Molecular Plant PathologyPathologyPeer-reviewed journal article
Genomic surveillance and molecular evolution of fungicide resistance in European populations of wheat powderymildew
Nikolaos MinadakisiD, Jigisha JigishaiD, Luca CornettiiD, Lukas KunziD, Marion Claudia MülleriD, Stefano F. F. TorrianiiD, Fabrizio MenardoiD
Abstract
Fungicides are used in agriculture to manage fungal infections and maintain crop yield and quality. In Europe, their application on cereals increased drastically starting from the mid 1970s, contributing to a significant improvement in yields. However, extensive usage has led to the rapid evolution of resistant pathogen populations within just a few years of fungicide deployment. Here we focus on wheat powdery mildew, a disease caused by the ascomycete fungus Blumeria graminis forma specialis tritici (Bgt). Previous research on Bgt documented the emergence of resistance to different fungicides and identified various resistance mechanisms. Yet, the frequency, distribution, and evolutionary dynamics of fungicide resistance in Bgt populations remain largely unexplored. In this study we leveraged extensive sampling and whole-genome sequencing of Bgt populations in Europe and the Mediterranean to investigate the population genetics and molecular epidemiology of fungicide resistance towards five major fungicide classes. We analyzed gene sequences and copy number variation of eight known fungicide target genes in 415 Bgt isolates sampled between 1980 and 2023. We observed that mutations conferring resistance to various fungicides increased in frequency over time, and had distinct geographic distributions, likely due to diverse deployment of fungicides across different regions. For demethylation inhibitor fungicides we identified multiple independent events of resistance emergence with distinct mutational profiles, and we tracked their rapid spread in the last decades. Overall, we revealed the evolutionary and epidemiological dynamics of fungicide resistance mutations in European Bgt populations. These results underscore the potential of genomic surveillance and population genetics to enhance our understanding of fungicide resistance.
- organism: fungi (other)
- assay: whole-genome sequencing
- assay: long-read sequencing
- research focus: phylogenetics / evolution
- research focus: chromatin & epigenetic regulation
- research focus: differential expression
- approach: statistical model
In-depth summary
Claude Opus 5.5 · xhigh effort · 8 Nov 2026 · from the full text of version 1 (31 Oct 2024) · CC BY 4.0
In 415 European and Mediterranean wheat powdery mildew genomes, fungicide-resistance mutations rose over time with regional patterns; demethylation-inhibitor resistance at cyp51 emerged at least twice independently and spread.
Wheat powdery mildew (Blumeria graminis f. sp. tritici, Bgt) evolves fungicide resistance quickly, yet the frequency and spread of resistance in Europe are largely unexplored; the only systematic monitoring is run by industry, and only summaries are made public. Reusing published genomes, this work screens eight fungicide target genes for sequence and copy-number variation and uses identity-by-descent analysis to trace resistance origins, which sensitivity screening cannot do.
Key results
ShownG143A in cytb, the quinone outside inhibitor (QoI) resistance substitution, is common but not fixed. It is near-universal in Northwestern Europe, intermediate in Central and Eastern Europe, rare in the south and absent in the Middle East, and it has increased over time in Western Europe.
G143A (nucleotide G428C) was found in 146 of 368 (39.7%) Europe+_recent isolates, and every sample from Ireland, England, Northern France and the Netherlands carried it. In the temporal dataset (UK, France, Switzerland), Fisher's exact test gave p < 0.001 comparing 1990-2001 with 2022-2023. · cytb – QoI resistance; Discussion
ShownMultiple cyp51 copies and Y136F, a mutation linked to demethylation inhibitor (DMI) resistance, are widespread in Northern Europe, Spain and Turkey. Y136F occurs only in multi-copy isolates and rose in frequency, but mean copy number showed no significant increase between pre-2001 and 2022-2023 samples.
