent and NRC-dependent cell death, respectively, and eGFP as a negative control for cell death (Figure 3A). We were able to reproduce the previously reported cell death triggered by RxNB-eGFP in WT N. benthamiana. This cell death was abolished in leaves of nrc2/3/4 KO N. benthamiana, indicating that RxNB-eGFP activates NRC-dependent hypersensitive cell death. This suggests that the NB domain of Rx is necessary and sufficient to activate downstream NRC helpers (Figure 3A).
(A) Photo of a representative leaves from WT and nrc2/3/4 KO N. benthamiana plants expressing RxNB-eGFP. NbZAR1D481V and RxD460V were included as controls for NRC-independent and NRC-dependent cell death, respectively. eGFP was included as a negative control for cell death. Leaves were agroinfiltrated to express the constructs indicated and photographed 5-7 days after infiltration. The experiment was repeated three times with at least 6 technical replicates per repeat, with similar results in all cases. (B) Schematic representation of RxNB-eGFP construct. Domain boundaries used are indicated above, and were the same as reported previously by Rairdan and colleagues (Rairdan et al., 2008). Position of highly conserved lysine (K) residue of the p-loop is indicated, with numbering corresponding to its position in full-length Rx. (C) Representative leaf of nrc2/3/4 KO N. benthamiana expressing RxNB-eGFP complemented with NbNRC2, NbNRC3 or NbNRC4. SlNRC0 was included as a negative control for complementation. NbZAR1D481V and RxD460V were included as controls for NRC-independent and NRC-dependent cell death, respectively. A quantitative analysis of the cell death can be found in Figure S6. (D) BN-PAGE assays with inactive and activated C-terminally 4xMyc-tagged NRC2EEE. NRC2EEE was activated either with Rx-6xHA/PVX CP-eGFP or RxNB-eGFP. Total protein extracts were run on native and denaturing PAGE assays in parallel and immunoblotted with the appropriate antisera labelled in the bottom left corner of each blot. Approximate molecular weights (kDa) of the proteins are shown on the right. Accompanying SDS-PAGE assays can be found in Figure S4. The experiment was repeated 3 times with similar results.
Mutating a highly conserved K residue in the p-loop motif typically makes NLRs non-functional (Mestre & Baulcombe, 2006). To test the p-loop dependency of RxNB-eGFP-mediated activation of NRCs, we generated RxNB-eGFP p-loop mutants (K to R substitution in position 176, as per numbering in full-length Rx, Figure 3B) and expressed them in leaves of WT and nrc2/3/4 KO N. benthamiana. Interestingly, much like RxNB-eGFP, the RxNB-eGFP p-loop mutants triggered NRC-dependent cell death. This indicates that unlike full-length Rx and the Rx halves expressed in trans (Moffett et al., 2002), cell death triggered by RxNB-eGFP does not require an intact p-loop (Figure S2).
To further test the NRC-dependency of RxNB-eGFP triggered cell death, we carried out complementation assays in leaves of nrc2/3/4 KO N. benthamiana. We co-expressed RxNB-eGFP with NbNRC2, NbNRC3 and NbNRC4 or SlNRC0, a tomato NRC that full-length Rx is unable to signal through (Sakai et al., 2023). The cell death mediated by the RxNB-eGFP was restored upon complementation with NbNRC2, NbNRC3 and NbNRC4, but not SlNRC0 (Figure 3C). We conclude that RxNB-eGFP is indeed activating cell death via NRC-dependent pathways and that it retains the specificity to activate the same set of NRC helpers that full-length Rx signals through in N. benthamiana.
To investigate the contributions of the C-terminal eGFP tag to the cell death mediated by RxNB-eGFP, we generated RxNB-mCherry-6xHA fusions and tested them for NRC-dependent cell death. These RxNB-mCherry-6xHA fusions also triggered NRC-dependent cell death, although this cell death was weaker than that triggered by RxNB-eGFP (Figure S3A). In complementation assays carried out in leaves of nrc2/3/4 KO N. benthamiana, cell death triggered by RxNB-mCherry-6xHA was also complemented by NbNRC2, NbNRC3 and NbNRC4 but not SlNRC0 (Figure S3B). We used western blot analysis to test for RxNB-mCherry-6xHA protein accumulation levels in nrc2/3/4 KO N. benthamiana. Strikingly, although this construct triggers NRC-dependent cell death in WT N. benthamiana, we were unable to detect any signal for RxNB-mCherry-6xHA (Figure S3C). Based on these results, we conclude that the cell death triggered by the RxNB truncations is not specific to RxNB-eGFP. Since RxNB-eGFP-triggered cell death is much stronger and easier to detect in planta, we decided to proceed with C-terminally eGFP-tagged variants for all subsequent experiments.
