Introduction

Plants activate powerful and effective defence responses upon pathogen perception by their cell-surface or intracellular immune receptors (Jones & Dangl, 2006). Microbes can be recognised via cell-surface pattern recognition receptors (PRRs) that detect pathogen-associated molecular patterns (PAMPs) (Chinchilla et al., 2007; Heese et al., 2007) and activate pattern-triggered immunity (PTI) (Boller & Felix, 2009). Pathogens in turn can suppress host recognition and defence via effectors (Toruño et al., 2016), which function either in the apoplast or are translocated into the host cell. These effectors may interfere with PAMP perception (Wawra et al., 2016) or PTI signalling (Fabro et al., 2011; Irieda et al., 2019). In turn, some pathogen effectors are detected by intracellular nucleotide-binding domain leucine-rich repeat (NLR) immune receptors (Jones & Dangl, 2006), leading to the activation of effector-triggered immunity (ETI), which often culminates in a hypersensitive cell death response (HR) that restricts pathogen invasion (Jones & Dangl, 2006). Nucleotide-binding domain leucine-rich repeats carry either a Toll/Interleukin 1/Resistance (TIR) domain (TIR–NLRs, or TNLs) or a coiled-coil (CC) domain at their N-termini (Jones & Dangl, 2006).

Effector detection by NLRs activates defences and thwarts pathogen growth (Armstrong et al., 2005; Rehmany et al., 2005). Effectors can be recognised by direct binding to an NLR, or indirectly by their interactions with a host protein, which is either ‘guarded’ by NLRs or has evolved to mimic a host target (a ‘decoy’) (Cui et al., 2015). Alternatively, some NLRs detect multiple sequence-unrelated effectors through an integrated domain or by forming protein complexes (Sarris et al., 2015; Guo et al., 2018; Martel et al., 2020). Effector recognition by some NLRs can involve a post-LRR (PL) domain that comprises a C-terminal jelly-roll and Ig-like domain (C-JID) that contributes to effector binding to form a resistosome, (a wheel-like pentameric structure formed by oligomerisation) that upon activation activates defence (Ma et al., 2020; Martin et al., 2020). Understanding the molecular basis of such interactions by identifying recognised effectors in the pathogen is important for elevating crop disease resistance.

White blister rust in crop and wild Brassicaceae species is caused by oomycete pathogens in the genus Albugo (Holub et al., 1995; Voglmayr & Riethmüller, 2006; Choi et al., 2007). Dispersal is by dehydrated sporangiospores (Heller & Thines, 2009). Approximately 24 physiological races of Albugo candida have been defined, primarily based on which Brassicaceae species they infect (Hiura, 1930; Pound & Williams, 1963; Saharan & Verma, 1992). For example, A. candida races 2 (Ac2V), 7 (Ac7V) and 9 (AcBoT) infect Brassica juncea, B. rapa and B. oleracea (Gupta et al., 2018). Race 4 infects Capsella bursa-pastoris and Arabidopsis thaliana (Fairhead, 2016) and can also infect Camelina sativa (Castel et al., 2021). Independent isolates of race 4 (AcEm2 and AcNc2) were found on C. bursa-pastoris in Kent (UK; Borhan et al., 2008) and infected field-grown A. thaliana plants in Norwich (UK; McMullan et al., 2015). Race 4 variant AcEx1 was isolated from wild A. halleri (Fairhead, 2016), which grows and sporulates on most A. thaliana accessions and on C. sativa (Castel et al., 2021). Jouet et al. (2019) verified that different physiological races of A. candida exhibit distinct host specificities within Brassicaceae. Genome comparisons by McMullan et al. (2015) revealed rare recombination events between these races.

Albugo sp. have a marked capacity to impose host immunosuppression (Cooper et al., 2008; Belhaj et al., 2017; Prince et al., 2017). Albugo infection enhances host susceptibility and enables growth of nonadapted fungal and oomycete pathogens (Cooper et al., 2008), including the potato late blight pathogen, Phytophthora infestans (Belhaj et al., 2017; Prince et al., 2017). The races studied in the work reported here are summarised in Table 1.

