An NLR paralog Pit2 generated from tandem duplication of Pit1 fine-tunes Pit1 localization and function

Abstract

NLR family proteins act as intracellular receptors. Gene duplication amplifies the number of NLR genes, and subsequent mutations occasionally provide modifications to the second gene that benefits immunity. However, evolutionary processes after gene duplication and functional relationships between duplicated NLRs remain largely unclear. Here, we report that the rice NLR protein Pit1 is associated with its paralogue Pit2. The two are required for the resistance to rice blast fungus but have different functions: Pit1 induces cell death, while Pit2 competitively suppresses Pit1-mediated cell death. During evolution, the suppression of Pit1 by Pit2 was probably generated through positive selection on two fate-determining residues in the NB-ARC domain of Pit2, which account for functional differences between Pit1 and Pit2. Consequently, Pit2 lost its plasma membrane localization but acquired a new function to interfere with Pit1 in the cytosol. These findings illuminate the evolutionary trajectory of tandemly duplicated NLR genes after gene duplication.

Fig. 1. Pit2 interacts with Pit1 in planta and suppresses Pit1-mediated cell death.

A In vivo interaction of full-length Pit1 and Pit2. The indicated tagged proteins were transiently expressed in N. benthamiana. Co-IP was performed and proteins were detected by Western blot. Ponceau staining of Rubisco serves as a loading control. Data was repeated three times with similar results. B The indicated proteins were transiently expressed in N. benthamiana. GUS serves as a negative control. Cell death was photographed at 3 dpi. C Activation of OsRac1 by Pit1 and Pit2 using Raichu-OsRac1 FRET in vivo. Color scale of emission ratio images (left) in rice protoplasts co-expressing Raichu-OsRac1 represents the level of OsRac1 activation. Scale bars, 5 μm. The bar graph (right) indicates statistical analysis of OsRac1 activation. Bars = mean ± s.d. (n  =  3 biological replicates). D The indicated combinations of proteins were transiently expressed in N. benthamiana leaves. Cell death was photographed at 3 dpi. E Cell death activity of indicated proteins in rice protoplasts. The indicated constructs were co-transfected with a luciferase reporter vector. Luciferase activity was measured 40 h after transfection. Relative luciferase activity (GUS = 1) is shown. Bars = mean ± s.d. (n  =  3 biological replicates). F Rice protoplasts from K59 WT or Pit1 KO were co-transfected with RNAi constructs against Pit2 and the luciferase reporter vector. Luciferase activity was measured 40 h after transfection. Relative luciferase activity (Mock = 1) is shown. Bars = mean ± s.d. (n  =  3 biological replicates). The asterisks indicate significant differences determined by one-way (C and E with Tukey’s test) or two-way (F with Šídák’s test) ANOVA. ^**P < 0.01, ^***P < 0.001, ^****P < 0.0001, ns indicates no significant difference. G Infection assays of Pit1 and Pit2 KO plants with the M. oryzae race 007.0. Photographs were taken at 7 dpi. Scale bar, 5 mm. H Biomass of M. oryzae was measured by qPCR and normalized with rice endogenous OsUbq. Relative infection ratio (K59 = 1) is shown. Bars = mean ± s.d. (n = 3 independent lines). Different letters indicate significant differences determined by one-way ANOVA (with Tukey’s test) (P < 0.05). Source data are provided as a Source Data file.

Fig. 2. Pit2 competes with Pit1 for binding to Pit1 to form a stable heterocomplex mainly in the cytosol.

A In vivo self-association of full-length Pit1 and Pit2. The indicated combinations of tagged proteins were transiently expressed in N. benthamiana. Co-IP was performed using anti-HA magnetic beads, and the proteins were detected by Western blot with relevant antibodies. Ponceau staining of Rubisco serves as a loading control. B In vitro pull-down assay of homo and heterocomplexes of Pit1 CC and Pit2 CC. GST, Pit1 CC-GST, or Pit2 CC-GST immobilized on GST beads was incubated with Pit1 CC-SUMO or Pit2 CC-SUMO. Input and pull-down proteins were detected by Western blot with anti-GST and anti-SUMO antibodies. C Pit2 competes with Pit1 to form heteromers in vivo. Pit1-GFP was transiently co-expressed with Pit1-HA in the presence or absence of Pit2-Myc in N. benthamiana. Co-IP was performed using anti-GFP agarose beads, and the proteins were detected by Western blot with relevant antibodies. Ponceau staining of Rubisco served as a loading control. D Pit2 CC competes with Pit1 CC to form a heteromer in vitro. Pit1 CC-GST and Pit1 CC-SUMO were co-incubated with anti-GST agarose beads and different amounts of Pit2 CC-Myc. Pull-down assay was carried out using anti-GST beads and proteins were detected by Western blot with corresponding antibodies. E Localization of Pit1 and Pit2 in rice protoplasts. Protoplasts were transfected with the fluorescent constructs Pit2-Venus with combinations of FLS2-mCherry, mCherry, or mCerulean-NLS. N, nucleus. BF, Bright-field. Scale bars, 5 μm. F Colocalization of Pit1 and Pit2 in rice protoplasts. Pit1-Venus was co-transfected with Pit2-mCherry. Scale bars, 5 μm. G Cell fractionation assay of Pit1 and Pit2. Pit1-HA, Pit2-Myc, or both were transiently expressed in N. benthamiana. Western blot was performed with indicated antibodies. T, S, and M indicate total extract, soluble fraction, and microsomal fraction, respectively. M(3X) indicates three-fold enrichment relative to T and S. Ponceau staining of Rubisco serves as a loading control. All images are representative of results repeated three times with similar results. Source data are provided as a Source Data file.

