Fig 1.
RBF1 expression is activated when Magnaporthe oryzae invades living plant cells.
(A) Quantitative RT-PCR analysis of RBF1 and PWL2 expression in conidia and inoculated rice leaf blades. The vertical axis indicates the amount of transcripts relative to that from the M. oryzae actin gene (MoACT1). Data are represented as mean values ± standard error (SE) (n = 3 plants). dpi, days post inoculation. (B) The dynamic expression of RBF1. RBF1 expression during the infection process in the rice leaf sheath epidermis was monitored by a long-term time-lapse imaging method using a fungal line transformed with RBF1p::GFP. After appressoria formation, GFP signals were captured at 20-min intervals. The z-series of optical sections corresponding to the outer half of the inner epidermal cells were stacked to generate maximum-intensity projection images. Images displayed were selected from S1 Movie. White and blue arrows indicate the induction of GFP expression in the appressorium and the invasive hyphae, respectively. Red arrows indicate the re-induction of the GFP expression in the hyphal cell that was about to invade the neighboring cell. hpi, hours post inoculation. Bar = 20 μm. (C) Confocal images of M. oryzae transformants introduced with RBF1p::GFP, PWL2p::GFP, or ChBD8p::GFP growing in living rice leaf sheaths (upper) and dead rice leaf sheaths (lower) at 24 hpi. GFP images were merged with differential interference contrast images. Asterisks indicate appressoria. Bar = 20 μm. (D) Quantitative RT-PCR analysis of RBF1 expression in the inoculated living leaf blades of rice and wheat, and dead rice leaf blades at 24 hpi. The vertical axis indicates the amount of transcripts relative to that from the M. oryzae actin gene (MoACT1). Data are represented as mean values ± SE (n = 4 plants).
Fig 2.
Rbf1 accumulates in the BIC and is translocated into rice cells.
(A) Co-localization of Rbf1:mCherry with Pwl2:GFP at the BIC. Rice leaf sheaths were inoculated with M. oryzae transformed with RBF1p::RBF1:mCherry and PWL2p::PWL2:GFP, and observed using a confocal microscope at 36 hpi. DIC, differential interference contrast image. (B) Accumulation of Rbf1:mCherry in the rice cytoplasm. Rice leaf sheaths were inoculated with M. oryzae transformed with RBF1p::RBF1:mCherry and IH at 36 hpi were observed after sucrose-induced plasmolysis. Confocal mCherry images were merged with DIC images. Data obtained using a transformant with PWL2p::PWL2:mCherry is shown as the control. (C) Accumulation of the Rbf1:mCherry fused with a nuclear localization signal (NLS) in the host nucleus. Rice leaf sheaths infected by the transformant containing RBF1p::RBF1:mCherry:NLS (upper) were observed using a confocal microscope at 24 hpi. Arrows indicate rice nuclei with mCherry signals. The transformant containing PWL2p::PWL2:mCherry:NLS (lower) is shown as the control. Asterisks indicate appressoria. Bar = 10 μm.
Fig 3.
RBF1 is a virulence determinant in Magnaporthe oryzae.
(A) Representative symptoms on the 6th leaf blades at 5 days after inoculation. Rice plants were sprayed with a conidial suspension of the wild-type strain (WT), an RBF1-knockout line (Δrbf1-1; KO), and a gene complementation line (KO+RBF1). Bar = 5 mm. (B) Evaluation of lesion types in leaf blades. Lesions formed at 5 days after spray-inoculation were counted according to the classifications displayed. Data are represented as the mean percentages ± SE (n = 5 plants). (C) Comparison of the development of invasive hyphae in rice leaf sheaths between WT and Δrbf1-1 (KO). Infection levels in leaf sheaths were assessed for each appressorium under a microscope and categorized as no invasion (S0), short invasive hyphae in one cell (S1), highly-branched invasive hyphae in one cell (S2), and multiple cell invasion (S3). To illustrate each category, typical images using a WT line transformed with TEFp::mCherry are displayed. Data are represented as the mean percentages ± SE [n = 14 plants (24 h post inoculation; hpi) and 23 plants (42 hpi)]. Student’s t-test was performed on arcsine-transformed data between WT and KO (*, P < 0.05; **, P < 0.005). The total numbers of appressoria observed per line were ~ 1,500 (24 hpi) and 3,000 (48 hpi). (D) Confocal images of rice epidermal cells in inoculated leaf blades. Transgenic rice plants constitutively expressing GFP under the CaMV 35S promoter were inoculated with the WT (left) or Δrbf1-2 line (KO; right) transformed with TEFp::mCherry. GFP and mCherry signals were merged. Note that the disappearance of the GFP signal (green) indicates host cell death. Arrows indicate invasive hyphae. Ap, appressorium. Bar = 20 μm. (E) Transverse sections of inoculated rice leaf blades at 6 dpi. To visualize invasive hyphae, the WT (upper) and Δrbf1-1 (KO; lower) were transformed with TEFp::GUS. Spot-inoculated rice leaf blades were stained for β-glucuronidase activity, hand-sectioned, and observed by light microscopy. Bar = 0.1 mm.