Table 1. Resistance and susceptibility phenotypes on adult leaves of Arabidopsis accessions to different Albugo candida races.

Albugo candida isolate	Race	Host	Infection phenotype on A. thaliana
			Col-0	Ws-2	Col-0-wrr4a-6	Col-0-wrr4b
Ac2V	Race 2	Brassica juncea	Resistant (GR)	Resistant (GR)	Resistant (GR)	Resistant (GR)
Ac7V	Race 7	Brassica rapa	Resistant (GR)	Resistant (GR)	Resistant (GR)	Resistant (GR)
AcBoT	Race 9	Brassica oleracea	Resistant (GR)	Resistant (GR)	Resistant (GR)	Resistant (GR)
AcNc2	Race 4	Arabidopsis thaliana	Resistant (GR)	Susceptible (S)	Resistant (NCR)	Resistant (GR)
AcEm2	Race 4	Capsella bursa-pastoris	Resistant (GR)	Susceptible (S)	Resistant (NCR)	Resistant (GR)
AcEx1	Race 4	Arabidopsis halleri	Susceptible (S)	Susceptible (S)	Susceptible (S)	Susceptible (S)

• Host specificity of different A. candida races and their observed infection phenotypes on different accession of A. thaliana (GR, green resistance; NCR, necrotic–chlorotic resistance; S, susceptibility). Host indicates the plant host from which the A. candida isolate was originally isolated.

Genome analysis of Albugo laibachii and A. candida revealed a class of secreted proteins that carry a ‘CHxC’ motif (Kemen et al., 2011; Furzer et al., 2022). Re-sequencing of the Ac2V isolate of A. candida from B. juncea using long reads, revealed a c. two-fold expansion of CHxC effector-like proteins in A. candida compared with A. laibachii (Kemen et al., 2011). These are now reclassified as CX₂CX₅G and abbreviated to CCG effectors (Furzer et al., 2022). Every A. candida race has c. 80–100 CCG proteins that comprise c. 10% of the secretome (Furzer et al., 2022). CCG genes show signatures of diversifying selection and display elevated rates of pseudogenisation and presence/absence variation, consistent with their selection for diversity while maintaining virulence functions (Furzer et al., 2022). We address here the question of whether secreted A. candida CCG proteins are effectors, by evaluating whether any are recognised by White Rust Resistance (WRR) genes, and testing if some confer enhanced disease susceptibility.

Infection phenotypes conferred by White Rust Resistance 4 (WRR4) are classified into resistant (green resistant, GR), partially resistant with chlorosis or necrosis but no pustules (necrotic–chlorotic resistant, NCR), and susceptible, with pustules (Susceptible, S; Cevik et al., 2019). WRR4 from A. thaliana confers resistance to A. candida (Borhan et al., 2008). The Col-0 locus contains three paralogues that encode TIR–NLR (TNL) immune receptors. The Col-0 allele of WRR4A confers resistance to four A. candida races (Borhan et al., 2008), and resistance to race Ac2V in transgenic B. juncea (Borhan et al., 2010). The Col-0 allele of WRR4B also confers resistance to A. candida races Ac2V, Ac7V and AcBoT (Cevik et al., 2019). Although resistance in Col-0 functions against multiple Brassica-infecting A. candida races, some variants of race 4 (e.g. AcEx1) can grow and sporulate on Col-0, but with a chlorotic phenotype (Fairhead, 2016; Jouet et al., 2019). Some accessions, for example Oystese (Oy-0), resist race 4 isolate AcEx1, due to an allele of WRR4A with a C-terminal extension of 80 amino acids (Fairhead, 2016; Castel et al., 2021).