RxNB mediates effector-independent oligomerization of the NRC2 helper
We next sought to investigate if RxNB is sufficient to trigger NRC2 oligomerization. To this end, we co-expressed RxNB and NbNRC2EEE in leaves of nrc2/3/4 KO N. benthamiana and performed BN-PAGE assays at different timepoints, with Rx/PVX CP as a positive control for NRC2 oligomerization. Interestingly, expression of RxNB-eGFP was sufficient to mediate the formation of NRC2 high molecular weight complexes of a similar size as the NRC2 oligomers formed upon activation with full-length Rx/PVX CP. Importantly, the absence of a shift in size of the NRC2 oligomers triggered by RxNB-eGFP compared to those triggered by Rx/PVX CP suggests that, much like full-length Rx, RxNB-eGFP does not form part of the mature NRC2 oligomer (Figure 3D). We noted that compared to the NRC2 oligomer signal induced by full-length Rx and PVX CP, RxNB-eGFP-activated NRC2 treatments showed a weak signal corresponding to inactive NRC2 in all timepoints analyzed, indicating that RxNB-eGFP mediated helper oligomerization may be less efficient (Figure 3D). We did not observe any signal for RxNB-eGFP at a molecular weight matching that of the NRC2 oligomer, which further suggests that the mature NRC2 resistosome does not include RxNB-eGFP (Figure 3D).
The nucleotide binding (NB) domains of the sensor NLRs Gpa2, Rpi-amr1e, Rpi-amr3 and Sw-5b are sufficient to activate downstream NRC helpers
To what degree does the activation of NRC helpers by the NB domain of sensor NLRs apply to other disease resistance proteins? We investigated this by generating NB domain-eGFP fusions of 8 well-studied disease resistance NLR proteins that are dependent on NRC helpers in the immune receptor network (Figure 4A, Figure S5). This panel covered the phylogenetic diversity of NRC-dependent sensors and was composed of the Rx-type sensors Gpa2, Rpi-amr1e, Rpi-amr3 and Bs2, and the SD-type sensors Mi-1.2, Rpi-blb2 and Sw-5b (Figure 4A) (Contreras et al., 2023a). Gpa2NB-eGFP, Rpi-amr1eNB-eGFP and Sw-5bNB-eGFP triggered NRC-dependent cell death to levels comparable to RxNB-eGFP (Figure 4B, Figure S6). Rpi-amr3NB-eGFP triggered weak NRC-dependent cell death in a subset of the leaves tested (6 out of 18 total infiltration spots) (Figure 4B, Figure S6). In contrast, Bs2NB-eGFP, Rpi-blb2NB-eGFP and Mi-1.2NB-eGFP did not trigger cell death (Figure 4B, Figure S6). In parallel, we determined the protein accumulation levels of each NB domain-eGFP fusion in planta. All of the sensor NLR NB domains that induced cell death accumulated in planta (Rx, Gpa2, Rpi-amr1e and Sw-5b). In contrast, the NB domain truncations that triggered weak or no cell death accumulated to comparatively lower levels (Rpiamr3, Bs2, Rpi-blb2 and Mi-1.2) (Figure S7). We conclude that NB domain-mediated activation of downstream NRC helpers is not exclusive to Rx and can also be triggered by other Rx-type and SD-type sensors of the NRC network.
(A) Schematic representation of sensor-helper signaling specificities in the NRC network. Rx-type sensors Rx, Gpa2, Rpi-amr1e and Rpi-amr3 are represented with three domains (CC, NB-ARC and LRR) whereas SD-type sensors Sw-5b, Rpi-blb2 and Mi-1.2 are represented with four domains (SD, CC, NB-ARC and LRR). (B) Representative photos of cell death assays with the constructs indicated in leaves of either WT or nrc2/3/4 KO N. benthamiana. NbZAR1D481V and eGFP were included as positive and negative controls for cell death, respectively. RxD460V was included as a control for NRC-dependent cell death. Leaves were agroinfiltrated to express the constructs indicated and photographed 5-7 days after infiltration. One representative leaf is shown. A quantitative analysis of the cell death assays can be found in Figure S6. (C) Representative photos of cell death assays with the constructs indicated in leaves of nrc2/3/4 KO N. benthamiana. NbZAR1D481V and RxD460V were included as positive and negative controls for cell death, respectively. Leaves were agroinfiltrated to express the constructs indicated and photographed 5-7 days after infiltration. One representative leaf is shown. A quantitative analysis of the cell death assays can be found in Figure S6.
