Fig 1.
PsCRN108 is a P. sojae secreted effector that acts in the nucleus of plant cells.
(A) Yeast invertase secretion assay of the predicted signal peptide of PsCRN108. The predicted signal peptide sequences and the following two amino acids (1–23) of PsAvr1b was used as the positive control to assay the predicted signal peptide sequences and the following four amino acids (1–21) of PsCRN108. CMD-W (minus Trp) plates were used to select yeast strain YTK12 carrying the pSUC2 vector. YPRAA media was used to indicate invertase secretion. An enzymatic activity test based on reduction of TTC to red-colored formazan was used to confirm invertase secretion. (B) Western blot analysis of proteins in culture supernatants of the P. sojae transgenic lines. PsCRN108:Avr1bCt indicates the full length of PsCRN108 fused to the Avr1b C-terminal regions (Avr1bCt aa 66–138, lacking the predicted signal peptide and RxLR host translocation domain). A P. sojae line expressing Avr1b:mRFP was used as the positive control. The His tag was attached to the C-terminus of all the fusion proteins. The non-transformed P. sojae P6497 line was used as the antibody-specificity control. Proteins extracted from mycelia (M) and culture supernatants (S) were analyzed by Western blotting using anti-His or anti-Actin antibodies. The protein sizes are expressed in kDa. (C) Western blot analysis of proteins isolated from P. sojae transgenic lines during infection. Avr1bCt:mRFP indicates Avr1b C-terminal regions fused to mRFP and is used as a control. The infected hypocotyl tissues (HARO13) including mycelia of P. sojae transgenic lines were collected at 12 hpi for protein extraction, and then Western blotting was performed using anti-His antibodies. The protein sizes are expressed in kDa. PS, Ponceau S staining. (D) Hypocotyl-inoculation assay using two soybean cultivars. Two soybean cultivars, HARO13 (Rps1b) and Williams (rps), were inoculated with the indicated P. sojae transgenic lines. Photographs were taken 2 dpi. Data (e.g. 2/83) indicate the number (2) of dead plants out of the total number (83) of plants tested. (E) Western blot analysis of GFP fusion proteins transiently expressed in N. benthamiana. The indicated proteins were extracted from leaves at 48 h after agoinfiltration and detected using anti-GFP antibodies. GFP:PsCRN108-NLSm and GFP:PsCRN108-HhHm are described in Fig 3. Dotted line indicates lanes not adjacent on the gel. The protein sizes are expressed in kDa. PS, Ponceau S staining (F) Nuclear localization of PsCRN108 in N. benthamiana. Epidermal cells of N. benthamiana leaves transiently expressing GFP alone or GFP:PsCRN108 were observed using confocal microscopy at 2 days post infiltration. DAPI staining was used to visualize the nuclei. Scale bars = 50 μm (upper panel) and 10 μm (lower panel).
Fig 2.
PsCRN108 contributes to the virulence of P. sojae.
In all panels, WT = wild type P6497; T17 is a non-silenced transformant; ST18 and ST19 are PsCRN108-silenced transformants. (A) Relative transcript levels of PsCRN108 in different transgenic P. sojae lines. The transcript levels of the PsCRN108 and PsCRN112a/b/c genes were measured by qRT-PCR and normalized to those in the WT using the actin gene as an internal reference. Bars represent standard errors from three independent biological replicates (**, P<0.01 compared with the WT; Dunnett's test). (B) Lengths of lesions on etiolated soybean hypocotyls inoculated with P. sojae. A susceptible soybean cultivar (Williams) was inoculated with ~100 zoospores of each P. sojae line. Lesion lengths were measured at 36 hpi in three independent biological replicates, each of which comprised at least nine plants (**, P<0.01 compared with WT; Dunnett's test). (C) Phenotypes of lesions on etiolated soybean hypocotyls. Photographs representative of three independent experiments were taken at 36 hpi. (D/E) Callose deposition in etiolated soybean hypocotyl epidermal cells infected with P. sojae. Representative micrographs (D) of etiolated soybean hypocotyls that were inoculated with P. sojae zoospores or water (Mock) as indicated and then stained by aniline blue at 10 hpi. Bar = 50 μm. Number of callose deposits (E) per microscopic field was quantitated using ImageJ software. The data are mean ± SEM of numbers of callose deposits per microscopic field in three independent biological replicates, each of which comprised at least three soybean hypocotyls (**, P<0.01 compared with WT; Dunnett's test).
