In the game process with various pathogens, plants have a two-layered immune system, including pattern-triggered immunity (PTI) and effector-triggered immunity (ETI) [1] . PTI and ETI have been demonstrated to interact with each other [2] [3] . PTI is initiated by pattern recognition receptors (PRRs) located on the cell membrane and can resist most pathogens. Some pathogens have evolved a type III secretion system, which transmits effectors to plant cells to inhibit PTI, leading to plant susceptibility. Once the corresponding effectors are directly or indirectly monitored through intracellular immune receptors, which are mostly proteins encoding nucleotide-binding leucine-rich repeat receptors (NLRs), ETI is activated and usually goes with characteristic cell death, known as hypersensitivity response (HR) [4] .

Simply put, according to the differences in the N-terminus, NLR receptors can be divided into two categories: TIR (toll/interleukin-1)-NBS-LRR (TNL)and CC (coiled-coil)-NBS-LRR (CNL) [5] . Transient expression of some NLR receptors N-termini in Nicotiana benthamiana is enough to generate severe hypersensitive response [6] - [11] . The NBS domain is relatively conserved in NLR receptors, containing many extremely conserved motifs such as P-loop and MHD (Met-His-Asp) motifs, which are considered molecular switches for NLR protein activity [12] . The LRR domain is usually unstable and variable, involving NLR receptor and effector interaction, and auto-inhibition of NLR receptors in the absence of pathogens [13] .

Many NLR receptors function on the plasma membrane (PM). Previous studies have shown that RPS5 (RESISTANCE TO PSEUDOMONAS SYRINGAE 5), L5 (AT1G12290) and R5L1 (RPS5-like 1) all function on the PM, and their N-terminal myristoylation site is important for their function and subcellular localization [11] [14] [15] . Recent reports have shown that ADR1 (ACTIVATED DISEASE RESISTANCE 1), NRG1.1 and NRG1.2 (N Requirement Gene 1s (NRG1s)) proteins share similar activation mechanisms; when activated, they generated advanced oligomeric permeable thoroughfare related to Ca²⁺ on PM [16] [17] . Notably, the CC domain of barley MLA10 (mildew locus a (Mla)) has been proven to form homologous dimers, and some conserved hydrophobic residue (I33, L36, and M43) alleles can abolish the self-interaction of CC module [6] . Interestingly, these hydrophobic sites are extremely conserved in many NLR proteins, including RPM1 (Resistance to Pseudomonas syringae pv. Maculicola 1). Mutations of the corresponding hydrophobic residues (I31, M34, and M41) can significantly affect the self-association of RPM1 CC and the function of full-length RPM1 [18] . It is currently unclear whether these hydrophobic sites are related to self-interaction, or whether their roles in each NLR are diverse.

Previous article reported a novel NLR protein SUT1, which is involved in autoimmunity initiated by type one protein phosphatase 4 mutation (topp4-1) in Arabidopsis [19] , however, its role in planta is still unclear. This study employed Nicotiana benthamiana, a model platform, to conduct an overall structural and functional analysis of SUT1 protein.

What is the activation mechanism of the disease-resistant protein SUT1 (SUPPRESSORS OF TOPP4-1, 1)?

As a model plant, N. benthamiana can complete gene expression regulation and protein translation modification within the plant, making it more suitable for functional verification of heterologous genes than E. coil and yeast. With the rapid development of Agrobacterium mediated gene transient transformation systems, it has become more convenient to express exogenous proteins in N. benthamiana instantaneously, rapidly, and in large quantities for functional research. Therefore, N. benthamiana is currently widely used in protein subcellular localization, protein interaction, biopharmaceutical production, and other fields.

We employed the N. benthamiana transient expression system to deeply analyze the activation mechanism of SUT1 from both structural and functional aspects.

The seeds of Arabidopsis and N. benthamiana are laboratory-owned (Key Laboratory of Chronic Diseases, Fuzhou Medical University, Fuzhou, China), and both model plants are cultured in specific greenhouses at 25˚C with a 16 hours light period.

