| The partial sterility of intersubspecific hybrids between Oryza sativa ssp. Indica and O. sativa ssp. Japonica is a major barrier for the utilization of the heterosis. Therefore, study on the genetics of intersubspecific hybrids sterility has great theoretically and practically meaningful.In this study, a F2 population composing 176 plants derived from zhenshan97B (Indic maintainer) crossed with xiushui13 (a Japonica variety form south China) was made. And a molecular linkage map containing 107 SSR markers and covering all 12 chromosomes was constructed. Its total length is 1574.5 cM with a average marker distance about 14.71 cM. Based on the map, QTLs underlined pollen and spikelet fertility were preliminarily mapped by means of the methods of composed interval mapping, interval mapping and single marker analysis, respectively. The mail results are summarized as follows:1 .Investigation showed that the percentage of pollen and spikelet fertility of zhenshan97B was 97.53% and 94.9%, xiushuil3 was 97.13% and 84.3%, and F1 plants was 59.44% and 48.01%, respectively. In F2 population, both pollen and spikelet fertility behaved the continuous segregation, indicating that they were genetically quantitative. Relation analysis showed that the relation coefficient of 0.3332 between pollen and spikelet fertility was significant at the level of 1%.2. By composed interval mapping, two QTLs, pf5 and pf12, for pollen fertility were detected on chromosome 5 and 12, respectively. pf5 was close to the markers between RM13-RM267 and pfl2 close to the markers between RM19-RM247. pf5 and pf12 with their additive effects of 0.253 and 0.202, can explain 21.3% and 13.4% of phenotypic variance, respectively. On the other hand, three QTLs, spf5, spf6 and spf8, for spikelet fertility were mapped between the markers of RM13-RM267, RM111-RM217 and RM72-RM407 on chromosome 5, 6 and 8, respectively. These three QTLs are with their additive effects of 0.267, 0.157 and 0.114, and can explain 25.33%, 13.5% and 8.9% of phenotypic variance, respectively. Because pf5and spf5 on chromosome 5 shared the same markers, it suggested that they are the same QTL. 3. Two QTLs,pf1 on chromosome 1 close to the markers of RM9-RM488 and pf5 on chromosome 5 close to RM13-RM267, were detected for pollen fertility by means of interval mapping. The pf1 has additive effect of 0.151, or 7.5% of phenotypic variance. Based on its position on chromosome and additive effect, the pf5 detected by both composed interval mapping and interval mapping may be the same QTL. As to the spikelet fertility, three QTLs, spf5, spf6 and spf8, were detected on chromosome 5, 6 and 8. Their additive effects are 0.264, 0.156 and 0.113, and explain 25%, 12.8% and 8.8% of phenotypic variance, respectively. When compared on their positions, additive effects and LOD value, these three QTLs mapped by interval mapping would be the same QTLs mapped by composed interval mapping.4. According to single marker analysis, 10 and 6 markers significantly relative to pollen and spikelet fertility were found, respectively. These markers are responsible for the linked markers of QTLs detected by composed interval mapping and interval mapping with the exception of marker RM107.5. Analysis indicated that interactions between loci could greatly affect both pollen fertility and spikelet fertility. At a significance level of 0.005, interactions between at least 46 and 38 pairs loci in this population played an important role in pollen and spikelet sterility expression, respectively, suggesting that epitasis also was one of major genetic components controlling indica/japonica hybrid sterility. |
PDF Full Text Request