Powdery mildew, caused by Erysiphe quercicola, is one of the most destructive foliar diseases affecting rubber tree (Hevea brasiliensis) plantations in China. Severe infections can result in yield losses of up to 45%, representing a major challenge to latex production, particularly in key rubber-producing regions (Li et al., 2023; Cao et al., 2024). The development of disease-resistant cultivars offers a sustainable solution for controlling this disease, making the identifying of resistance genes and understanding their underlying molecular mechanisms essential.

In plants, Nucleotide-binding leucine-rich repeat (NLR) proteins play a pivotal role as primary immune receptors, recognizing pathogen effectors and triggering effector-triggered immunity (ETI) (Duggan et al., 2021). This often leads to a hypersensitive response (HR), a form of programmed cell death that limits pathogen spread and enhances resistance (Huang et al., 2023; Sun et al., 2024).

The RAR1 (required for MLA12 resistance) gene was first identified in barley during studies of MLA12-mediated resistance to powdery mildew (Shen et al., 2003; Bieri et al., 2004). The RAR1 protein contains two highly conserved zinc-binding domains, termed CHORD-I and CHORD-II (cysteine- and histidine-rich domains), which are separated by a 20-amino acid linker harboring three cysteine residues and one histidine (CCCH motif) (Zhang et al., 2010; Hong et al., 2013). Each CHORD domain coordinates one zinc ion, and these zinc ions are essential for proper protein folding and structural stability (Hong et al., 2013). Accumulating evidence indicates that RAR1 is a critical component of immune signaling in both monocotyledonous and dicotyledonous plants. It functions as a key regulator in powdery mildew resistance pathways and is required for the activity of multiple resistance (R) genes against diverse pathogens (Shen et al., 2003). In barley and rice, rar1 mutants fail to accumulate reactive oxygen species and do not exhibit an HR upon pathogen challenge, resulting in heightened susceptibility to Blumeria graminis and Magnaporthe grisea (Shen et al., 2003; Jarosch et al., 2005).

Accumulating studies have established that the molecular chaperone complex composed of HSP90, SGT1, and RAR1 serves as a central hub in regulating plant innate immunity (Seo et al., 2008; Ito et al., 2015; Zhang et al., 2015; Katsuyama et al., 2021; Yuan et al., 2021). This complex plays a critical role in plant defense by stabilizing NLR immune receptors and modulating their activation during pathogen invasion. RAR1 interacts directly with both HSP90 and SGT1. This interaction is essential for maintaining the stability of certain resistance proteins (e.g., RPS2 in Arabidopsis, Rx in potato), thereby contributing significantly to immune signaling (Takahashi et al., 2003; Boter et al., 2007). In wheat, for example, Wang et al. demonstrated that the interaction between TaRAR1, TaSGT1, and TaHSP90 enhances R gene-mediated resistance to stripe rust (Wang et al., 2015). Similarly, in barley, resistance to powdery mildew triggered by the MLA6 NLR receptor is dependent on functional RAR1 and SGT1 (Shen et al., 2003). In Solanum lycopersicum, the NLR protein Sw-5b confers resistance to tomato spotted wilt virus (TSWV) in an SGT1-dependent manner (Chen et al., 2021). These findings collectively highlight that while the involvement of HSP90, SGT1, and RAR1 in immune responses is broadly conserved, the modes of their interaction and their specific roles in immunity can vary depending on the NLR protein and the context of the pathogen.

In rubber tree, eight HSP90 genes have been identified, with HbHSP90.1 functioning as a molecular chaperone for SGT1 and playing a key role in disease resistance (Liu et al., 2022). HbSGT1b interacts with HbHSP90.1 in the nucleus, and HbHSP90.1 also associates with the NLR protein HbCNL2, further supporting the involvement of this chaperone complex in immune signaling (Wang et al., 2023). Additionally, the NLR protein HbCNL4 is specifically induced in rubber tree epidermal cells following pathogen invasion, and it triggers HR in multiple plant species. This suggests HbCNL4 likely functions downstream of the HSP90–SGT1–RAR1 complex (Liang et al., 2023; Zhang et al., 2025). However, the mechanism by which the HSP90–SGT1–RAR1 complex interacts with NLR proteins in woody plants remains poorly understood, and the specific role of RAR1 in activating CNL-type NLRs in the rubber tree is unknown. Elucidating these interactions will deepen our understanding of immune receptor regulation in perennial species and provide molecular targets for breeding disease-resistant rubber trees.

RAR1 is a core component of plant immune signaling and functions in multiple NLR-mediated pathways. However, its role in rubber tree has not been fully explored. In this study, we identified two homologs of the RAR1 gene, HbRAR1–1 and HbRAR1-2, both of which contain two conserved CHORD domains typical of RAR1 proteins. The proteins differ by only four amino acids within these domains, reflecting a high degree of structural conservation. Phylogenetic analysis indicates that HbRAR1s are most closely related to MeRAR1, confirming the evolutionary conservation of RAR1 across plant species (Muskett et al., 2002; Bieri et al., 2004). These findings support the hypothesis that RAR1 proteins play a conserved role in plant immunity.