All plasmid vectors in this paper were constructed through the Gateway cloning strategy (Thermo Fisher Scientific). Briefly, using cDNA from wild-type Arabidopsis leaves as the initial template. The gel recovery products from multiple DNA fragments, including full-length (FL) SUT1 and its mutants, various SUT1 truncated fragments, and CC domain mutants, were cloned into PENTR/D (Thermo Fisher Scientific) through a step cloning kit (Vazyme, #C112-02), respectively. After complete sequencing, they were transferred to the expression vector pEarleygate101 for YFP (Yellow Fluorescent Protein)-HA tagged through Gateway cloning technology (Thermo Fisher Scientific). The expression plasmids were electro-transferred into agrobacterium strain GV3101.

Agrobacteria (GV3101) carrying the constructs were grown overnight in Luria Bertani medium (LB) with suitable antibiotics. Three milliliters of Agrobacterium culture were centrifuged and resuspended in MES buffer (10 mM MgCl₂, 10 mM MES pH 5.6, and 150 μM acetosyringone). Agrobacteria were incubated at room temperature for 1 h and infiltrated into the leaves of 4-week-old N. benthamiana at specific OD₆₀₀ values. Agrobacteria containing the P19 construct were co-infiltrated at an OD₆₀₀ of 0.2. Cell death phenotypes were photographed at indicated time point. The white solid circle represents obvious cell death, and the white dashed circle represents no visible cell death.

Three 8-mm leaf disks were harvested and ground to powder in liquid nitrogen. Total protein was extracted with 100 μL extraction buffer (20 mM Tris-HCl pH 8.0, 5 mM ethylene diamine tetraacetic acid (EDTA), 1% SDS, and 10 mM DL-dithiothreitol (DTT)). Lysate was boiled at 98˚C with 1× protein loading buffer for 10 min. The total protein extract was cleared by centrifuge at 15,000 g for 10 min. Then, the supernatant was separated by 10% SDS–PAGE gels and detected with corresponding antibodies. Antibodies used for immunoblotting include anti-HA (Roche) anti-H⁺-ATPase (Agrisera) and anti-β-actin (Abbkine).

In brief, sucrose buffer [20 mM Tris (pH 8.0), 0.33 M sucrose, 1 mM EDTA, 5 mM DTT, and 1× Sigma plant protease inhibitor cocktail] was added to the homogenized tissue at a ratio of 5 µL per mg (FW) tissue. The extract was centrifuged at 5000 × g for 10 min at 4˚C; then, the supernatant was transferred to a new tube and designated as total protein (T). Cytoplasmic fraction (C) was prepared by harvesting the supernatant after spinning the total protein fraction at 20,000 × g for 1 h at 4˚C. The total membrane fraction was prepared from the resulting pellet by resuspending in 200 µL of buffer B (Minute™ PM protein isolation kit, Invent Biotechnologies). After centrifugation at 7800 × g for 15 min at 4˚C, the supernatant was transferred to 2 mL Eppendorf tube and mixed with 1.6 mL cold phosphate belanced solution (PBS) buffer mixed by vortexing and spun at 16,000 × g for 1 h at 4˚C to pellet the PM fraction. The pellet was resuspended with sucrose buffer in 4 times less volume than the soluble fraction. The resulting fraction was labeled as the PM-enriched/microsomal fraction. Protein fractions were run on SDS-PAGE gels and analyzed by western blotting.

In short, a laser confocal microscope (Leica SP8) and a fluorescence microscope (Olympus BX53) were employed to observe images of live cells on the abaxial sides of N. benthamiana leaves at 28 or 40 h post-target protein expression. YFP fluorescence was excited at 514 nm.

Live images of N. benthamiana and specific infection steps refer to the literature [20] and website (https://bio-protocol.org/en/bpdetail?id=2063&type=0).

