Bacterial spot of tomato is a disease complex caused by at least four species of Xanthomonas and leads to severe yield and quality losses in humid growing conditions in the world. Five physiological species (T1-T5) have been defined by their virulence on tomato varieties. These races are associated with three species: X. euvesicatoria (T1), X. vesicatoria (T2), and X. perforans (T3-T5). Recent epidemics of X. gardneri has occurred in the Midwest United States. In this study, we developed germplasm with resistance to multiple species of bacterial spot. Six advanced breeding lines with at least three different source of resistance were crossed and their F1 hybrids were inter-mated to produce a complex breeding population consisting over 1,100 progeny. Three lines (OH7663, OH7667 and 2k7-6117-S2) were selected by field evaluations of the population against T1, T2, T3, and X. gardneri. Graphical genotypes demonstrated that these breeding lines contain a QTL and Rx-4/Xv3 in coupling phase on chromosome 11 as well as Rx-3 on chromosome 5. In order to test the combining ability of the lines, we developed hybrids from multiple crosses and conducted replicated field trials to evaluate bacterial spot resistance and yield. As a male parent, OH7663 showed acceptable combining ability for yield and for resistance against multiple species of Xanthomonas. Several hybrids produced yields that were not significantly different from yields of commercial varieties.

Bacterial spot of tomato (Solanum lycopersicum L.) is caused by at least four species of Xanthomonas with multiple physiological races. In this study, we developed a mapping population for association analysis of bacterial spot resistance. For this population, six advanced breeding lines with distinct sources of resistance were first crossed in all combinations and their F1 hybrids were intercrossed. The 1,100 segregating progeny from these crosses were evaluated in the field against T1 strains. Based on this individual evaluation, we selected 5% of the most resistant and 5% of the most susceptible progeny for evaluation as plots in two subsequent replicate field trials inoculated with T1 and T3 strains. A total of 461 markers across 12 chromosomes were used for genotyping these selections. Of these markers, an optimized subset of 384 SNPs was derived from the 7,720 SNP Infinium array developed by the Solanaceae Coordinated Agricultural Project (SolCAP). For association analysis to detect known resistance loci and additional novel loci, we used the mixed models with correction for population structure, and found that accounting for kinship appeared to be sufficient. Detection of known loci was not improved by adding a correction for structure using either a Q matrix from model-based clustering or covariate matrix from Principal Component Analysis. Both single-point and haplotype analyses identified strong associations in the region of the genome known to carry Rx-3 (chromosome 5) and Rx-4/Xv3 (chromosome 11). Additional QTL associated with resistance were detected on chromosomes 1, 3, 4, 6 and 7 for T1 resistance and chromosomes 2, 4, and 6 for T3 resistance. Haplotype analysis improved our ability to trace the origin of positive alleles. These results demonstrate that both known and novel associations can be identified using complex breeding populations that have experienced directional selection.