The spikelets of the grass tribe Andropogoneae, to which maize belongs, are borne in pairs, one sessile and one pedicellate. This paired arrangement is also found in maize ears, although the highly compact nature of the maize ear makes it difficult, but not impossible, to distinguish the sessile and pedicellate spikelets since both are essentially sessile (H. C. Cutler, Bot. Mus. Leaflet, Harvard Univ. 12:257-291). We describe a dominant mutant that suppresses the formation of sessile spikelets in maize ears. Seeds of the mutant were taken from an ear showing the mutant phenotype in the University of Wisconsin Herbarium. According to Dr. Hugh Iltis (Director of the Herbarium), he obtained the ear from Dr. John Lonnquist who reported to him that the mutant arose spontaneously in a maize population.

Seeds originally obtained from the herbarium specimen were selfed for several generations until a line (91-31) that was true-breeding for the single spikelet trait was obtained. This line also had a poor tassel with few short branches. Subsequently, 91-31 was crossed to W22 and W22-TGA (a W22 derivative carrying a segment of teosinte chromosome 4 and the teosinte allele at tga1 within this segment). The F1's of both crosses were grown and found to exhibit the single spikelet trait, indicating that this trait was dominant to the normal maize condition of paired spikelets. The 91-31xW22 F1 was backcrossed to W22, and the 91-31xW22-TGA F1 was selfed to produce an F2 population.

During the summer of 1992, we analyzed the 91-31xW22 backcross population. This was one of the coldest summers on record in Minnesota and plant growth was poor. Among the 58 progeny analyzed, 31 exhibited paired spikelets and normal tassels. The remaining 27 plants were barren (without ears) and had tassels that consisted only of a central spike without any branches. The 31:27 ratio suggested a 1:1 ratio expected if a single locus controlled the differences in tassel structure and barrenness. The absence of the single spikelet trait among those plants that had ears prevented us from scoring this trait. One possibility we considered was that the cold weather had induced the barrenness in plants carrying the factor that causes the single spikelet trait and that this factor also affects the production of tassel branches. Working on this assumption, we analyzed the 58 plants for RFLP markers, using one marker per chromosome arm. We detected linkage between an RFLP marker and barrenness/unbranched tassels on chromosome arm 4S. After mapping additional RFLPs on this chromosome arm, we mapped these traits between php20725 and bnl5.46.
        php20725 - 2.6 - barrenness/unbranched tassels - 4.4 - bnl5.46
Distances are the recombination fractions.

In the summer of 1993, we again attempted mapping the single spikelet trait using the 91-31xW22-TGA F2 population. Plant growth this year was better and, in a population of 58 F2 plants, we observed 46 with single spikelets and 12 with paired spikelets in the ear. These numbers do not differ from the expected 3:1 ratio if the single spikelet trait is controlled by a single dominant locus (X2 = 0.32, d.f = 1, p > 0.5). We designate this locus Suppressor of sessile spikelets1 (Sos1). RFLP analysis demonstrated that Sos1 maps between php20725 and bnl5.46.
        php20725 - 3.7 - Sos1 - 9.6 - bnl5.46
Because the single spikelet trait mapped to the same location as barrenness/unbranched tassels in the 1992 backcross population and because all these traits behaved as dominants, we infer that Sos1 alone controls these traits and that Sos1 plants have a tendency toward barrenness under poor growth conditions. We should also note that the introgressed teosinte chromosome segment in W22-TGA does not extend between php200725 and bnl5.46, and, thus, Sos1-Ref was segregating with sos1+W22 in 91-31xW22-TGA F2 population.

Figure 1 shows an ear carrying Sos1-Ref. The absence of the sessile spikelets leaves a gap between the rows of pedicellate spikelets. We were able to confirm that it is the sessile spikelet that is suppressed because this spikelet arises as a branch of the primordium that forms the pedicellate spikelet. Examination of ear primordia of Sos1-Ref maize revealed that this branch is not formed. Sos1-Ref also affects the formation of tassel branches, formation of sessile spikelets in the tassel, and the number of rows of cupules in the ear. The wild type function of sos1 may be in some aspect of the formation of inflorescence primordia, such as governing the number of cells committed to each branch primordium.

Our interest in Sos1 arose because the single spikelet trait of this mutant seems to resemble the single spikelets of teosinte ears, the probable ancestor of maize. Thus, Sos1 is a candidate for a gene involved in the evolution of maize from teosinte. Three observations suggest that this is not the case. (1) In teosinte ears, it is the pedicellate and not the sessile spikelet that is lacking. (2) In teosinte ears, both the pedicellate and sessile spikelet primordia are formed, but then the pedicellate spikelet is aborted. With Sos1, only a single spikelet primordium is formed. (3) In our QTL (quantitative trait locus) mapping studies (Doebley and Stec, Genetics 134:559-570, 1993), the QTLs controlling the difference of paired versus single spikelets between maize and teosinte does not map to this region of chromosome arm 4S. We are currently testing whether the sos1 allele of teosinte is equivalent to the maize wild type allele (sos1+maize) by examining an F2 population derived from a teosinte x Sos1-Ref maize cross. If the teosinte allele is functionally the same as sos1+maize, then we should recover plants with paired spikelets in the F2 population.

Figure 1. Ears of Sos1-Ref and sos1+W22 maize.

Return to the MNL 68 On-Line Index
Return to the Maize Newsletter Index
Return to the MaizeGDB Homepage