RXLR Effector AVR2 Up-Regulates a Brassinosteroid-Responsive bHLH Transcription Factor to Suppress Immunity

Abstract

An emerging area in plant research focuses on antagonism between regulatory systems governing growth and immunity. Such cross talk represents a point of vulnerability for pathogens to exploit. AVR2, an RXLR effector secreted by the potato blight pathogen Phytophthora infestans, interacts with potato BSL1, a putative phosphatase implicated in growth-promoting brassinosteroid (BR) hormone signaling. Transgenic potato (Solanum tuberosum) plants expressing the effector exhibit transcriptional and phenotypic hallmarks of overactive BR signaling and show enhanced susceptibility to P. infestans Microarray analysis was used to identify a set of BR-responsive marker genes in potato, all of which are constitutively expressed to BR-induced levels in AVR2 transgenic lines. One of these genes was a bHLH transcription factor, designated StCHL1, homologous to AtCIB1 and AtHBI1, which are known to facilitate antagonism between BR and immune responses. Transient expression of either AVR2 or CHL1 enhanced leaf colonization by P. infestans and compromised immune cell death activated by perception of the elicitin Infestin1 (INF1). Knockdown of CHL1 transcript using Virus-Induced Gene Silencing (VIGS) reduced colonization of P. infestans on Nicotiana benthamiana Moreover, the ability of AVR2 to suppress INF1-triggered cell death was attenuated in NbCHL1-silenced plants, indicating that NbCHL1 was important for this effector activity. Thus, AVR2 exploits cross talk between BR signaling and innate immunity in Solanum species, representing a novel, indirect mode of innate immune suppression by a filamentous pathogen effector.

Figure 1.

Transgenic potato cv Desiree lines expressing 35S:AVR2 show morphological hallmarks of an overactive BR pathway. A, Growth morphology of 35S:AVR2 plants (#29 and #39) showing twisted stems and curled leaves, compared to untransformed potato cv Desiree (wild type [WT]). B, Leaf phenotype of 35S:AVR2 plants showing reduced compound formation and loss of symmetry. C, Reduced percentage of stomata in 35S:AVR2 potato plants. Stomata count was expressed as percent of total epidermal cells counted per 500 µm. Results combine three biological replicates, each consisting of epidermal leaf prints from three or more plants. Error bars indicate SEM; letters denote significant difference (P < 0.001 in one-way ANOVA, Holm-Sidak). D, Confocal microscopy showing reduced stomatal frequency in 35S:AVR2 potato and enlargement of stomata relative to wild-type plants. Images are of representative leaves stained with calcofluor white. Scale bar = 100 μm.

Figure 2.

Microarray analysis of BR (EBL treatment) response in potato cv Desiree. A, Microarray validation by quantitiative real-time PCR (qRT-PCR) of two independent biological replicates plotted in one graph. Fold-change from microarray data plotted against fold-change from qRT-PCR for five selected BR marker genes examined at 24 hpt. Fold-change log2 transformed to allow symmetry of up- and down-regulation. Linear regression was used to determine a coefficient of determination (R²). B, Table of selected marker genes showing significant differential expression with BR treatment. Fold-change values are shown from the microarray data, qRT-PCR validation, and an independent biological replicate. C, Relative expression of BR-regulated genes in untreated potato cv Desiree (wild type [WT]; given a value of 1), wild type at 24 h after treatment with EBL, and constitutive levels of expression in 35S:AVR2 potato plants, assessed by qRT-PCR. Expression was normalized to StUbi and shown relative to wild-type untreated plants. Graph shows the average of three technical replicates ± sd, with similar trend observed in two independent biological replicates.

Figure 3.

BR-responsive genes are suppressed by PTI. A, Treatment of potato cv Desiree with P. infestans culture filtrate (CF) results in transcript accumulation of PTI marker genes StWRKY7 and StACRE31 by 1 h after treatment (CF1h), but reduced transcript abundance of BR (EBL)-induced genes StCHL1, StEXP8, StSAUR67, StCAB50, and StP69F and of BR biosynthesis-associated genes StDWF4 and StSTDH. B, Treatment of potato cv Desiree with the bacterial PAMP flg22 results in the same opposing patterns of transcript abundance by 1 h after treatment (F1h) for PTI, BR, and BR biosynthesis markers as observed in A. Error bars represent SD across three technical replicates, with a similar trend observed in two independent biological replicates.

Figure 4.

AVR2 expression in potato results in increased susceptibility to P. infestans. A, Lesion size of transgenic P. infestans isolate 88069 expressing tdTomato (McLellan et al., 2013; diameter in mm) on 35S:AVR2 potato at 7 d postinoculation of sporangia suspension. Data shown combines two independent replicates, each comprising 10 or more leaves per plant line, taken from three or more individual plants, with two inoculations per leaf. Error bars represent SEM; letters denote significant difference (P < 0.001, one-way ANOVA with Holm-Sidak). B, Representative leaf images showing increased lesion size of transgenic tdTomato expressing P. infestans isolate 88069 (McLellan et al., 2013) on 35S:AVR2 potato compared to untransformed wild-type (WT) potato. Images are taken under UV light.