Fig 3.
PsCRN108 suppresses defense responses in N. benthamiana and A. thaliana and promotes susceptibility.
(A) Representative photographs of N. benthamiana leaves inoculated with P. capsici. The leaves were infiltrated with A. tumefaciens carrying genes encoding GFP (as a control), PsCRN108, or the indicated mutants. PsCRN108-NLSm, the substituted mutation of the predicted NLS (PsCRN108KKRR120-123AAAA); PsCRN108-HhHm, the substituted mutation of the predicted HhH motif (PsCRN108L766A;G768A;V769A). P. capsici zoospores (~100) were inoculated into the infiltrated area 48 hours after infiltration, and photographs were taken at 36 hpi. Smaller circles indicate the inoculation sites and the dotted margins indicate the lesion areas. (B) Lesion areas of inoculated N. benthamiana leaves. Lesion areas (cm2) were measured at 24 and 36 hpi, and calculated from three independent biological replicates using at least six leaves each (**, P<0.01 compared with GFP; Dunnett's test). (C) Representative photographs of transgenic Arabidopsis leaves inoculated with P. capsici. Stable transgenic Arabidopsis leaves expressing GFP or PsCRN108 were inoculated with ~100 P. capsici zoospores, and photographs were taken at 36 hpi. Smaller circles indicate the inoculation sites and the dotted margins indicate the lesion areas. (D) Disease rating distributions of transgenic Arabidopsis plant leaves. Total 150 leaves inoculated with P. capsici as in (C) were graded on a scale 0–4 as described in the Methods. **, P<0.01 for the comparison of PsCRN108 to GFP; Wilcoxon rank sum test. (E-H) Callose deposition in P. capsici-inoculated leaves. P. capsici-inoculated leaves of N. benthamiana as in (A) and of transgenic A. thaliana as in (C) were stained with aniline blue at 10 hpi to visualize callose. (E, G) Representative images of callose deposition in N. benthamiana (E) and Arabidopsis (G) tissues. Scale bars = 50 μm. (F, H) The numbers of callose deposits were quantified using ImageJ software. The data indicate the mean ± SEM numbers of callose deposits per microscopic field in three independent biological replicates, each of which comprised at least three leaves and four fields were counted per leaf (**, P<0.01 compared with GFP, Dunnett's test in N. benthamiana (F); **, P<0.01, t test in A. thaliana (H)). (I) Confocal imaging of non-inoculated N. benthamiana epidermal cells expressing each of the four proteins. Photographs were taken 48 hours post-infiltration. Scale bars = 10 μm.
Table 1.
Transcriptional levels of HSP genes in transgenic Arabidopsis.
Fig 4.
PsCRN108 interferes with plant HSP gene expression.