The nucleotide binding (NB) domains of the sensor NLRs Gpa2, Rpi-amr1e and Sw-5b retain the NRC helper specificities of their full-length counterparts
Sensor NLRs in the NRC network can signal via different subsets of downstream NRC helpers. For example, PVX CP-activated Rx, RBP1-activated Gpa2 and NSm-activated Sw-5b can signal interchangeably via NRC2, NRC3 and NRC4, whereas AVRamr1-activated Rpi-amr1e can signal via NRC2 and NRC3 but not NRC4 (Figure 4A, Figure S8) (Contreras et al., 2023b; Contreras et al., 2023c; Derevnina et al., 2021; Lin et al., 2022; Witek et al., 2021; Wu et al., 2017). We next sought to determine whether, Gpa2NB-eGFP, Rpi-amr1eNB-eGFP and Sw-5bNB-eGFP retained the same downstream NRC specificities of their full-length counterparts, similar to RxNB-eGFP. To this end, we performed complementation assays in leaves of nrc2/3/4 KO N. benthamiana co-expressing each of these NB domains together with NbNRC2, NbNRC3 or NbNRC4, respectively. We included SlNRC0 as a negative control for complementation. We did not include Rpi-amr3 in these experiments as Rpi-amr3NB-eGFP accumulated poorly in planta and triggered comparatively weaker cell death, making it harder to draw robust conclusions (Figure 4, Figure S6, Figure S7). The cell death triggered by Gpa2NB-eGFP and Sw-5bNB-eGFP was restored upon co-expression with NbNRC2, NbNRC3 and NbNRC4. Rpi-amr1eNB-eGFP cell death was restored upon co-expression with NbNRC2 and NbNRC3 but not with NbNRC4. (Figure 4C). None of the NB domains was complemented by SlNRC0. This indicates that NB domain-eGFP fusions retain the same helper NRC specificity profiles of their full-length counterparts.
RxNB and its helper NRC2 form a minimal functional unit that can be transferred from Solanaceae to the Asteraceae plant lettuce
We leveraged transient expression in lettuce (Lactuca sativa) by agroinfiltration (Wróblewski et al, 2018) to investigate the capacity of the Rx/NRC2 pair to function in a species distantly related to the Solanaceae. Lettuce is an Asteraceae in the Campanulid lineage and is estimated to have split from Nicotiana (Solanaceae, Lamiid) about 102 Million years ago (Zeng et al, 2014). Lettuce predates the expansion of the NRC helpers that occurred in Solanaceae and other lamiids and therefore lacks orthologs of NRC2, NRC3 and NRC4 and other Solanaceae NRCs (Goh et al., 2023; Sakai et al., 2023). We first determined whether Rx/CP and NRC2 can trigger cell death in lettuce by co-expressing these three proteins in lettuce leaves. We included an autoactive variant of the previously published lettuce CCG10-NLR RGC2B (RGC2BD470V) as a positive control for cell death (Meyers et al, 1998; Shen et al, 2002). We observed a cell death response indicative of immune activation when all three components, CP, Rx and NRC2 were co-expressed in leaves of lettuce. This cell death was abolished when any one of these three components was missing (Figure 5A).
Photo of representative leaves of lettuce (L. sativa) var. “Fenston”. (A) Lettuce leaf showing cell death after co-expression of PVX CP, Rx and NbNRC2. No cell death was observed when any one of the three components (PVX CP, Rx or NbNRC2) was absent. (B) Photo of representative leaves of lettuce (L. sativa) var. “Fenston” showing cell death after co-expression of RxNB-eGFP and NbNRC2. No cell death was observed with eGFP or with RxNB-eGFP and NbNRC2 when expressed individually. (A-B) Images were taken 5-7 days after agroinfiltration, and leaves were imaged from both adaxial (left side of panel) and abaxial (right side of panel) sides. Images from abaxial side were included as HR cell death in lettuce is more visible on abaxial than adaxial sides. The constitutively active variant of the lettuce NLR RGC2B (RGC2BD470V) was used as a positive control for cell death in lettuce. Experiments were repeated three times with similar results.