Isolates carried 1 to 7 cyp51 copies; 88% (324/368) had more than one, and single-copy isolates came mostly from Southern Europe and the Middle East. Y136F was in 298/368 (81%), Fisher p < 0.001 (1980-2001 vs 2022-2023). For copy number, Welch t-test p = 0.351. Estimates agreed with a PacBio assembly (CHVD042201: four identical tandem copies) and with ddPCR for two isolates (two copies each). · Copy number variation; Amino acid mutations in cyp51; cyp51 - DMI resistance
SuggestedIdentity-by-descent (IBD) clustering at cyp51 points to at least two independent origins of DMI resistance that then spread. Cluster i has more copies, a short repeat unit and homoallelic Y136F with S79T and K175N; cluster ii has fewer copies, a longer unit and heteroallelic Y136F and S509T.
isoRelate on a 3 Mb window around cyp51 in 368 isolates gave two large clusters of 129 and 75 isolates (Europe and Turkey). Cluster i had more copies (Welch p < 0.001), and all its members shared a repeat unit of about 2.5 Kb. In cluster ii, 72/75 isolates were heteroallelic at position 136 and 65 at position 509. S509T (62 isolates, 16%) was found, among the two largest clusters, only in cluster ii. · Amino acid mutations in cyp51; Recent evolution of cyp51 in Europe; cyp51 - DMI resistance; IsoRelate and haplotype networks
SuggestedIn erg24 (a morpholine target), V295L is common but almost entirely limited to Northern Europe. Two nearby mutations, D291N (first report in Bgt) and the new F289H, were seen only in recent northern samples, and the authors propose both as likely morpholine-resistance mutations.
Eight amino acid mutations were found. V295L was in 141/368 (38%), with a temporal trend of Fisher p = 0.075. D291N was in 8 isolates from 2022-2023; F289H was in 36 N_EUR isolates and absent before 2007 (p = 0.195). Y165F did not change over time (p = 0.732). In the haplotype network, the wildtype was the most common haplotype overall, especially abundant in Southern Europe, Turkey and the Middle East. · erg24 – morpholine resistance; erg24 - Morpholine resistance
ShownThe targets of succinate dehydrogenase inhibitors (sdhB, sdhC, sdhD), of morpholines (erg2) and of methyl benzimidazole carbamates (beta tubulin) show almost no variation, so no candidate resistance signal was found in these genes.
There were only a few amino acid mutations, at frequencies between 0.3% and 2.5%, and virtually no copy number variation across these five genes. · Results; Other target genes
Methods, data and reproducibility
Short-read genomes of 415 Bgt isolates (1980-2023) were mapped to reference 96224 and called as haploid with GATK. For eight fungicide target genes the authors built amino acid alignments and coverage-based copy numbers, tested change over time (Fisher's exact, Welch t-tests), and traced cyp51 history with isoRelate IBD clusters and tandem-repeat coverage profiles, and erg24 history with a haplotype network.
- Europe+: short-read whole-genome sequences of Bgt isolates from Europe and the Mediterranean, 1980-2023 (from Jigisha et al. 2024) · Blumeria graminis f. sp. tritici · n = 415 isolates
- Europe+_recent: subset collected from 2015 to 2023, the main analysis set · Blumeria graminis f. sp. tritici · n = 368 isolates
- Temporal dataset: isolates from the UK, France and Switzerland spanning more than 40 years · Blumeria graminis f. sp. tritici · n = 83 samples
- Nuclear genome annotation for the 11 chromosomes · Blumeria graminis f. sp. tritici · PRJEB28180
- Mitochondrial genome annotation (Zaccaron and Stergiopoulos 2021) · Blumeria graminis f. sp. tritici · MT880591
- RNA-seq of isolates 96224 and 94202, used to check gene annotations · Blumeria graminis f. sp. tritici · n = 2 isolates · GSE108405
- PacBio long-read assembly of isolate CHVD042201, used to validate cyp51 copy number · Blumeria graminis f. sp. tritici · n = 1 isolate · PRJNA1131794
- Methods & tools
- fastp v0.23.2, bwa-mem, samtools v1.17, GATK v4.4.0, bcftools v1.17, STAR v2.7.11b, IGV v2.16.2, blastn v2.9.0+, redotable v1.2, isoRelate, pegas (haploNet), igraph
- Code
- https://github.com/fmenardo/Bgt_fungicides_2024https://github.com/fmenardo/Bgt_popgen_Europe_2024/blob/Bgt_ms/WGS_pipeline/WGS_pipeline.md
- Data & accessions
- https://github.com/fmenardo/Bgt_fungicides_2024/tree/main/Alignmentshttps://github.com/fmenardo/Bgt_fungicides_2024/tree/main/Fungicide_targets
- Software & environment
- The text gives tool versions: fastp v0.23.2, samtools and bcftools v1.17, GATK v4.4.0, STAR v2.7.11b, IGV v2.16.2, blastn v2.9.0+, redotable v1.2, R 4.3 with tidyverse. No container or requirements file is mentioned.