Fig 4.
RBF1 is required to suppress the activation of host immune responses.
(A) Enhanced activation of rice defense-related genes by KO. Rice leaf blades were spotted with a conidial suspension of the WT fungus or Δrbf1-1 (KO). RNA was extracted from the inoculated leaves at 2 dpi and subjected to qRT-PCR analysis. The vertical axis indicates the amount of transcripts relative to that from rice eEF-1α (OsEF1). Bars represent the mean values ± SE (n = 4 plants). PR, pathogenesis-related genes; Mock, spotted with water. (B) Enhanced accumulation of rice diterpenoid phytoalexins by KO. Momilactones and phytocassanes in inoculated leaf blades were quantified using an HPLC-MS/MS spectrometer. Data from five to seven independent extracts of two inoculation assays are represented as mean values ± SE. Asterisks indicate significant differences compared with the WT data (Student’s t-test, P < 0.05).
Fig 5.
The RBF1-disrupted blast fungus infects rice plants with impaired immunity.
(A) Symptoms on the spot-inoculated rice leaf blades at 5 dpi (upper) and the GUS staining (lower). Both the WT and Δrbf1-1 (KO) were transformed with TEFp::GUS to visualize invasive hyphae. NT, non-transgenic rice; +ABA, inoculated with 30 μM abscisic acid; NahG, transgenic rice expressing the salicylate hydroxylase gene. Bar = 5 mm. (B) Proliferation of M. oryzae in leaf blades at 6 dpi evaluated by quantitative PCR. DNA amount of M. oryzae 28SrDNA (Mo28S) relative to rice eEF-1α (OsEF1) in spot-inoculated leaf fragments were measured. Data are represented as mean values ± SE (n = 7 plants for NT and +ABA, and n = 5 plants for NahG samples). Different letters above bars indicate significant differences at P < 0.05 (Student’s t-test of paired comparison).
Fig 6.
Rbf1 is required for the focal BIC formation and normal hyphal development.
Confocal images of rice leaf sheath cells infected by the WT or KO line harboring PWL2p::PWL2:mCherry at 24 hpi (A) and 36 hpi. (B). Note that the coding region of RBF1 in the genome was replaced with GFP in the Δrbf1-1 used (KO), thus the KO-based transformants express free GFP (green) driven by the RBF1 promoter. Bar = 10 μm. (C) Confocal images of the extra-invasive hyphal membranes (EIHM) and a BIC-localizing effector protein at 30 hpi. Rice leaf sheaths transformed with 35S::GFP:LTI6b were inoculated with the WT or KO line harboring PWL2p::PWL2:mCherry. Arrow indicates the aggregation of EIHM at the BIC position. Note that the KO-invaded rice cell shows the broad distribution of the BIC marker effector around the IH and no accumulation of the GFP signals at the mCherry signals. Bar = 10 μm. (D) Comparison of invasive hyphal shape. Rice leaf sheaths were inoculated with the WT or Δrbf1-2 (KO) line harboring both PWL2p::PWL2:GFP and BAS4p::BAS4:mCherry and observed using a confocal microscope at 30 hpi. The z-series of optical sections were stacked to generate maximum-intensity projection images. Confocal images of the representative infection sites are shown with illustrations indicating the hyphal parts measured. Red, blue, and black arrows indicate the length and width of the primary IH (PIH), and the width of the BIC-associated cell (BAC), respectively. Arrowhead indicates the BIC. Bar = 20 μm. Data of the IH sizes measured using ImageJ (http://imagej.nih.gov/ij) are represented as mean values ± standard deviation (n = 57 infection sites). Asterisks above bars indicate significant differences compared with the WT data (Student’s t-test, P < 0.01). DIC, differential interference contrast image; mC, mCherry image; G, GFP image. Asterisks, appressoria.