Next, we tested whether RxNB-eGFP and NRC2 were also capable of mediating cell death in lettuce. We included RGC2BD470V and eGFP as positive and negative controls for cell death, respectively. Co-expression of RxNB-eGFP and NbNRC2 triggered cell death in lettuce, whereas expression of either RxNB-eGFP or NbNRC2 alone did not (Figure 5B). This result suggests that no additional proteins are required for RxNB-eGFP to activate NRC2. Alternatively, additional proteins that may be required are conserved between N. benthamiana (Lamiid) and lettuce (Campanulid). We conclude that RxNB-eGFP and NRC2 are most likely a minimal two-component system that can be transferred from N. benthamiana to lettuce.
Discussion
In this study, we build upon findings by Moffett, Rairdan, and colleagues regarding the ability of individual domains of the virus disease resistance protein Rx to activate immunity in trans, or as a nucleotide-binding (NB) domain (Moffett et al., 2002; Rairdan et al., 2008). We revisited this work within the framework of the activation-and-release biochemical model for sensor-helper activation in the NRC network of CC-NLRs (Contreras et al., 2023c). Our findings demonstrates that PVX CP-triggered cell death, mediated by the CC-NBARC (RxCCNBARC) and LRR (RxLRR) domains of Rx when expressed in trans, as well as the effector-independent cell death mediated by RxNB-eGFP, are dependent on NRC helpers and involves the formation of NRC2 resistosomes. Similar to the full-length Rx, the Rx halves and RxNB-eGFP do not integrate into the mature NRC2 helper oligomer (Figure 1, Figure 2, Figure 3). Furthermore, our work reveals that the ∼150 amino acid NB domain fragments of other sensor NLRs, such as Gpa2, Rpi-amr1e, Rpi-amr3 and Sw-5b, also induce constitutive cell death. Importantly, RxNB, Gpa2NB, Rpi-amr1eNB, and Sw-5bNB maintain the downstream helper specificity profiles of their full-length counterparts (Figure 3, Figure 4). Finally, we demonstrate that both full-length PVX CP-activated Rx and RxNB-eGFP can initiate cell death in lettuce when co-expressed with NRC2, suggesting that this represents a minimal, two-component system transferable to unrelated plant taxa (Figure 5). Our data support a working model in which the conditional exposure of sensor NLR NB domains upon pathogen perception serves as a signal for downstream NRCs that triggers helper oligomerization and resistosome formation (Figure 6).
Schematic representation of Rx halves and RxNB-eGFP-mediated NRC activation. Prior to effector-triggered activation, NRC-dependent sensors such as Rx are held in an inactive conformation by intramolecular interactions. These intramolecular interactions between the different Rx domains hide the sensor NLR NB domain, preventing its signaling, and contribute to the previously reported association of the RxCCNBARC and RxLRR halves (Moffett et al., 2002). Upon perception of PVX CP, Rx undergoes a series of p-loop dependent conformational changes that expose the NB domain and lead to dissociation of the Rx halves expressed in trans (Moffett et al., 2002). This conditional NB domain exposure allows it to signal to downstream helpers such as NRC2 leading to its homo-oligomerization and resistosome formation. Following activation by their upstream sensor, NRC helper oligomers signal via their N-terminal CC domain. In the case of the RxNB-eGFP and other sensor NB domain-eGFP fusions, the NB domain is exposed and therefore constitutively activates downstream helpers. Because no intramolecular rearrangements are required to relieve autoinhibitory intramolecular interactions, the p-loop mutation does not abolish this cell death.
The prevailing paradigm for NLR activation is that their N-terminal domains mediate immune signaling upon resistosome assembly. This is the case for the N-terminal CC domains of singleton and helper CC-NLRs and for the N-terminal TIR domains of sensor TIR-NLRs (Bentham et al., 2018; Contreras et al., 2023a). Based on our data, we propose that for NRC dependent sensor NLRs, the NB domain region of the central NB-ARC, not the N-terminal CC domain, executes signaling to downstream helpers. How does the NB domain of sensor NLRs initiate the activation of NRC helpers? Is this activation achieved through direct binding or does it necessitate enzymatic activity, as is the case for TIR-NLRs and their downstream CCR-NLR helpers (Huang et al, 2022; Ma et al, 2020)? The activation of full-length Rx and the Rx halves expressed in trans requires an intact p-loop (Moffett et al., 2002). In sharp contrast, RxNB activation of NRCs is p-loop independent (Figure S2). This observation indicates that while ATP-binding is required for conditional NB domain exposure, it does not play a role in the recognition of the sensor NB domain by the helper once the NB domain is exposed, making sensor NB domain-catalyzed ATP hydrolysis an improbable mechanism of helper activation.