Figure 5.

AVR2 negatively regulates immunity to P. infestans and suppresses INF1 cell death. A, Average lesion size (diameter) of P. infestans 88069 colonization on N. benthamiana, inoculated 24 h after Agrobacterium-mediated transient expression of GFP-AVR2 or empty vector (EV) control. Results combine three biological reps, consisting of at least six plants each with six infiltrations per construct. Error bars show SEM; letters denote significant difference (P ≤ 0.001) using one-way ANOVA (Holm-Sidak). B, Representative leaf images showing increased P. infestans 88069 lesion size with GFP-AVR2 expression. Images were taken under UV light to show full extent of infection. C, Percentage of inoculation sites leading to cell death following coexpression of INF1 with AVR2, AVR3a (positive control) or an empty vector (negative control) in N. benthamiana. Error bars show SEM; letters denote significant difference (P ≤ 0.001) using one-way ANOVA (Holm-Sidak). Results combine at least three experimental replicates, consisting of four or more plants, with three or more infiltrations per plant, per combination. D, Representative leaf image showing suppression of INF1-triggered cell death when AVR2, or AVR3a are coexpressed

Figure 6.

StCHL1 suppresses immunity and increases susceptibility to P. infestans in N. benthamiana. A, Graph shows percentage of leaf infiltration sites at 5 dpi resulting in cell death following Agrobacterium-mediated coexpression of INF1 with either StCHL1 or an empty vector (EV) control. Error bars show SEM, a≠b (P ≤ 0.001) in one-way ANOVA (Holm-Sidak). Results are combined from four experimental replicates consisting of at least four plants, each, with at least six infiltrations per plant per expression combination. B, Representative leaf image showing suppression of INF1 cell death when CHL1 is coexpressed. C, StCHL1 or an empty vector control were transiently expressed in N. benthamiana. Sites were inoculated with P. infestans 88069 sporangia suspension 24 h later, with lesions measured (diameter in mm) at 7 dpi. Error bars show SEM; letters denote significant difference (P ≤ 0.001 in one-way ANOVA, Holm-Sidak). Results are combined from four experimental replicates. D, Representative leaf image showing increased P. infestans. colonization following StCHL1 expression in N. benthamiana. E, PTI marker gene (NbWRKY7, NbACRE31) expression in N. benthamiana 1 h after P. infestans CF treatment, relative to untreated plants (which were given a value of 1) by qRT-PCR. Treatment occurred 2 d after Agrobacterium-mediated transient expression of AVR2, StCHL1, or an empty vector control. Expression levels were normalized to NbEF1a. Data represents the average of three technical replicates, each from two experimental replicates combined, ±SD.

Figure 7.

Silencing of NbCHL1 in N. benthamiana compromises P. infestans infection and perturbs the ability of AVR2 to suppress INF1 cell death. A, Silencing of NbCHL1 using two independent VIGS constructs (TRV:NbCHL1 5′ and TRV:NbCHL1 3′) significantly reduced (one-way ANOVA, P < 0.001, n = 136) sporulation of pathogen on NbCHL1-silenced N. benthamiana compared to TRV:GFP control. B, P. infestans 88069 lesion diameter also showed a significant reduction (one-way ANOVA, P ≤ 0.05, n = 223) compared to TRV:GFP control. C, The graph shows the percentage of inoculation sites leading to cell death following coexpression of INF1 with AVR2, AVR3a (positive control), or an empty vector (negative control) in CHL1-silenced N. benthamiana and GFP control. Knockdown of NbCHL1 transcript compromised the ability of AVR2 to attenuate INF1 HR. A significant increase is seen in INF1 HR in TRV:NbCHL1 5′ and TRV:NbCHL1 3′ plants (one-way ANOVA, P ≤ 0.001, using 41 biological replicates and at least 6 inoculations per replicate), compared with the TRV:GFP control at 5 dpi. No significant change is seen in AVR3a suppression of INF1 cell death between the NbCHL1 silenced and GFP control plants. The significance is denoted by lowercase letters and Error bars shown represent SEM.

Figure 8.

Proposed model indicating how AVR2 tips the balance between growth and immunity to promote potato late blight disease. Perception of BR by the receptor BRI1 activates the BR signaling pathway, inducing CHL1 (black arrows), which is proposed to stimulate growth and development. Conversely, we show that CHL1 suppresses immunity triggered by perception of the oomycete PAMP INF1 by receptor ELR. Transgenic potato plants expressing AVR2 lead to activation of the BR signaling pathway and up-regulation of CHL1 to suppress immunity (red arrows). We propose that AVR2 activates BR signaling by stimulating BSL1 activity (red question mark).