Molecular evolution of the Pi-d2 gene conferring resistance to rice blast in Oryza

Abstract

The exploitation of plant disease resistance (R) genes in breeding programs is an effective strategy for coping with pathogens. An understanding of R gene variation is the basis for this strategy. Rice blast disease, caused by the Magnaporthe oryzae fungus, is a destructive disease of rice. The rice blast resistance gene Pi-d2 represents a new class of plant R gene because of its novel extracellular domain. We investigated the nucleotide polymorphism, phylogenetic topology and evolution patterns of the Pi-d2 gene among 67 cultivated and wild rice relatives. The Pi-d2 gene originated early in the basal Poales and has remained as a single gene without expansion. The striking finding is that susceptible Pi-d2 alleles might be derived from a single nucleotide substitution of the resistant alleles after the split of Oryza subspecies. Functional pleiotropy and linkage effects are proposed for the evolution and retention of the disease-susceptible alleles in rice populations. One set of DNA primers was developed from the polymorphic position to detect the functional nucleotide polymorphism for disease resistance of the Pi-d2 gene based on conventional Polymerase Chain Reaction. The nucleotide diversity level varied between different domains of the Pi-d2 gene, which might be related to distinct functions of each domain in the disease defense response. Directional (or purifying) selection appears dominant in the molecular evolution of the Pi-d2 gene and has shaped its conserved variation pattern.

Keywords: Pi-d2; evolutionary history; origin; rice blast resistance gene; selection.

FIGURE 1

Sliding window of nucleotide diversity and positive-selection sites of Pi-d2 in Oryza sativa ssp. Japonica, O. sativa ssp. indica and O. rufipogon subgroups. (A) The nucleotide diversity of Pi-d2 in three subgroups. The X-axis represents the nucleotide position of Pi-d2; the Y-axis indicates value of nucleotide diversity per site. Values were assigned to the nucleotide at the midpoint of 5 bp. (B) The positive-selection sites of the Pi-d2 allelic variants under MEC model. The X-axis represents the position of the Pi-d2 amino acids; the Y-axis indicates the ratio of non-synonymous substitution (Ka) to the rate of synonymous substitution (Ks) (Ka/Ks); the B-lectin, PAN-AP, and S-TKc domains are marked on the corresponding region above the sliding windows. Structure of the Pi-d2 coding region is shown at the top. O. sativa ssp. Japonica, O. sativa ssp. indica, O. rufipogon and the all group are represented by blue, red, black and gray lines, respectively.

FIGURE 2

A haplotype network based on nucleotide polymorphisms of the Pi-d2 coding region of 66 accessions of 11 Oryza species: O. sativa ssp. indica, O. sativa ssp. Japonica, O. rufipogon, O. nivara, O. meridionallis, O. glaberrima, O. barthii, and O. glumaepatula, O. punctata and O. brachyantha. Each group of haplotypes is shown as a solid circle. Each branch represents a single mutational step. Branches with small solid circles indicate that there is more than a single mutational step between haplotypes. A number next to a branch represents the length of the mutational steps. Different sizes of circles represent the different numbers of each haplotype.

FIGURE 3

Reconstruction of the evolutionary history of Pi-d2 (left) and OSPUB15 (right) in Oryza with Leersia perreri as outgroup. The tree shows a set of inferred nucleotides (states) at the Pi-d2 polymorphic position 1383. Non-synonymous changes at the codon are depicted in black (ATA) or red (ATG) next to their corresponding node. Nucleotides (guanine in red and adenine in black) at position 1383 of Pi-d2 in different accessions are indicated at the end of the name. Pi-d2 and OSPUB15 identified in the same accessions are connected by dotted lines (the assumed susceptible alleles are dotted in red, while the assumed resistant alleles are dotted in blue).

FIGURE 4

Development and validation of the FNP-based molecular marker. (A) Primer set (Pi-d2-S/AS) designed to target the FNP site that distinguished resistant and susceptible alleles with an expected PCR-amplified product of 616 bp. The numbers shown at the left of each sequence were referred to the landraces in Supplementary Table S2. The functional nucleotide polymorphic site of landrace 3, 4, 7–11, 13–16 was adenine (A) highlighted in blue. (B) PCR based validation of the primer set in eighteen landraces by PCR. M: marker. 1–18: landraces referred in Supplementary Table S2.