(A) Relative transcript levels of AtHSP90.1 in transgenic Arabidopsis leaves. The transcript levels of AtHSP90.1 (AT5G52640) in stable transgenic Arabidopsis leaves expressing GFP or PsCRN108 were measured by qRT-PCR using RNA samples extracted from P. capsici inoculated leaves (10 hpi) and mock-treated leaves. The relative transcript levels during infection were normalized to those of mock-treated leaves expressing GFP using Actin2 as an internal reference. Values are means ± SEM of three independent biological replicates (**, P<0.01; t-test). (B) Relative transcript levels of NbHsp90-1. The transcript levels of NbHsp90-1 in N. benthamiana leaves expressing GFP, PsCRN108 or the indicated PsCRN108 variants were measured by qRT-PCR using RNA samples extracted from P. capsici inoculated leaves (10 hpi) and mock-treated leaves. The relative transcript levels during infection were normalized to those of mock-treated leaves expressing GFP using NbEF1α gene as an internal reference. Values are means ± SEM of three independent biological replicates (**, P<0.01 compared with GFP; Dunnett's test). (C) Relative transcript levels of GmHsp90A1. Hypocotyls of etiolated soybean cultivar (Williams) were challenged with zoospores of the indicated P. sojae lines, and RNA was extracted at 10 hpi. The transcript levels of GmHsp90A1 in soybean inoculated with transgenic P. sojae lines were normalized to levels of those of mock-treated soybean using ACT20 gene as an internal reference. Values are the means ± SEM of three independent biological replicates (**, P<0.01; *, P<0.05 compared with the WT; Dunnett's test). (D) Schematic of the constructs used in the GUS activity assay. A -2000 bp promoter region of AtHSP90.1 were fused to GUS. “ATG” is the translational start site for the GUS gene. The 35S min promoter fused to GUS (p35S min:GUS) was used for normalization of the relative GUS activity. (E) Relative GUS enzyme activity in N. benthamiana. The constructs of pAtHSP90.1:GUS and p35S min:GUS were transiently introduced into N. benthamiana leaves together with constructs encoding GFP, PsCRN108 or the indicated PsCRN108 variants. The leaves were immersed in a solution containing P. capsici zoospores 48 h post infiltration, and the protein was extracted at 3 hpi. GUS activity levels were measured using a fluorometric assay. Each column represents the ratio from the pAtHSP90.1:GUS construct relative to that from the p35S min:GUS construct. Values are the mean ± SEM of three independent biological replicates with three technological replicates each (**, P<0.01 compared with GFP; Dunnett's test).
Fig 5.
PsCRN108 interacts with Arabidopsis HSP gene promoters and HSE’s.
(A) Enrichment of AtHSP gene promoters as determined by ChIP assays. Transgenic Arabidopsis plants expressing GFP or GFP:PsCRN108 infected with P. capsici at 10 hpi (I) or without (N) infection were used for ChIP assays. Specific primers were designed for four fragments (S1-4 are described in S7 Fig) in the five selected Arabidopsis HSP promoters (AtHSP20, AT1G53540; AtHSP70, AT3G12580; AtHSP90, AT5G52640; AtHSP101, AT1G74310; and AT5G56010, another HSP90 gene used as a negative control). qPCR analysis was performed on immunoprecipitated DNA using anti-GFP antibodies. Values for the ChIP samples were firstly normalized to the input control and then divided by the no-antibody control to obtain the fold enrichment values. Values are the means ± SEM of three independent biological replicates (**, P<0.01; *, P<0.05 compared with GFP without infection; Dunnett’s test). (B) Yeast one-hybrid (Y1H) assay of PsCRN108 with the AtHSP90.1 promoter region and HSE. A 500-bp (-1 to -500) fragment of the AtHSP90.1 promoter, a 19 bp HSE (CCAGAAGCTTCCAGAAGCC), or a 19 bp HSE mutant (HSEm, CCAtAAGCTTaCAtAAGCC) were integrated into the genome of yeast upstream of the AUR1-C gene to produce yeast bait strains. The lowest concentrations of Aureobasidin A (AbA) that limited the growth of yeast bait strains were determined before transformation with the pGAD plasmid (AtHSP90.1–500, 200 ng ml-1; HSE, 500 ng ml-1; HSEm, 500 ng ml-1) and were used to assess the pGAD transformants. Yeast growth on selective medium (-Leu +AbA) was recorded on day 3 as an indicator of protein—DNA interactions. pGAD:AtHsfA1a served as the positive control and pGAD as the negative control.
Fig 6.
PsCRN108–HSE interaction inhibits binding to HSE of plant AtHsfA1a.