Our transient expression assays in lettuce, along with recently published data demonstrating that Campanulid NRC sensor-helper pairs can initiate immune signaling in N. benthamiana (Goh et al., 2023; Sakai et al., 2023), lead us to favor a model in which sensor NB domains and NRC helpers directly interact, with no additional components required for sensor-helper communication. While it remains possible that additional conserved components exist between N. benthamiana and lettuce, the evolutionary divergence between these two plant species—estimated at 102 million years— makes this hypothesis less likely (Zeng et al., 2014). Importantly, BN-PAGE assays do not reveal any evidence of NB domain integration into the NRC2 resistosome, suggesting that this interaction is transient, as previously suggested for full-length sensor activation of NRCs (Contreras et al., 2023c). To date, conclusive evidence that NRC-dependent sensors and their NRC helpers form stable complexes has not been obtained, possibly because of the transient nature of this interaction. Notably, that the NB domain alone is sufficient to activate NRCs suggests that sensor NLRs are unlikely to be required as scaffolds for helper NLR polymerization during sensor-helper activation, as is the case for the mammalian NAIP/NLRC4 sensor-helper pairs (Zhang et al, 2015). Further biochemical and structural analyses will provide insights into the precise dynamics of sensor-helper interactions during activation.
According to a well-established paradigm of plant immunity, NLRs are known to guard critical host components, known as guardees or decoys, that are perturbed during pathogen infection (Dangl & Jones, 2001; Van Der Biezen & Jones, 1998; van der Hoorn & Kamoun, 2008). We propose that helper NLRs, such as the NRCs, may have evolved to guard the conformational state of the NB-ARC domains in their upstream sensors, enabling them to detect any pathogen-triggered modifications. The NB-ARC domain stands out as the most conserved feature within the NLR protein family, and NB-ARC rearrangements have been demonstrated as critical for NLR activation across various life forms (Chou et al., 2023; Duxbury et al., 2021). It is plausible that NRC helpers have evolved to monitor the activation status of the NB-ARC domains of their upstream sensors, recognizing NB-ARC conformational states indicative of effector perception and immune activation. This concept is reminiscent of how CCR-NLR helpers, such as NRG1 and ADR1, have evolved to sense the activation of their upstream TIR-NLR sensors by detecting the formation of EDS1-SAG101/PAD4 heterodimers induced following TIR-NLR resistosome assembly and small molecule production (Feehan et al., 2023; Huang et al., 2022; Jia et al., 2022). We propose that helper NLRs in the NRC family function as “guards of sensors”, monitoring sensor NLR activation. In this model, sensor NLRs could even be considered as decoys, as they have degenerated features such as non-functional MADA motifs and N-terminal domain integrations (SD) and have lost the capacity to trigger cell death on their own (Adachi et al., 2019b; Contreras et al., 2023a).
The discovery that Rx and NRC2 can effectively function in lettuce carries significant implications for disease resistance breeding and bioengineering (Marchal et al, 2022). These findings raise the possibility of transferring NRC sensor-helper pairs across crop species at least within Asterid plants. This potential transfer may even extend to agronomically important species outside of the Asteraceae, such as soybean, wheat, or rice, which lack NRCs altogether (Wu et al., 2017). Furthermore, it opens the door to the prospect of transferring entire networks of NRC sensors and helpers across species boundaries, capitalizing on the resilience that networked signaling architectures offer to immune systems (Wu et al., 2018). The transfer of the NRC network across species is particularly exciting given that it confers resistance to multiple species of pathogens and pests (Derevnina et al., 2021; Kourelis et al., 2022; Wu et al., 2017). Additionally, the observation that Rx can recognize PVX CP in lettuce, triggering cell death, prompts important inquiries about the recognition mechanism of PVX CP. Previous work suggested that Rx indirectly recognizes PVX CP and that this recognition necessitates the host component RanGAP2 (Moffett et al., 2002; Sacco et al., 2007; Tameling & Baulcombe, 2007). The fact that Rx can still respond to PVX CP in lettuce raises the possibility that either Rx retains its proper functionality with the lettuce RanGAP2 ortholog or that Rx recognition of PVX CP might involve direct, albeit transient, interactions. Further research is required to elucidate the precise role of RanGAP2 in Rx functionality and to unravel the specific mechanism by which Rx perceives PVX CP.