Leaves of A. thaliana accession Wassilewskija (Ws-2) are resistant to A. candida race 2 and race 7. WRR4A in Ws-2 carries deletions compared with Col-0. Ws-2 contains two divergent WRR4 paralogues (Van de Weyer et al., 2019) and one of these (a Ws-2 allele of WRR4B) confers resistance to A. candida race 2 (from B. juncea). Both Col-0 and Ws-2 alleles of WRR4B confer resistance in transgenic Brassicas to A. candida race 2 (Cevik et al., 2019). Allelic variation of TIR–NLR paralogues at the WRR4 locus therefore provides multiple genes that could control white rust in major Brassica crops. Identifying A. candida effectors recognised by WRR4A and WRR4B would help choose the most effective transgene combinations for Albugo control.

Here, we compared the genomes of different races of A. candida with the goal of identifying the cognate recognised effectors for the Col-0 alleles of WRR4A and WRR4B. We screened multiple CCG secreted proteins, mainly from A. candida races 2 and 4, using Agrobacterium-mediated transient co-expression to identify pairwise combinations of effector and NLR variant that activate an HR. Twelve CCG candidates – eight recognised by WRR4A and four by WRR4B – show activation of HR when transiently co-expressed, and were validated for their recognition by a bombardment assay in Col-0 wild-type and mutants (wrr4a-6 or wrr4b) of A. thaliana. Several of these CCGs are absent and some show expression or allelic polymorphism in the Col-0 virulent isolate AcEx1. For WRR4A-recognised CCGs, the N-terminal 100 amino acids are sufficient for recognition. Our data reveal that two distinct WRR4 paralogues confer broad-spectrum resistance by recognition of multiple CCG effectors across distinct clades in the A. candida CCG effectorome. Moreover, some of the CCGs confer enhanced susceptibility to another oomycete pathogen Hyaloperonospora arabidopsidis (Hpa), consistent with the idea that CCG secreted proteins are authentic A. candida effectors.

中文翻译：

---

拟南芥 WRR4A 和 WRR4B 旁系同源 NLR 蛋白均赋予对多个 Albugo 念珠菌效应器的识别

介绍

植物通过其细胞表面或细胞内免疫受体感知病原体后，会激活强大而有效的防御反应 (Jones & Dangl, 2006 )。微生物可以通过细胞表面模式识别受体 (PRR) 识别，该受体检测病原体相关分子模式 (PAMP)（Chinchilla等人，2007 年；Heese等人，2007 年）并激活模式触发免疫 (PTI)（Boller & 费利克斯，2009 年）。病原体反过来可以通过效应器抑制宿主识别和防御（Toruño等人，2016 年）), 它们在质外体中起作用或易位到宿主细胞中。这些效应器可能会干扰 PAMP 感知（Wawra等人，2016 年）或 PTI 信号（Fabro等人，2011 年；Irieda等人，2019 年）。反过来，一些病原体效应子被细胞内核苷酸结合域富含亮氨酸重复序列 (NLR) 免疫受体检测到 (Jones & Dangl, 2006 )，导致效应子触发免疫 (ETI) 的激活，这通常最终导致过敏限制病原体入侵的细胞死亡反应 (HR) (Jones & Dangl, 2006). 核苷酸结合域富含亮氨酸的重复序列在其 N 末端携带一个 Toll/白细胞介素 1/抗性 (TIR) 域（TIR–NLRs 或 TNLs）或一个卷曲螺旋 (CC) 域（Jones & Dangl, 2006） .

NLR 的效应子检测可激活防御并阻止病原体生长（Armstrong等人，2005 年；Rehmany等人，2005 年）。效应器可以通过直接结合 NLR 或间接通过它们与宿主蛋白的相互作用来识别，宿主蛋白要么被 NLRs“保护”，要么已经进化为模仿宿主目标（“诱饵”）（Cui 等人，2015年） ). 或者，一些 NLR 通过整合域或通过形成蛋白质复合物检测多个与序列无关的效应子（Sarris等人，2015 年；Guo等人，2018 年；Martel等人，2020 年）). 一些 NLR 对效应子的识别可能涉及 LRR 后 (PL) 结构域，该结构域包含 C 末端果冻卷和 Ig 样结构域 (C-JID)，后者有助于效应子结合形成抗性体（轮状五聚体）寡聚化形成的结构）激活后激活防御（Ma等人，2020 年；Martin等人，2020 年）。通过识别病原体中公认的效应子来了解这种相互作用的分子基础对于提高作物抗病性非常重要。