(A) Schematic of the constructs used in the GUS activity assay. The HSE and HSEm were fused to the 35S minimal promoter and GUS. “ATG” is the translational start site for the GUS gene. The 35S minimal promoter fused to GUS was used for normalization of the relative GUS activity. (B) HSE-driven GUS activity in N. benthamiana was suppressed by PsCRN108. Leaves transiently expressing PsCRN108 or PsCRN108 mutants together with the GUS constructs were challenged with P. capsici zoospores for 3 h. Then, GUS activities were measured by fluorometric assays. The data in each column are fold changes normalized to the control (p35S min:GUS). Bars indicate standard errors from three biological replicates (*, P<0.05 compared with GFP; Dunnett’s test). (C) Pull-down of PsCRN108 by HSE. Biotinylated HSE (100 fmol) was incubated with 150 μg indicated proteins in the presence of streptavidin agarose. Upper panel, Western blot analysis of the input proteins using an anti-MBP antibody. Middle panel, Western blot detection of un-bound proteins before elution. Lower panel, proteins pulled-down by HSE. MBP and PsCRN108-HhHm were used as negative controls and AtHsfA1a as the positive control. (D) Inhibition of binding between AtHsfA1a and HSE by PsCRN108 as determined by DNA pull-down assay. Biotinylated HSE (100 fmol) was first incubated with 150 μg GST:PsCRN108 proteins (first panel) in the presence of streptavidin agarose, and then 150 μg MBP (negative control) or 150 μg MBP:AtHsfA1a (second panel) proteins were added after washing of the unbound protein. PsCRN108 (third panel), but not AtHsfA1a (fourth panel), was found to bind to HSE by Western blotting. (E) Western blot analysis of GFP- and HA- fusion proteins transiently expressed in N. benthamiana 48 h after infiltration. The protein sizes are expressed in kDa. PS, Ponceau S staining. (F) Inhibition of binding between AtHsfA1a and HSE by PsCRN108 as determined by in vivo ChIP assay. N. benthamiana leaves expressing the constructs as indicated were infected with P. capsici zoospores for 10 h, and then used for ChIP assays with anti-GFP antibodies. qPCR analysis was performed on immunoprecipitated DNA using primers specific for the NbHsp90-1 promoter (S7 Fig). Values for the ChIP samples were first normalized to the input control and then divided by the no-antibody control to obtain the fold enrichment values. Values are the means ± SEM of three independent biological replicates (**, P<0.01 compared with the indicated control; Dunnett’s test).
Fig 7.
The NbHsp90 gene family is required for P. capsici resistance in N. benthamiana.
(A) Relative transcript levels of three NbHsp90 genes. N. benthamiana leaves were infiltrated with Agrobacterium strains harboring the pTRV1 vector combined with pTRV2:NbHsp90 or pTRV2:GFP (as a negative control). Total RNA samples were extracted 2 weeks after infiltration and subjected to qRT-PCR analysis. Transcriptional levels of three genes, NbHsp90-1/2/3, were normalized to the levels in N. benthamiana infiltrated with TRV:GFP using the NbEF1α gene as an internal reference. Bars represent standard errors from three independent biological replicates (**, P<0.01; t-test). (B) Lesion areas of TRV-N. benthamiana leaves inoculated with P. capsici. ~100 P. capsici zoospores were inoculated into the indicated silenced leaves and lesion areas were measured at 36 hpi and 48 hpi. Values (cm2) are the means ± SEM of three independent biological replicates, each of which comprised five leaves (**, P<0.01, t-test). (C) Phenotypes of TRV-infected N. benthamiana leaves inoculated with P. capsici. Representative photographs were taken at 48 h post-inoculation with P. capsici zoospores. Smaller circles indicate the inoculation sites and the dotted margins indicate the lesion areas. (D) Model of the involvement of PsCRN108 in suppression of HSP gene transcription and plant defense. During exposure to heat or other stresses, including biotic stresses, free cytoplasmic AtHsfA1a may translocate to the nucleus and bind as a trimer to HSE (a conserved element in the promoter regions of HSP genes), resulting in activation of gene expression [28,39]. P. sojae secretes the PsCRN108 effector to bind to the same binding region as AtHsfA1a in an HhH-dependent manner. The binding of PsCRN108 to HSE inhibits the association of AtHsfA1a with HSE, resulting in the suppression of HSP gene transcription and inhibition of plant defense responses.