A significant, unresolved question is deciphering the molecular basis of sensor-helper specificity, a fundamental characteristic of the NRC network. While certain sensors can activate multiple helpers within a given host, others selectively activate a subset of the NRC helper repertoire. Furthermore, across evolutionary timescales, sensors from one species can lose the ability to communicate with NRC helpers from distantly or even closely related species, as is the case of species-specific sensor-helper NRC pairs in lamiids (Goh et al., 2023; Sakai et al., 2023). Our data suggests that the NB domain plays a crucial role as a molecular determinant for sensor-helper specificity on the sensor side. However, although the sensor NB is necessary and sufficient for NRC activation, the contributions of other sensor NLR domains to downstream helper activation remain unclear. Furthermore, the domains within the helper proteins involved in monitoring sensor NLR activation are yet to be identified. Understanding the domains and interfaces involved in sensor and helper protein interactions, and their evolutionary dynamics, will provide additional insights into the mechanisms through which paired and networked CC-NLRs activate. Identifying the precise molecular communication between sensors and helpers could have substantial implications for disease resistance engineering, potentially enabling the synthetic expansion and optimization of sensor-helper interactions, resulting in more robust and efficient immune receptor networks.
Materials and Methods
Plant growth conditions
Wild-type, nrc2/3, nrc4a/b and nrc2/3/4 KO Nicotiana benthamiana lines as well as lettuce (L. sativa) var. “Fenston” were grown in a controlled environment growth chamber with a temperature range of 22–25°C, humidity of 45–65% and a 16/8-h light/dark cycle.
Plasmid construction
The Golden Gate Modular Cloning (MoClo) kit (Weber et al, 2011) and the MoClo plant parts kit (Engler et al, 2014) were used for cloning, and all vectors are from this kit unless specified otherwise. Cloning design and sequence analysis were done using Geneious Prime (v2021.2.2; https://www.geneious.com). Rx, RxD460V, CP-eGFP and NRC2EEE constructs used were previously described (Contreras et al., 2023c). NbZAR1D481V construct was previously described (Harant et al., 2022). All NRC constructs used were previously described (Derevnina et al., 2021; Sakai et al., 2023). Rx halves were cloned into pICH86988 acceptor with integrated 35S promoter and OCS terminator. All NB domain constructs were cloned into the pJK268c vector with pICSL51288 (2×35S promoter) and pICSL41414 (35S terminator). Domain boundaries for all sensor NB domains were based on amino acid sequence alignment to the original RxNB domain boundaries reported by Rairdan and colleagues (Figure S5) (Rairdan et al., 2008), and subsequently ordered as synthetic gene fragments (Azenta/Genewiz). RGC2BD470V construct was synthesized and cloned into pJK001c acceptor with pICSL51288 (2×35S promoter) and pICSL41414 (35S terminator). C-terminal tag modules used were pICSL50012 (V5), pICSL5007 (3xFLAG), pICSL50034 (eGFP), pICSL5009 (6xHA) or pICSL50010 (4xMyc), as indicated.
Cell death assays by agroinfiltration
Proteins of interest were transiently expressed in N. benthamiana and lettuce according to previously described methods (Bos et al, 2006). Briefly, leaves from 4–5-week-old plants were infiltrated with suspensions of A. tumefaciens, using the GV3101 pM90 strain for N. benthamiana and the C58C1 strain for lettuce. Strains were transformed with expression vectors coding for different proteins indicated. Final OD600 of all A. tumefaciens suspensions were adjusted in infiltration buffer (10 mM MES, 10 mM MgCl2, and 150 μM acetosyringone (pH 5.6)). Final OD600 used was 0.3 for each construct. Whenever multiple constructs were co-infiltrated into an individual spot, the total concentration of bacteria was kept constant across treatments by adding untransformed A. tumefaciens when necessary. This was to avoid an effect from differences in the total OD600 of bacteria in each treatment.