Fig. 3. Three residues in the NB-ARC domain determine the functional difference between Pit1 and Pit2.

A Schematic architecture and cell death phenotype of Pit1 D485V, Pit2 D484V and domain-swapping mutants. Red and blue indicate domains derived from Pit1 and Pit2, respectively. The indicated proteins were transiently expressed in N. benthamiana, and cell death was photographed at 3 dpi. B Cell death phenotype of Pit1 D485V and the indicated amino acid substitution mutants in N. benthamiana, photographed at 3 dpi. Red and blue characters indicate residues derived from Pit1 and Pit2, respectively. C Infection assays of Pit1 mutant plants with the incompatible M. oryzae race 007.0. Photographs show phenotypes of representative Pit1 WT and Pit1 mutant plants at 7 dpi. Scale bar, 5 mm. D Biomass of the incompatible M. oryzae race was measured by qPCR and normalized with endogenous OsUbq. Relative infection ratio (Nipponbare (Nip.) = 1) is shown. Bars represent the mean ± s.d. (n  =  3 independent lines). Different letters above bars indicate a significant difference determined by one-way ANOVA (with Tukey’s test) (P < 0.05). E Cell death phenotype of Pit2 D484V and Pit2 mutants in N. benthamiana, photographed at 3 dpi. Red and blue characters indicate residues derived from Pit1 and Pit2, respectively. F Infection assays of Nipponbare expressing Pit2 mutants with the incompatible M. oryzae race 007.0. Photographs show phenotypes of representative Pit1 WT, Pit2 WT, and Pit2 LCG plants at 7 dpi. Scale bar, 5 mm. G Biomass of the incompatible M. oryzae race was measured by qPCR and normalized with endogenous OsUbq. Relative infection ratio (Nip. = 1) is shown. Bars represent the mean ± s.d. (n  =  3 independent lines). Different letters above bars indicate a significant difference determined by one-way ANOVA (with Tukey’s test) (P < 0.01). Source data are provided as a Source Data file.

Fig. 4. Pit1 L301 and C416 are important for OsRac1 activation and plasma membrane localization.

A, B Monitoring OsRac1 activation by amino acid-substituted Pit1 (A) and Pit2 (B) mutants using Raichu-OsRac1 FRET in vivo. Statistical analysis of OsRac1 activation by Raichu-OsRac1 with normalized emission ratios of Venus to CFP. PFW, Pit1 L301P C416F G479W. Bars represent the mean ± s.d. (n  =  3 biological replicates). The asterisks indicate significant differences determined by one-way ANOVA (with Tukey’s test) (^****P < 0.0001). C Localization of Pit1, Pit2, and the indicated amino acid substitution mutants in rice protoplasts. Protoplasts were co-transfected with the indicated Venus-tagged fluorescent constructs and the plasma membrane marker FLS2-mCherry. Scale bars, 5 μm. This experiment was repeated three times with similar results. D Cell fractionation assay showing the localization of Pit1, Pit2, and the indicated mutants; the HA-tagged proteins were transiently expressed in N. benthamiana. Western blot was performed with anti-HA, anti-cAPX (cytosolic marker), and anti-H⁺ATPase (plasma membrane marker) antibodies. M(3x) is three times enrichment relative to T or S. T, S, and M indicate total extract, soluble fraction, and microsomal fraction, respectively. Ponceau staining of Rubisco served as a loading control. This experiment was repeated three times with similar results. E Quantification of localization of Pit1, Pit1 PFW, Pit2 and Pit2 LCG. One hundred transfected cells were counted under a microscope. PM, plasma membrane. Cyt, Cytosol. Source data are provided as a Source Data file.

Fig. 5. Pit homologs in O. sativa and other species.

A Phylogenetic analysis of Pit homologs in Poaceae species. Three residues corresponding to L301, C416 and G479 of Pit1 in different homologs are shown to the right. The inset highlights the estimated duplication time of Pit1 and Pit2. B The three residues corresponding to L301, C416 and G479 of Pit1 in 13 domesticated and wild rice species. The red dotted box indicates that Pit genes in L. perrieri and O. brachyantha are outside the Pit1 and Pit2 clades (Supplementary Fig. 12). This phylogenetic tree is adapted from Stein et al.. C Cell death activity of Pit from L. perrieri in rice protoplasts. The indicated constructs were co-transfected with a luciferase reporter vector in rice protoplasts, and LUC activity was measured at 40 h after transfection. Relative luciferase activity (GUS = 1) is shown. Bars represent the mean ± s.d. (n  =  3 biological replicates). The asterisks indicate significant differences determined by one-way ANOVA (with Tukey’s test) (^****P < 0.0001). D Localization of Pit from L. perrieri in rice protoplasts. Protoplasts were transfected with the fluorescent constructs Pit1-Venus or Pit^per-Venus in combination with the control fluorescent protein FLS2-mCherry, and fluorescence in the protoplasts was observed at 12–16 h after transfection. Scale bars, 5 μm. Source data are provided as a Source Data file.

Fig. 6. Model of the evolution and function of Pit1 and Pit2.

The ancient Pit gene is a conserved resistance gene in Poaceae species and appears to have been duplicated in Oryza species. After duplication, Pit2 is under positive selection and the three important residues have evolved from LCG in Pit1 to PFW in Pit2, leading to Pit2 losing the ability to localize in the plasma membrane but gaining a new function to suppress Pit1-mediated cell death.