作物和野生十字花科物种中的白色疱锈病是由Albugo属的卵菌病原体引起的（Holub等人，1995 年；Voglmayr 和 Riethmüller，2006 年；Choi等人，2007 年）。分散是通过脱水的孢子囊孢子进行的（Heller & Thines，2009）。已经定义了大约 24 个白斑念珠菌的生理小种，主要基于它们感染的十字花科物种（Hiura，1930 年；Pound 和 Williams，1963 年；Saharan 和 Verma，1992 年）。例如，A. candida种族 2 (Ac2V)、7 (Ac7V) 和 9 (AcBoT) 感染Brassica juncea、B. rapa和B. oleracea（Gupta等人，2018 年）。Race 4 感染荠菜和拟南芥(Fairhead, 2016 )，也可以感染亚麻荠(Castel et al ., 2021 )。种族 4 的独立分离株（AcEm2 和 AcNc2）在肯特（英国；Borhan等人，2008 年）和诺里奇（英国；McMullan等人，2015 年）受感染的田间种植的拟南芥植物中发现). 第 4 种变体 AcEx1 是从野生A. halleri (Fairhead, 2016 ) 中分离出来的，它在大多数拟南芥种质和C. sativa上生长并形成孢子（Castel等人，2021）。Jouet等人。( 2019 ) 验证了A. candida的不同生理种族在十字花科中表现出不同的宿主特异性。McMullan等人的基因组比较。( 2015 ) 揭示了这些种族之间罕见的重组事件。

白蛋白服务提供商。具有施加宿主免疫抑制的显着能力（Cooper等人，2008 年；Belhaj等人，2017 年；Prince等人，2017 年）。Albugo感染增强了宿主的易感性并促进了不适应的真菌和卵菌病原体的生长（Cooper等人，2008 年），包括马铃薯晚疫病病原体致病疫霉（Belhaj等人，2017 年；Prince等人，2017 年）。此处报告的工作中研究的种族总结在表 1 中。

表 1.拟南芥种质成年叶片对不同白色念珠菌种群的抗性和敏感性表型。

白色念珠菌分离物	种族	主持人	拟南芥的感染表型
			Col-0	WS-2	Col-0- wrr4a-6	Col-0- wrr4b
交流电压	比赛 2	芥菜	耐药 (GR)	耐药 (GR)	耐药 (GR)	耐药 (GR)
AC7V	第 7 场比赛	白菜	耐药 (GR)	耐药 (GR)	耐药 (GR)	耐药 (GR)
机器人	第 9 场	甘蓝	耐药 (GR)	耐药 (GR)	耐药 (GR)	耐药 (GR)
AcNc2	第 4 场比赛	拟南芥	耐药 (GR)	易感 (S)	耐药 (NCR)	耐药 (GR)
AcEm2	第 4 场比赛	荠菜	耐药 (GR)	易感 (S)	耐药 (NCR)	耐药 (GR)
AcEx1	第 4 场比赛	拟南芥	易感 (S)	易感 (S)	易感 (S)	易感 (S)

• 不同拟南芥假丝酵母种群的宿主特异性及其观察到的感染表型（GR，绿色抗性；NCR，坏死-褪绿抗性；S，易感性）。Host 表示最初分离出A. candida分离株的植物宿主。