Extraction of total proteins for BN-PAGE and SDS–PAGE assays
Four to five-week-old plants were agroinfiltrated as described above with constructs of interest and leaf tissue was collected 3 days post agroinfiltration. Final OD600 used was 0.3 for each NLR immune receptor, Rx halves or NB domain, and 0.2 for eGFP or CP-eGFP. BN-PAGE was performed using the Bis-Tris Native PAGE system (Invitrogen) according to the manufacturer’s instructions. Leaf tissue was ground using a Geno/Grinder tissue homogenizer and total protein was subsequently extracted and homogenized extraction buffer. GTMN extraction buffer was used (10% glycerol, 50 mM Tris–HCl (pH 7.5), 5 mM MgCl2 and 50 mM NaCl), supplemented with 10 mM DTT, 1x protease inhibitor cocktail (SIGMA) and 0.2% Triton X-100 (SIGMA). Samples were incubated in extraction buffer on ice for 10 min with short vortex mixing every 2 min. Following incubation, samples were centrifuged at 5,000xg for 15 min and the supernatant was used for BN-PAGE and SDS–PAGE assays.
BN-PAGE assays
BN-PAGE assays were done as described previously (Ahn et al., 2023; Contreras et al., 2023c). Brieflly samples extracted as detailed above were diluted as per the manufacturer’s instructions by adding NativePAGE 5% G-250 sample additive, 4x Sample Buffer and water. After dilution, samples were loaded and run on Native PAGE 3–12% Bis-Tris gels alongside either NativeMark unstained protein standard (Invitrogen) or SERVA Native Marker (SERVA). The proteins were then transferred to polyvinylidene difluoride membranes using NuPAGE Transfer Buffer using a Trans-Blot Turbo Transfer System (Bio-Rad) as per the manufacturer’s instructions. Proteins were fixed to the membranes by incubating with 8% acetic acid for 15 min, washed with water and left to dry. Membranes were subsequently re-activated with methanol to correctly visualize the unstained native protein marker. Membranes were immunoblotted as described below.
SDS–PAGE assays
For SDS–PAGE, samples were diluted using a 3:1 ratio of sample to SDS loading dye and denatured at 72 °C for 10 min. Denatured samples were spun down at 5,000 g for 3 min and supernatant was run on 4–20% Bio-Rad 4–20% Mini-PROTEAN TGX gels alongside a PageRuler Plus prestained protein ladder (Thermo Scientific). The proteins were then transferred to polyvinylidene difluoride membranes using Trans-Blot Turbo Transfer Buffer using a Trans-Blot Turbo Transfer System (Bio-Rad) as per the manufacturer’s instructions. Membranes were immunoblotted as described below.
Immunoblotting and detection of BN-PAGE and SDS–PAGE assays
Blotted membranes were blocked with 5% milk in Tris-buffered saline plus 0.01% Tween 20 (TBS-T) for an hour at room temperature and subsequently incubated with desired antibodies at 4°C overnight. Antibodies used were anti-GFP (B-2) HRP (Santa Cruz Biotechnology), anti-HA (3F10) HRP (Roche), anti-Myc (9E10) HRP (Roche), anti-FLAG (M2) HRP (Sigma), and anti-V5 (V2260) HRP (Roche), all used in a 1:5,000 dilution in 5% milk in TBS-T. To visualize proteins, we used Pierce ECL Western (32106, Thermo Fisher Scientific), supplementing with up to 50% SuperSignal West Femto Maximum Sensitivity Substrate (34095, Thermo Fishes Scientific) when necessary. Membrane imaging was carried out with an ImageQuant LAS 4000 or an ImageQuant 800 luminescent imager (GE Healthcare Life Sciences, Piscataway, NJ). Rubisco loading control was stained using Ponceau S (Sigma) or Ponceau 4R (Irn Bru, AG Barr).
Data availability
All relevant data are within the article and in the Supplementary materials. This study includes no data that would need to be deposited in external repositories.
Author contributions
Mauricio P Contreras: Conceptualization; data curation; formal analysis; supervision; validation; investigation; visualization; methodology; writing – original draft; project administration; writing— review and editing. Hsuan Pai: Conceptualization; Data curation; formal analysis; supervision; validation; investigation; methodology; writing—review and editing. Rebecca Thompson: Data curation; formal analysis; validation; investigation; visualization. Jules Claeys: Data curation; formal analysis; validation; investigation. Hiroaki Adachi: Conceptualization; investigation; writing—review and editing. Sophien Kamoun: Conceptualization; resources; supervision; funding acquisition; visualization; writing—original draft; project administration; writing—review and editing.
Disclosure and competing interest statement
S.K. receives funding from industry on NLR biology and has cofounded a start-up company (Resurrect Bio Ltd.) related to NLR biology. S.K. and M.P.C. have filed patents on NLR biology. M.P.C. has received fees from Resurrect Bio Ltd.