In plants, HSP90, RAR1, and SGT1 form a dynamic chaperone complex that stabilizes NLR receptors and maintains their signaling activity (Seo et al., 2008; Ito et al., 2015). In Arabidopsis and barley, RAR1 binds the HATPase domain via its CHORD-I domain, whereas SGT1 associates with the same domain through its CS region, and both CHORD domains also interact with SGT1 (Takahashi et al., 2003). By contrast, our results in rubber tree indicate that both CHORD domains of HbRAR1 can interact with HbHSP90.1 and HbSGT1b, suggesting an adaptable interaction pattern that may have evolved to accommodate specific NLR clients. Biologically, this interaction diversity implies that the HSP90–SGT1–RAR1 chaperone machinery is fine-tuned for different immune receptors, rather than following a rigid one-size-fits-all mechanism.

In H. brasiliensis, HbHSP90.1 was previously shown to interact with HbSGT1b (Liang et al., 2023; Wang et al., 2023), but whether RAR1 is involved in this regulatory network was unknown. Here, Y2H and BiFC assays demonstrated that HbRAR1–1 and HbRAR1–2 interact with HbHSP90.1 and HbSGT1b. Domain mapping revealed that both CHORD-I and CHORD-II domains of HbRAR1 bind the HATPase domain and the CS domain. Interestingly, full-length HbRAR1 and HbSGT1b interact in both the cytoplasm and nucleus, while HbRAR1–HbHSP90.1 interactions predominantly occur in the nucleus. The HbSGT1b–HbHSP90.1 interaction displayed dynamic subcellular relocalization over time, and all domain-level interactions were confined to the nucleus. In tobacco, phosphorylation-mimicking mutations in the MAPK-targeted region of SGT1 promote its nuclear accumulation and restrict the N protein to the nucleus, thereby altering NLR nucleo-cytoplasmic partitioning (Hoser et al., 2013). The HSP90–RAR1–SGT1 complex has been implicated in both stabilizing and degrading NLRs to balance immune activation and repression (Azevedo et al., 2002; Kadota and Shirasu, 2012; Chapman et al., 2022). Thus, we propose that post-translational modification of HbSGT1b may regulate HbSGT1b–HbHSP90.1 localization and, consequently, modulate NLR-mediated immunity.

The interaction between HbRAR1 and HbHSP90.1/HbSGT1b is critical not only for maintaining NLR protein stability and conformational readiness but also for facilitating their activation and downstream defense responses.

Our previous research identified HbHSP90.1 and HbSGT1b as chaperones associated with powdery mildew resistance in the rubber tree. Specifically, HbHSP90.1 interacts with HbCNL2, a member of the same family as HbCNL4, playing a critical role in the accumulation and stability of HbCNL2 (Liu et al., 2022; Liang et al., 2023; Wang et al., 2023). This observation suggests a potentially similar mechanism for HbCNL4. Furthermore, literature reviews indicate that the HSP90–SGT1–RAR1 ternary complex is vital in other NLR protein-mediated hypersensitive responses. For instance, Zhang et al. demonstrated that the hypersensitive response mediated by the tomato Rx4 protein against Xanthomonas euvesicatoria pv. perforans race T3 relies on SGT1 and RAR1 (Zhang et al., 2020). In N. benthamiana, NbNRC2, NbNRC3, and NbNRC4 are NLR proteins integral to plant immunity, and research has shown that their proper function necessitates SGT1 and HSP90 (Dong et al., 2022). In this study, we demonstrated that HbCNL4 interacts with both HbHSP90.1 and HbSGT1b. Moreover, HR and ion leakage assays showed that the combined presence of HbHSP90.1, HbSGT1b, and HbRAR1 significantly amplifies HbCNL4-triggered cell death, indicating the ternary complex of HbHSP90.1–HbSGT1b–HbRAR1 plays a crucial role in the HbCNL4-triggered HR response through its interaction with HbCNL4.

However, the role of RAR1 in the regulation of NLRs may not be universal across all plant species or immune pathways. Previous studies revealed that some NLR-mediated resistances require both SGT1 and RAR1, others depend on only one, and some function independently of either protein (Oh et al., 2014; Chapman et al., 2021; Wu et al., 2025). For instance, barley MLA6-mediated resistance to powdery mildew requires both RAR1 and SGT1, whereas MLA1, despite 95% sequence identity with MLA6, requires neither (Azevedo et al., 2002). Tomato Sw-5b confers resistance to TSWV in an SGT1-dependent but RAR1-independent manner (Chen et al., 2021). This specificity in chaperone dependence underscores a nuanced layer of immune regulation in plants, whereby certain resistance pathways rely on general chaperone support while others have become self-sufficient. Importantly, such insights carry practical implications for crop improvement. In the context of rubber tree, knowing whether a candidate NLR gene requires the HSP90–SGT1–RAR1 complex can inform breeding strategies: by stacking or selecting immune receptors with differing co-chaperone requirements, breeders could ensure that no single pathogen strategy (e.g. targeting a co-chaperone) can extinguish all defenses, ultimately leading to more robust and durable disease resistance in H. brasiliensis.

Based on these findings, we propose a molecular model (Figure 4E). In rubber tree, HbRAR1 interacts with HbHSP90.1 and HbSGT1b through their two CHORD domains, binding to the HATPase domain of HbHSP90.1 and the CS domain of HbSGT1b, respectively. These interactions facilitate the assembly of the HbHSP90.1–HbSGT1b–HbRAR1 ternary complex. Furthermore, the core components of this complex, HbHSP90.1 and HbSGT1b, associate with HbCNL4, thereby maintaining the stability of the HbCNL4 protein and potentially playing a crucial role in HbCNL4-mediated HR. Our work provides new insights into the molecular mechanisms of disease resistance in rubber trees and offers a foundation for future efforts to exploit the HSP90–SGT1–RAR1 complex in improving crop resilience.