Molecular characterization of iron deficiency chlorosis in soybean

Iron deficiency chlorosis (IDC) of soybean occurs on calcareous soils when a cultivar is unable to utilize the iron (Fe) in the soil. The objectives of this study were to map gene(s) for Fe chlorosis resistance in two Glycine mm x G. maw populations, and to evaluate the use of marker-assisted selection in breeding for improved Fe chlorosis resistance in soybean. Chlorosis symptoms were evaluated by visual scores and spectrometric chlorophyll determinations at the V4 stage (third trifoliate leaf fully developed) in each of two years. Ninety restriction fragment length polymorphism (RFLP) and 10 simple sequence repeat (SSR) markers in the Pride B216 x A15 population, and 82 RFLP, 14 SSR and one morphological (hilum color) markers in the Anoka x A7 population were used to map quantitative trail loci (QTL) affecting IDC. In the Pride population, 120 random F-2 plants were used, 92 were used in the Anoka population. In the Pride population, QTL with minor effects were detected in three linkage groups for visual scores, and for chlorophyll concentration, for a total of five different linkage groups for both traits, indicating a polygene mechanisms for IDC. In the Anoka population, two QTL were each mapped for visual scores and chlorophyll concentration. One of the QTL had major effect and was mapped in the same interval of linkage group N using both visual scores and chlorophyll concentrations, verifying that one major gene is involved in segregation for IDC in this population. Different genetic linkage groups in soybean have been identified by letters, and as such will be used throughout the paper. Two QTL on linkage groups I and N were common to the Pride and Anoka populations and were considered for use in marker-assisted selection. A thorough analysis conducted in both populations, however, indicated the impossibility of using these markers for marker-assisted selection. In no cases were both of the QTL-flanking markers present in one population also present in the other population. On the basis of these results we concluded the markers identified in this study can not be used in marker-assisted selection because of a general lack of common markers between the two populations.