Fig 7.
Focal BIC formation correlates with the virulence of Magnaporthe oryzae.
(A) Symptoms on the spot-inoculated rice leaf blades at 6 dpi. The Δrbf1-1 (KO) was further transformed with RBF1p:RBF1:mCherry or RBF1p::RBF1Δ20:mCherry. Rbf1Δ20:mCherry has a 20 amino acid-deletion (corresponding to Pro320-Gly339). Bar = 5 mm. (B) Proliferation of M. oryzae in leaf blades at 6 dpi evaluated by quantitative PCR. DNA amount of M. oryzae 28SrDNA (Mo28S) relative to rice eEF-1α (OsEF1) in spot-inoculated leaf fragments were measured. Data are represented as mean values ± SE (n = 4 plants). Confocal images of leaf sheath cells invaded by the Δrbf1-1 lines (KO) transformed with RBF1p::RBF1:mCherry. (C) or RBF1p::RBF1Δ20:mCherry. (D) Both transformants express GFP (green) owing to the replacement of the coding region of the endogenous RBF1 with GFP in Δrbf1-1. Arrows indicate the focal localization of Rbf1:mCherry in the BIC. Bar = 10 μm. (E) Confocal image of a leaf sheath cell invaded by the WT line transformed with RBF1p::RBF1Δ20:mCherry at 36 hpi. Bar = 10 μm.
Fig 8.
Host immunity does not affect the dispersed BIC formation in the RBF1-disrupted fungus.
Rice leaf sheaths were inoculated with the WT or Δrbf1-2 line transformed with both PWL2p::PWL2:GFP and BAS4p::BAS4:mCherry and observed using a confocal microscope at 30 hpi. The z-series of optical sections were stacked to generate maximum-intensity projection images. NT, non-transgenic rice; +ABA, inoculated with 30 μM abscisic acid; NahG, transgenic rice expressing the salicylate hydroxylase gene. Bar = 10 μm.
Fig 9.
Translocation levels of a symplastic effector are impaired by the RBF1 disruption.
(A) Confocal images of rice leaf sheath cells invaded by the Δrbf1-1 (KO) lines containing PWL2p::PWL2:mCherry (left) or PWL2p::PWL2:mCherry:NLS (right) at 24 hpi. Arrowheads and arrows indicate host cytoplasm and nuclei with mCherry signals (red), respectively. Asterisks, appressoria. Bar = 20 μm. (B) Categories of the mCherry signal pattern. L0, no mCherry signals in the host nucleus; L1, mCherry signals only in the nucleus of the invaded cell; L2, mCherry signals in the first invaded cell and the neighboring uninvaded cells. (C) A lack of RBF1 causes a reduction in the spread of Pwl2. Rice leaf sheaths inoculated with the WT or KO line containing PWL2p::PWL2:mCherry:NLS were observed at 24 hpi, and mCherry signal patterns were classified into the three categories illustrated in (B). Data are represented as the mean percentages ± SE [n = 4 tests (WT) and 5 tests (KO)]. Asterisks above the bars indicate significant differences compared with the data of the WT line (P < 0.05, Student’s t-test on arcsine-transformed data).
Fig 10.
Summary of phenotypes shown in the wild-type Magnaporthe oryzae and the RBF1-disrupted mutant.
The biotrophic interfacial complex (BIC) is a specialized region of the EIHMx focally formed at the tip of the tubular invasive hypha that differentiates into the bulbous pseudohyphae. The BIC comprises the aggregated EIHM in which symplastic effectors detected as a cluster of puncta and the BIC base in which apoplastic effectors also preferentially accumulate. From the characterization of the knockout mutants (Δrbf1), it is deduced that Rbf1 plays a crucial role in the development of the focal BIC structure and the hyphal differentiation, which is required to lower the activation of host immune response, thus allowing the establishment of the biotrophic invasion. Ap, appressorium; CW, host cell wall; IH, invasive hyphae; PM, host plasma membrane.