We inferred that SUT1 G2 and C4 sites are necessary for targeting SUT1 to the PM. To verify this hypothesis, we designed a SUT1 version with mutations in the important sites, SUT1_G2AC4S. The subcellular localization pattern of SUT1-YFP (Yellow Fluorescent Protein)-HA and SUT1_G2AC4S-YFP-HA were examined. As expected, we observed that SUT1 exhibited typical membrane localization, while the introduction of G2AC4S resulted in SUT1 mutant exhibiting significant cytoplasmic distribution (Figure 1(A)), suggesting that the two acylation sites were prominent in SUT1 PM distribution.

To prove the fluorescence results, we carried out membrane fractionation assays. We employed H⁺-ATPase and Ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco) serve as markers of membrane and soluble fractions, respectively. Immunoblot results indicated that wild-type SUT1 mainly existed in the microsomal fraction containing PM, while on the contrary, acylation deficient variant SUT1_G2AC4S was mainly found in the soluble fraction (Figure 1(B)). Thus, SUT1 is anchored to the PM through two acylation residues.

SUT1 containing several domains. In order to further determine the region that can attract cell death, we designed a series of truncated fragments and expressed

them in N. benthamiana, respectively (Figure 2(A)). We observed that only the CC domain has cell-death-inducing activity, leading to a significant hypersensitivity response in N. benthamiana leaves (Figure 2(B)). The immunoblot results showed that all variants exhibited comparable protein abundance, ruling out the possibility that these fragments could not mediate cell death due to incorrect expression (Figure 2(C)).

We further analyzed the localization of these fragments employing a fluorescence microscope, and found that any fragment containing two acylation sites exhibited membrane-localized, while fragments lacking these two residues displayed significant cytoplasmic and nuclear distribution (Figure 2(D)). Thus, the membrane-localized SUT1 CC module was enough to cause cell death in N. benthamiana.

We found through sequence alignment that these hydrophobic residues are highly conserved in receptors including RPM1, Sr33, and SUT1 (Figure 3(A)). Naturally, we introduced alleles of these residues in the CC domain to analyze the impact of these hydrophobic sites on the functionality of the SUT1 CC module. The results declared that both CC (L36E) and CC (L39E) lost their activity to trigger cell death, and CC (I46E) significantly weakened its function (Figure 3(B)), although none of the versions affected the protein accumulation of the SUT1 CC domain (Figure 3(C)), promulgating that the three hydrophobic residues in CC domain were important for its function.

Furthermore, we investigated the subcellular distribution of these CC variants, and discovered that only CC (L36E) emerged evident cytoplasmic positioning (Figure 3(D)). We inferred that these important hydrophobic residues were likely involved in the self-connection of the SUT1 CC domain. Thus, the function and localization of the CC domain are affected by three hydrophobic residue mutations.

The motifs in the NBS domain, such as P-loop and MHD, are highly conserved in NLR receptors. To directly verify whether SUT1 protein is a typical CNL receptor, we designed SUT1 (D473V) and SUT1 (K187R/D473V) variants, which correspond to MHD motif mutant, as well as P-loop and MHD motifs double mutant, respectively (Figure 4(A)). The results of transient expression in N. benthamiana manifested that SUT1 (D473V) provoked a strong hypersensitivity response, while P-loop mutation completely abolished the cell death caused by SUT1 (D473V). Immunoblotting indicated that both proteins mentioned above can be expressed correctly (Figure 4(B) and Figure 4(C)). In summary, SUT1 protein is a typical CNL receptor.

We further analyze whether the hydrophobic residue mutations that affect the

function of the SUT1 CC domain impact on the function and subcellular localization of the full-length [FL] SUT1. Fluorescence data revealed that all mutations significantly altered the PM localization of wild-type SUT1 (Figure 4(D)). We observed that co-expression of SUT1 and topp4-1, rather than wild-type TOPP4, can induce cell death in N. benthamiana (Figure 4(E)). Naturally, we co-expressed three SUT1 alleles with topp4-1 in N. benthamiana, and identified that all mutants abolished or compromised the function of wild-type SUT1 (Figure 4(F)). Thus, functional loss mutations in the CC domain influence the function of FL SUT1.