Albugo laibachii和A. candida的基因组分析揭示了一类携带“CHxC”基序的分泌蛋白（Kemen等人，2011 年；Furzer等人，2022 年）。使用长读长对来自芥菜的念珠菌A. candida的 Ac2V 分离株进行重新测序，发现 a c。与A. laibachii相比，念珠菌中 CHxC 效应样蛋白的两倍扩增（Kemen等人，2011 年）。这些现在被重新分类为 CX ₂ CX ₅ G 并缩写为 CCG 效应器（Furzer等人，2022 年）。每个A. candida种族都有c。包含c的 80–100 个 CCG 蛋白。10% 的分泌蛋白组 (Furzer et al ., 2022 )。CCG基因表现出多样化选择的特征，并表现出较高的假基因化率和存在/不存在变异，这与它们在保持毒力功能的同时对多样性的选择一致（Furzer 等人，2022年）。我们在这里解决分泌的念珠菌CCG 蛋白是否是效应物的问题，方法是评估是否有任何蛋白被白锈病抗性( WRR)识别) 基因，并测试某些基因是否增强了疾病易感性。

White Rust Resistance 4 ( WRR4 ) 赋予的感染表型分为耐药（green resistant, GR）、部分耐药伴萎黄或坏死但无脓疱（necrotic–chlorotic resistant, NCR）和易感，伴有脓疱（Susceptible, S ; Cevik等人，2019 年）。来自拟南芥的WRR4赋予对念珠菌的抗性（Borhan等人，2008 年）。Col-0 基因座包含三个编码 TIR-NLR (TNL) 免疫受体的旁系同源物。WRR4A的 Col-0 等位基因赋予对四种念珠菌种群的抗性（Borhan等人., 2008 ), 以及转基因芥菜对 Ac2V 种族的抗性(Borhan et al ., 2010 )。WRR4B的 Col-0 等位基因也赋予对念珠菌Ac2V、Ac7V 和 AcBoT 的抗性（Cevik等人，2019 年）。尽管 Col-0 中的抗性对多个感染芸苔属的念珠菌小种起作用，但第 4 种小种的某些变体（例如 AcEx1）可以在 Col-0 上生长并形成孢子，但具有褪绿表型（Fairhead，2016 年；Jouet等人，2019). 一些种质，例如 Oystese (Oy-0)，由于WRR4A的等位基因具有 80 个氨基酸的 C 端延伸，因此抵抗种族 4 分离株 AcEx1（Fairhead，2016 年；Castel等人，2021 年）。

A. thaliana accession Wassilewskija (Ws-2)的叶子对A. candida race 2 和 race 7具有抗性。与 Col-0 相比，Ws-2 中的WRR4A携带缺失。Ws-2 包含两个不同的WRR4旁系同源物（Van de Weyer等人，2019 年），其中一个（ WRR4B的 Ws-2 等位基因）赋予对A. candida race 2（来自B. juncea）的抗性。WRR4B的 Col-0 和 Ws-2 等位基因均赋予转基因芸苔属植物对念珠菌小种 2 的抗性（Cevik等人，2019 年）。TIR-NLR 旁系同源物的等位变异因此， WRR4基因座提供了多个可以控制主要芸苔属作物白锈病的基因。识别WRR4A和WRR4B识别的念珠菌效应器将有助于选择最有效的转基因组合来控制Albugo。

Here, we compared the genomes of different races of A. candida with the goal of identifying the cognate recognised effectors for the Col-0 alleles of WRR4A and WRR4B. We screened multiple CCG secreted proteins, mainly from A. candida races 2 and 4, using Agrobacterium-mediated transient co-expression to identify pairwise combinations of effector and NLR variant that activate an HR. Twelve CCG candidates – eight recognised by WRR4A and four by WRR4B – show activation of HR when transiently co-expressed, and were validated for their recognition by a bombardment assay in Col-0 wild-type and mutants (wrr4a-6 or wrr4b) of A. thaliana. Several of these CCGs are absent and some show expression or allelic polymorphism in the Col-0 virulent isolate AcEx1. For WRR4A-recognised CCGs, the N-terminal 100 amino acids are sufficient for recognition. Our data reveal that two distinct WRR4 paralogues confer broad-spectrum resistance by recognition of multiple CCG effectors across distinct clades in the A. candida CCG effectorome. Moreover, some of the CCGs confer enhanced susceptibility to another oomycete pathogen Hyaloperonospora arabidopsidis (Hpa), consistent with the idea that CCG secreted proteins are authentic A. candida effectors.