- Reproducibility gaps
- Read accessions for the 415 isolates are listed only in Supplementary Data S1, not in the main text. The source of the genetic map used to filter SNPs for isoRelate is a citation that is not resolved in the text.
Limitations (6)
- There are no up-to-date, region-level data on fungicide use, so the roles of migration and selection in the frequencies of G143A and V295L cannot be separated. · stated by the authors
- Short reads cannot resolve the individual cyp51 copies, so a detected mutation is known only to be in at least one copy; the authors say long-read data would help with complex loci like cyp51. · stated by the authors
- isoRelate detects IBD only within about 25 sexual generations, while DMIs date from the 1970s, so it cannot be determined whether the smaller clusters (for example v, vi, vii) are truly independent origins. · stated by the authors
- F289H and D291N are proposed as morpholine-resistance mutations without sensitivity testing; the authors call for sensitivity assays combined with erg24 sequencing to test this. · stated by the authors
- Expression of the target genes was not measured, although expression could contribute to insensitivity and be under selection. · stated by the authors
- The temporal comparisons rest on 83 isolates from three Western European countries, and some of the erg24 trends cited as increases (V295L p = 0.075, F289H p = 0.195) are weak statistically. · noted in this summary
Read this if you work on surveillance of fungicide resistance in plant pathogens, or want a worked example of using population whole-genome sequencing and IBD analysis to trace how resistance alleles and copy-number variants arose and spread.
- Citations
- 10
- FWCI
- 4.65
- Field percentile
- top 5%
Reproducibility ledger
Nobody has reported trying to reproduce this paper yet. Did you replicate an experiment or run the code? Your report saves the next lab a week.
Sign in to file a reproducibility report.
Related papers
- Population genomics and molecular epidemiology of wheat powdery mildew in Europe
Jigisha Jigisha, Jeanine Ly, … Fabrizio Menardo·PLoS BiologyPeer-reviewed·May 2, 2025
- The genetic architecture of emerging fungicide resistance in populations of a global wheat pathogen
Danilo Pereira, Bruce Alan McDonald, Daniel Croll·Genome Biology and EvolutionPeer-reviewed·Sep 24, 2020
- A large European diversity panel reveals complex azole fungicide resistance gains of a major wheat pathogen
Guido Puccetti, Thomas Badet, … Daniel Croll·mBioPeer-reviewed·Nov 24, 2025
- Cultivar-specific fungicide resistance emerges during a growing season in field populations of Zymoseptoria tritici
Firas Talas, Jessica Stapley, Bruce Alan McDonald·Pest Management SciencePeer-reviewed·Jan 19, 2026
- Large-scale study validates that regional fungicide applications are major determinants of resistance evolution in the wheat pathogen Zymoseptoria tritici in France.
M. Garnault, C. Duplaix, … F. Carpentier·bioRxiv·Jul 17, 2020
- High-throughput discovery of emerging antifungal resistance in crop pathogens
Guido Puccetti, Daniel Flückiger, … Daniel Croll·bioRxiv·Jul 18, 2025