Genomic surveillance uncovers a pandemic clonal lineage of the wheat blast fungus

Abstract

Wheat, one of the most important food crops, is threatened by a blast disease pandemic. Here, we show that a clonal lineage of the wheat blast fungus recently spread to Asia and Africa following two independent introductions from South America. Through a combination of genome analyses and laboratory experiments, we show that the decade-old blast pandemic lineage can be controlled by the Rmg8 disease resistance gene and is sensitive to strobilurin fungicides. However, we also highlight the potential of the pandemic clone to evolve fungicide-insensitive variants and sexually recombine with African lineages. This underscores the urgent need for genomic surveillance to track and mitigate the spread of wheat blast outside of South America and to guide preemptive wheat breeding for blast resistance.

Fig 1. The emergence of wheat blast in Bangladesh and Zambia was caused by the B71 genetic lineage of Magnaporthe oryzae.

(A) Neighbor-joining tree of 537 worldwide distributed M. oryzae isolates based on 84 concatenated SNPs obtained by multiplex amplicon sequence and/or genome sequences. The topology corresponds to the optimal tree drawn from 1,000 bootstrap replicates. Isolates that belong to the B71 lineage are shown with orange (13 Zambian isolates) and red (71 Bangladeshi isolates and the Bolivian B71) background shades. (B) Geographical distribution of M. oryzae samples used in the tree. The colored points represent the approximate geographical origin of the isolates. Colors in (A) and (B) correspond to the plant host organism (upper inset). The base map was created with the R package rworldmap v.1.3.4 (data from: Natural Earth data v.1.4.0). (C) Distribution of the collection year of M. oryzae wheat-infecting isolates used in the phylogenetic analysis. The data underlying this figure can be found in https://doi.org/10.5281/zenodo.7590238. SNP, single nucleotide polymorphism.

Fig 2. Wheat blast outbreaks in Zambia and Bangladesh originated by independent introductions of the recently emerged B71 clonal lineage.

(A) The B71 lineage shows reduced genetic diversity in comparison with South American wheat-infecting isolates. PCA was performed based on genome-wide pairwise Hamming distances of 71 isolates from South America, Asia, and Africa. The colors of the points indicate the provenance of each isolate (see inset). The circular shaded area indicates isolates from the Brazilian cluster (PY0925) that is the closest to the B71 cluster. Axes labels indicate the percentage of total variation explained by each PC. The magnified area shows isolates that are part of the B71 lineage. (B) The B71 cluster is a clonal lineage. The scatter plots show pairwise LD (measured as r²) between SNPs that are at most two megabases apart. The red solid line in the South American cluster represents a fitted exponential decay model using nonlinear least squares. The histograms display LD expressed as D’ for genome-wide SNPs. The points and bars are colored as indicated in the inset. (C) The B71 clonal lineage has recently expanded with independent introductions in Zambia and Bangladesh. The scatter plot shows the linear regression (dotted line) of root-to-tip patristic distances (y-axis) versus collection dates (x-axis) for the isolates of the B71 clonal lineage. Maximum likelihood tip-calibrated time tree of the B71 cluster isolates (the PY0925 cluster was used as an outgroup). The horizontal gray and green bars indicate 95% CI for divergence dates (in calendar years) calculated using BacDating and BEAST2, respectively. The points and isolate names are colored as indicated in the inset. The data underlying this figure can be found in https://doi.org/10.5281/zenodo.7590238. CI, confidence interval; LD, linkage disequilibrium; PCA, principal component analysis; SNP, single nucleotide polymorphism.

Fig 3. Rmg8 confers resistance against Zambian wheat blast isolates.

(A) The wheat blast lineage contains 5 AVR-Rmg8 variants. Neighbor-joining tree based on amino acid sequences of all nonredundant AVR-Rmg8 variants of 71 wheat blast lineage isolates (left). Representative isolate IDs are shown for each branch. Schematic representation of polymorphic amino acids in AVR-Rmg8 variants of the wheat blast lineage (center). Virulence phenotype associated with each AVR-Rmg8 variant on Rmg8 containing host plants (right). (B) The resistance gene Rmg8 is effective against wheat blast isolates collected in Zambia. Leaves from two weeks old seedlings of Fielder (-Rmg8, upper panel) and S-165 (+Rmg8, lower panel) wheat cultivars were inoculated with spores from Zambian wheat blast isolates, the rice blast isolate Guy11 (non-adapted, avirulent control), PY6047 (virulent control; AVR-Rmg8 eII”‘-carrier), and BTJP4-16 (avirulent on Rmg8 carrying host plants, AVR-Rmg8 eI carrier). Disease and lesion size 5 days’ post-infection. Photo taken by the authors. (C) Quantification of lesions size (in mm) of 10 leaves and three independent experiments. The data underlying this figure can be found in https://doi.org/10.5281/zenodo.7590238.

Fig 4. Zambian wheat blast isolates are susceptible to strobilurin fungicides but at risk from resistance development and can mate with prevailing finger millet blast isolates.

(A) The tree describes, based on nuclear SNPs, the phylogenetic relationship among wheat-infecting blast isolates belonging to three clonal lineages: B71, PY0925, and P29. The tree was rooted in the midpoint. The colored dots next to each isolate label represent the resistant-type allele of the mitochondrially encoded cyt b gene associated with a susceptible or resistant predicted phenotype as shown in the inset. (B, C) 4 × 10⁶ spores of the Zambian strain ZMW20-14 were exposed to 100 μg ml¹ azoxystrobin to obtain an azoxystrobin-resistant strain with the G143A mutation (SR1). This strain, the rice infecting strain Guy11 (G11), and the wheat-infecting strains numbered 1–31, respectively, BTBa-B1, BTBA-B2, BTGS6e, BTGS6f, BTGS6g, BTGS6h, BTGP1-b, BTMP-S13-1, PY6047, BTJP4-1, BTJP4-2, BTJP4-3, BTJP4-5, BTJP4-6, BTJP4-9, BTJP4-11, BTJP4-12, BTJP4-15, BTJP4-16, BTJP4-18, ZMW18-06, ZMW20-07, ZMW18-08, ZMW19-09, ZMW18-10, ZMW18-11, ZMW19-13, ZMW20-14, ZMW20-15, ZMW20-16, and ZMW19-17 were grown for 4 days at 25°C on CM in the absence (A) and presence (B) of azoxystrobin indicating that all the Zambian and Bangladeshi isolates have the “strobilurin susceptible” genotype as anticipated by their CYTB sequences. (D) Zambian isolate ZMW20-7 (MAT-1-2) successfully produced perithecia when crossed with a finger millet blast isolate K1 (MAT-1-1) but (E) ZMW20-7 was unable to produce perithecia when crossed with a finger millet blast isolate T26 of the same mating type (MAT-1-2). Photos taken by the authors. The data underlying this figure can be found in https://doi.org/10.5281/zenodo.7590238. SNP, single nucleotide polymorphism.