Supplementary Figures
Protein extracts were run on SDS-PAGE assays and immunoblotted with the appropriate antisera labelled on the left. Approximate molecular weights (kDa) of the proteins are shown on the right. Rubisco loading control was carried out using Ponceau stain (PS). Experiment was repeated three times with similar results.
Photo of representative leaves of WT and nrc2/3/4 KO N. benthamiana plants expressing RxNB-eGFP and an RxNB-eGFP variant with a mutation in the conserved p-loop motif (K to R amino acid substitution). Experiment was repeated three times with at least 6 technical replicates for each repeat. All replicates showed similar results.
(A) Photo of representative leaves of WT and nrc2/3/4 KO N. benthamiana plants expressing RxNB-mCherry-6xHA. NbZAR1D481V and RxNB-eGFP were included as controls for NRC-independent and NRC-dependent cell death, respectively. mCherry-6xHA was included as a negative control for cell death. Images were taken 5-7 days post agroinfiltration. (B) Photo of representative leaves of nrc2/3/4 KO N. benthamiana plants co-expressing RxNB-mCherry-6xHA with NbNRC2, NbNRC3 or NbNRC4. Complementation with SlNRC0 was included as a negative control. NbZAR1D481V and RxD460V were included as controls for NRC-independent and NRC-dependent cell death, respectively. Images were taken 5 days post agroinfiltration. (C) Protein extracts were run on SDS-PAGE assays and immunoblotted with the appropriate antisera labelled on the left. Approximate molecular weights (kDa) of the proteins are shown on the right. Rubisco loading control was carried out using Ponceau stain (PS). Experiment was repeated three times with similar results.
Protein extracts were run on SDS-PAGE assays and immunoblotted with the appropriate antisera labelled on the left. Approximate molecular weights (kDa) of the proteins are shown on the right. Rubisco loading control was carried out using Ponceau stain (PS). Experiment was repeated three times with similar results.
NB domain boundaries of sensors cloned in this study were determined via alignment to the original RxNB domain truncation reported by Rairdan and colleagues (Rairdan et al., 2008). Amino acid positions indicated correspond to the NB domain alone, with residue in position 1 in RxNB being equivalent to residue in position 139 for full-length Rx. Alignments were generated using Clustal Omega (Sievers et al, 2011).
(A) Quantitative analysis of cell death assays shown in Figure 4B. (B) Quantitative analysis of cell death assays shown in Figure 3C and Figure 4C. (A-B) HR cell death was scored based on a modified 0-7 scale (Segretin et al, 2014) at 5-7 days post agroinfiltration. HR scores are presented as dot plots, where the size of each dot is proportional to the number of samples with the same score (Count). Results are based on 3 biological replicates.
Protein extracts from leaves of nrc2/3/4 KO N. benthamiana expressing proteins of interest were run on SDS-PAGE assays and immunoblotted with the appropriate antisera labelled on the left. Approximate molecular weights (kDa) of the proteins are shown on the right. Rubisco loading control was carried out using Ponceau 4R staining (Irn-BruTM). Experiment was repeated two times with similar results.
(A) Photo of a representative leaf of WT N. benthamiana showing cell death after co-expression of Gpa2 and RBP1. (B) Photo of a representative leaf of nrc2/3/4 KO N. benthamiana showing cell death after co-expression of Gpa2/RBP1 and NRC2 or NRC4. The S. lycopersicum helper SlNRC0 was included as a negative control for complementation. GFP was used as a negative control for RBP1-GFP. (A-B) Images were taken 5-7 days after agroinfiltration. Experiment was repeated 2 times with similar results.
Acknowledgements
We thank D. Lüdke and C. Marchal for valuable comments on this article. We thank members of the Jones lab (The Sainsbury Laboratory), including Camille-Madeleine Szymansky, Hee-Kyung Ahn and Jonathan Jones, for valuable scientific discussions. We thank Tania Toruño and Peter-Paul Damen (Rijk Zwaan, Netherlands) for providing seeds of lettuce (L. sativa) var. ‘Fenston’ and valuable scientific discussions. We thank all members of the TSL Support Services for their invaluable assistance. M.P.C. thanks S. Scorza for support and L. A. “El Flaco” Spinetta for inspiration. The Kamoun Lab is funded primarily from the Gatsby Charitable Foundation, Biotechnology and Biological Sciences Research Council (BBSRC, UK, BB/WW002221/1, BB/V002937/1, BBS/E/J/000PR9795 and BBS/E/J/000PR9796) and the European Research Council (BLASTOFF).