Fig 1.
(A) Schematic representation of the Pik-1 (Pikp-1, Pik*-1, Pikm-1) and RGA5 Nucleotide-binding Leucine Rich Repeat Receptors (NLRs) and small HMA (sHMA) proteins of rice. CC: coiled-coil domain; NB-ARC: nucleotide binding domain; LRR: leucine rich repeat; PRD: proline-rich domain. Amino acid sequence alignment of a subset of HMA proteins of rice. The black bar highlights the putative metal-binding motif MxCxxC. Interaction with AVR-PikD is indicated for yeast two-hybrid (Y2H: red dot: binding; white dot: non-binding) and AlphaScreen (AlphaScreen Signal [AS]: strength of interaction signal as compared to that of AVR-PikD/OsHIPP19 interaction is given in the inset). (B) A maximum likelihood tree of the HMA domains of 87 sHMA of rice. Amino acid sequences of the HMA domains were aligned and used for reconstruction of the phylogenetic tree. The dots on the branches indicate bootstrap values after 1,000 replications. Clade A and Clade B are indicated by red and blue branches, respectively. Note Pi21 (OsHIPP05) belongs to Clade B. Y2H and AlphaScreen results (AS) are as shown as in (A). Bar graphs in purple color show expression level of each gene in leaves as revealed by RNA-seq (log2(TPM+1) value). Bar graphs in green color show the induction level of each gene in rice suspension cultured cells after chitin treatment (log2[TPM of chitin-treated cultured cells+1] + 0.01)/ (log2[TPM of mock-treated cultured cells+1] + 0.01). (C) Results of co-immunoprecipitation of AVR-PikD with OsHIPP19 and OsHIPP20 transiently expressed in Nicotiana benthamiana leaves. (D) Y2H interactions of the AVR-Pik variants, AVR-PikA, AVR-PikC, AVR-PikD, AVR-PikE, to OsHIPP19 and OsHIPP20.
Fig 2.
Three regions of sHMAs are predicted to have close contact with AVR-PikD.
(A) Amino acid sequence alignment of sHMAs (top) and a matrix of distance of each amino acid residue to AVR-PikD protein as predicted by ColabFold (bottom). Predicted atomic distance below 5Å are indicated by blue tiles. Three regions indicated by A, B and C are predicted to be in close contact with AVR-PikD. (B) Binding structures between AVR-PikF (green) and OsHIPP19 (light orange) (a: Crystal structure; Maidment et al. 2021) [29], predicted structure between AVR-PikD (light green) and OsHIPP20 (grey) (b: AlphaFold-multimer model), predicted structure between AVR-PikD (light green) and OsHIPP16_D68DELS80K (dark yellow) (c: AlphaFold-multimer model), and predicted structure between AVR-PikD (light green) and OsHPP02_D79V (yellow) (d: AlphaFold-multimer model). (C) Y2H interactions between the variants of OsHIPP16 (OsHIPP16, OsHIPP16_D68DEL, OsHIPP16_S80K, OsHIPP16_D68DEL_S80K) and AVR-PikD. (D) AlphaScreen interactions between the variants of OsHIPP16 (OsHIPP16, OsHIPP16_D68DEL, OsHIPP16_S80K, OsHIPP16_D68DEL_S80K) and AVR-PikD. The values are relative AlphaScreen signals (AS) to that of OsHIPP16/DHFR interaction signal (negative control). The error bars represent SD of 3 replications.
Fig 3.
AVR-PikD stabilizes sHMA proteins and alters sHMA subcellular localization in N. benthamiana.
(A) sHMA proteins (Myc:OsHIPP19, Myc:OsHIPP20 and Myc:OsHIPP17) were transiently expressed in N. benthamiana leaves together with either GUS, HA:AVR-Pii or AVR-PikD:HA and were detected by an anti-Myc antibody. The result for supernatant fraction after fractionation of leaf extract is shown. The result for pellet fraction is shown in S9 Fig. OsHIPP19 and OsHIPP20 bound by AVR-PikD remain largely stable, whereas OsHIPP19 and OsHIPP20 expressed with GUS or AVR-Pii were degraded to a lower mass fragment. OsHIPP17 binds AVR-PikD only weakly and is degraded even in the presence of the effector. AVR-PikD seems unstable when unbound to target proteins. (B) GFP:OsHIPP19 and GFP:OsHIPP20 seem to accumulate at plasmodesmata. Plasmodesmata are stained by aniline blue (blue color). White arrows indicate colocalization of GFP and aniline blue (Cyan color). Co-expression of AVR-PikD:HA relocates GFP:OsHIPP19 and GFP:OsHIPP20 from plasmodesmata. GFP:OsHIPP17 co-expressed with AVR-PikD:HA shows no relocation. Scale bar: 20 μm. (C) A magnified view of GFP-OsHIPP20 in an inset square of (B) for GFP (left), aniline blue (center) and merged view of GFP and aniline blue (right). Scale bar: 5 μm. (D) GFP:OsHIPP19 and GFP:OsHIPP20 accumulate to punctae-like structures in the cells when expressed with GUS or HA:AVR-Pii, whereas these proteins were evenly distributed in the cytoplasm when expressed with AVR-PikD:HA. We obtained similar results in three independent experiments. Scale bar: 200 μm.
Fig 4.
OsHIPP20 is a susceptibility gene (S-gene).
(A) Results of punch inoculation of conidia of a compatible isolate Sasa2 to the wild-type Sasanishiki (WT) and T2 generation of two homozygous KO rice groups (guide RNA-1 [g1] and guide RNA-2 [g2], each with two replicates [g1-1, g1-2 and g2-1 and g2-2]). Box plots show lesion area sizes in the rice lines (top). Statistical significance is shown after Wilcoxon rank sum test. Photos of typical lesions developed on the leaves after inoculation of M. oryzae (bottom). The number (N) of leaves used for experiments are indicated below. (B) Results of punch inoculation of Sasa2 conidia to T2 progeny segregated to the wild type allele (W/W) and KO-type allele (m/m) from a T1 heterozygous KO line (g1-3). Statistical significance is shown according to Wilcoxon rank sum test. (C) Results of spray inoculation of conidia of a compatible isolate Ken53-33 to the wild-type Sasanishiki (WT) and T2 generation of the OsHIPP20-knockout lines g1-2 and g2-1. Box plots show the relative content of fungal actin gene DNA (M. oryzae actin gene DNA / rice actin gene DNA) as determined by quantitative PCR. Statistical significance is shown according to Wilcoxon rank sum test. (D) No growth defect in OsHIPP20 KO line as compared to the wild type one month after seed sowing. Top: Overview of the W/W and m/m plants. Bottom: Box plot showing plant height distribution of W/W and m/m plants segregated from the g1-3 heterozygous KO line. Statistical significance is shown according to Wilcoxon rank sum test. Scale bar: 5cm.
Fig 5.
Schematic representation of a model showing molecular interactions between the AVR-Pik effector, rice sHMA proteins and Pik NLRs.
In the compatible interaction (susceptible, left), AVR-Pik binds rice Clade A sHMA proteins, stabilizes and relocalizes them, possibly enhancing pathogen virulence. OsHIPP20 is a S-protein required for effective M. oryzae invasion. In the incompatible interaction (resistant, right), AVR-Pik interacts with integrated HMA domains of the Pik-1 NLRs which, together with Pik-2, triggers disease resistance by the hypersensitive response (HR). AVR-Pik and Pik seem involved in arms-race coevolution (selective force to enhancing interaction in Pik and evading interaction in AVR-Pik) by each generating multiple variants.