Genome-wide association studies of resistance to dieback diseases in a pseudo-F2 mango population
Lasiodiplodia theobromae and Neofusicoccum parvum are important fungi affecting mango trees in Northeast Brazil, a prominent region of mango export. Genome-wide association studies in a ‘Haden’ × ‘Tommy Atkins’ mango pseudo-F2 population were performed for symptoms of both fungal diseases to support the development of new cultivars by applying marker-assisted selection. ‘Haden’ is resistant while ‘Tommy Atkins’ is susceptible to both fungal diseases. Single nucleotide polymorphism (SNP) and microsatellite data of 95 progenies were analyzed by allelic and genotypic association and by general (GLM) and mixed linear models (MLM). Artificial pathogen inoculation was performed on 15-year-old progenies by manually spraying a 103 conidia mL-1 suspension on young branches and leaves. The plants were considered resistant when the absence of symptoms was ≥ 90% over three different evaluations. Consensus genomic associations were identified on chromosome 12; position 10.60 Mb (Mi_0096) and 1 (Mango_rep_c1316) position 14.67 Mb, with a significant association with L. theobromae symptoms, accounting for 20% of the total variation. Additional regions identified exclusively by GLM and MLM analysis, in chromosomes 11 and 8 (positions 24.57 Mb and 9.34 MB, respectively), explain 36% of the symptoms variation of this disease. Consensus genomic associations were identified on chromosomes 2, position 20.49 Mb (Mango_rep_c9407) and 9, position 15.01 Mb (Mango_rep_c8984), with a significant association with N. parvum symptoms, accounting for 21% of total resistance to this fungus. An additional region identified exclusively by GLM and MLM analysis, in chromosome 12 (Mango_rep_c7620, position 14.29 MB), explains 29% of the total variation of this disease. Qualitative and quantitative genome association methods run together enabled the identification of consensus chromosomal regions controlling resistance to these diseases. These chromosome regions are candidates for saturation with SNPs or further genome data mining to apply marker-assisted selection in mangoes.