Fig 1.
Characteristics of the PUF family.
(A) Maximum-likelihood phylogenetic tree for PUF proteins in oomycetes, fungi, Arabidopsis thaliana, and humans. (B) Domain architectures of four P. ultimum PUF proteins predicted using SMART analysis based on their amino acid sequences (shown in yellow). The missing Pumilio repeat (highlighted in green) was computationally modeled to complete the characteristic structural framework depicted in (C) Ribbon diagrams showing the structural features of the PUF proteins PuPuf1 (classical), PuPuf2 (classical), PuPuf3 (C-shaped), and PuPuf4 (L-shaped). http://smart.embl-heidelberg.de/. (D) RNA-seq analysis was performed to quantify the transcript levels of Puf genes during the infection of soybean hypocotyls at 3, 6, 12, 24, and 36 hours post-inoculation (hpi). MY represents the mycelium control, while IF3, IF6, IF12, IF24, and IF36 denote the samples collected at 3, 6, 12, 24, and 36 hpi, respectively.
Fig 2.
ΔPuPuf4 mutants are defective in virulence and growth.
(A) Infection lesions on soybean leaves at 48 hpi and soybean hypocotyls at 24 hpi. (B), (C) Relative P. ultimum biomass detected through qRT-PCR at 48 h after leaf infection (B) and 24 h after hypocotyl infection (C). (D) Growth rates on V8 medium. (E) Relative inhibition rate on V8 medium containing 0.5M sorbitol, 0.4M NaCl, and 6.5 mM H2O2. (F) Oospore number after culturing for 2 d, 5 d, and 14 d. Asterisks indicate significant differences relative to the WT at P < 0.05 (*) or P < 0.01 (**).
Fig 3.
PuPuf4 is involved in 45S pre-rRNA processing.
(A) Subcellular localization of PuPuf4-GFP. DAPI (4′,6-diamidino-2-phenylindole) staining was performed by adding DAPI to the cultures 5 min prior to microscopic analysis. DIC: differential interference contrast; Merge: overlay of DIC, GFP fluorescence and DAPI staining. Bar, 10 μm. (B) Flow chart of RIP-seq. P. ultimum culture is lysed, incubated with anti-GFP beads, and then enriched with the protein–RNA mixture using a magnetic rack. Complexes are removed from the beads and then RNA is eluted and prepared for sequencing and qPCR. (C) Integrative Genomics Viewer results showing PuPuf4 binding peaks on rRNA precursors. The control represents the RNA segment where GFP binds. In comparison to GFP, PuPuf exhibits binding peaks in specific regions, including the 5′ETS, ITS1, 5.8S, ITS2, and 25S regions. (D) qRT-PCR analysis of RNA coimmunoprecipitated with PuPuf4-GFP protein with anti-GFP antibodies. Enrichment of pre-rRNA processing intermediates containing specific regions (regions 1 to 6) through co-IP was calculated using values obtained from the PuPuf4-GFP sample compared to those from the GFP sample. Error bars represent standard deviation (n = 3). Asterisks indicate statistically significant differences between values obtained with GFP (control) and the WT expressing PuPuf4-GFP based on Student’s t-test (**P < 0.01). (E) Relative levels of processing intermediates in the WT (wild type), ΔPuPuf4 (knockout mutant), and ΔPuPuf4-C (complemented line). Asterisks indicate statistically significant differences between the mutants and WT based on Student’s t-test (**P < 0.01).
Fig 4.
(A) Electrostatic surface representation of the structure of PuPuf4. Three distinct basic patches are indicated with dotted ovals. (B) EMSA results showing that PuPuf4 bound H68 rRNA. (C) MST results showing that PuPuf4 bound H68 rRNA (Kd = 0.24 μM, green curve), while mutation of R111, F147, and K368 in PuPuf4 attenuated its binding to H68 rRNA (Kd = 0.62 μM, red curve). (D) Superposition of PuPuf4, APUM24, and ScPuf6. K368 is located on patch 2, and R111 and F147 are located on patch 1. (E) Partial sequence alignment of non-canonical PUF proteins including PuPuf4, PsPuf4, APUM24, and ScPuf6. The residues R111, F147, and K368, indicated by asterisks, were replaced with A. (F) PuPuf4R111A, F147A, K368A nearly abolished detectable binding to H68 rRNA but formed a weak RNA–protein complex at high levels of PuPuf4R111A, F147A, K368A.
Fig 5.
Genes associated with ribosome biogenesis are bound and regulated by PuPuf4.
(A) Sequence of the enriched motif among the peak sequences of PuPuf4 target genes. (B) In vitro binding of PuPuf4 to three motifs in the EMSA assay. (C) EMSA assay showing that non-canonical PUF proteins including ScPuf6 and APUM24 bound AGAAGAA. (D) Principal component analysis of transcriptome data and overview of DEGs in the PuPuf4 knockout compared to WT. (E) GO enrichment results of DEGs between WT and ΔPuPuf4. (F) Distribution of PuPuf4-bound peaks within protein-coding gene bodies divided into 5′-untranslated regions (UTRs), coding sequences (CDSs), 3′-UTRs, and intronic regions. Venn diagram showing the overlap of DEGs and motif-related genes with GO enrichment results of the overlapping genes.
Fig 6.
Application of exogenous Puf4-dsRNAs impairs the virulence of Pythium aphanidermatum and Phytophthora sojae.
(A), (E) Observation of dsRNA uptake efficiencies in mycelium and zoospores of P. aphanidermatum and Ph. sojae. Cy3-labeled dsRNA (500 bp, 150 ng μL−1) was added and micrococcal nuclease treatment was performed 30 min before image acquisition using a Leica SP8 confocal microscope with excitation at 555 nm. Scanning was performed with the filter set to 570 nm. Cy3 signals were detected inside the target cells. Pictures were taken at 12 h post-treatment. Bar, 10 μm. (B) Fluorescence intensity in mycelium and zoospores treated with dsRNA was quantified using ImageJ software. Four images of each treatment were analyzed. Asterisks (**) indicate statistical significance compared with mycelium (P < 0.01). (C) Lesions on cucumber slices or blocks inoculated with P. aphanidermatum zoospores. The lesion areas were measured with ImageJ. Asterisks indicate significant differences compared with GFP-dsRNA at P < 0.01 (**). (D) Etiolated soybean seedlings inoculated with Ph. sojae zoospore suspension. Pictures were taken at 48 hpi. Data were subjected to statistical analysis using two-tailed t-test. Asterisks indicate significant differences relative to H2O at P < 0.05 (*) or P < 0.01 (**). (F) The level of GFP protein was dramatically reduced in Ph. sojae treated with corresponding GFP-dsRNA. Etiolated soybean seedlings were inoculated with Ph. sojae (overexpressing GFP) following treatment with GFP-dsRNA. The relative protein level was detected through western blotting of total protein extracted from the pathosystem. After quantification using ImageJ, relative band intensity was calculated. Ponceau staining indicated that total protein contents were consistent among samples from different treatments.
Fig 7.
A proposed model for Puf4 function.
PuPuf4 regulates the ribosome biogenesis by binding to the H68 component of 25S rRNA and AG-rich mRNA, influencing the pathogenicity and growth of the P. ultimum. Utilizing RNA interference technology, double-stranded RNA targeting PsPuf4 inhibited the translation of the PsPuf4, significantly reducing the pathogenicity of Ph. sojae.