Developmental Genetic Analysis For Seed Quantitative Traits At Different Growth Periods In Rapeseed (Brassica Napus L.)

Rapeseed (Brassica napus L.) is an important oilseed crop, ranked third in the world behind palm and soybean oil. For improving the oil content (OC) or modifying different fatty acids (FA) profile in order to meet consumer or health demands, it is prerogative to have information on the genetic mechanism of these seed quality traits. Kilo seed weight (KSW), OC and FA contents are seed quantitative traits and as such are affected by genetic main effects and genotype×environment (GE) interaction effects from diploid embryo (cotyledon) nuclear genes, cytoplasmic genes and diploid maternal plant nuclear genes. Besides, varied gene expressions taking place during different seed developmental stages can also affect the performance of these quantitative traits in rapeseed. In the present experiment, developmental genetic models and their corresponding statistical approaches for quantitative traits of seeds in diploid plants were used to analyze the genetic inheritance mechanisms for KSW, OC and FA contents including palmitic acid content (PAC), oleic acid content (OAC), linoleic acid content (LAC), linolenic acid content (LLAC), eicosenoic acid content (EIAC) and erucic acid content (EAC) in rapeseed. A complete diallel mating design without reciprocal crosses was used for the field experiments to obtain the seed samples of parents, F₁ and F₂ generations in the two years, which were then analysed for quality traits. The unconditional genetic analysis method was used to analyze the cumulative genetic effects (0→t) along developmental times, while the conditional genetic analysis method was used to analyze the net genetic effects of new expressions of gene(s) activated in the special period from t-1 to time t (t - 1→t). The main results were as follows:1.The phenotypic means of parents, F₁ and F₂ generations at different developmental periods for KSW, OC and EAC gradually increased along with the development times of rapeseed and reached its highest value at seed maturity (43d after flowering). A sharp rise from flowering to 22d was observed for OAC accumulation while EIAC increased from flowering to 29d, but thereafter slight increase or decrease was observed for both traits at other developmental times. For PAC and LAC, the highest levels were attained at 15d after flowering followed by a gradual decrease, with 43d recording the lowest values for both FAs. In contrast, LLAC in rapeseed was highest at 15d after flowering and later decreased to lowest at 22d, but thereafter remained some slight increase at 29, 36 and 43d. Therefore, variations for the analyzed seed traits among developmental times for different generations and environments indicated that these seed traits were simultaneously affected by genetic effects and environment.2.Unconditional variance analysis showed that KSW, OC and FA contents were simultaneously governed by genetic main effects and GE interaction effects from diploid embryo nuclear genes, cytoplasmic genes and diploid maternal plant nuclear genes, while the results from conditional variance analysis showed that the new expression of gene(s) from these genetic systems were relevant at different developmental stages. Among them KSW at 22d, 36d and 43d, OC, OAC and EAC at 36d and 43d, EIAC at 15d, 29d, 36d and 43d, LAC at 15d and 43d, LLAC at 15d were mainly governed by genetic main effects, while the performance at other developmental times for these seed traits and the whole developmental period for PAC was affected by GE interaction effects respectively. Genetic analysis for the different genetic systems showed that KSW at 15d, 22d and 29d, OC at 15d, 29d and 43d, LAC and PAC at 15d, 22d, 29d and 43d, were mainly governed by maternal and cytoplasmic effects combined, while embryo effects for OAC at 29d, 36d and 43d, EIAC at all developmental times, LLAC at 36d and 43d, and EAC at 29d, 36d and 43d were more important. The conditional variance analysis revealed that the new expression of gene(s) especially at 30～36d for KSW and EIAC, 23～29d for OC, OAC and EAC, 1～15d for LAC and LLAC, and 16～22d for PAC were relevant as the highest genetic activity was found during these stages. Additionally, net GE interaction effects from conditional analysis were more important for all traits at most developmental times indicating that the new expression of gene(s) was to a great extent influenced by the environment.3.Additive effects were more prominent for all seed traits at most development times and so depending on varietal performance selection could be conducted in earlier generation(s) in rapeseed breeding programs. Heritability analysis showed that the general heritabilities were important for KSW, EIAC and EAC at almost all development times, OC at 36d and 43d, OAC at 15d, 36d and 43d, and LAC at 15d and 43d, and so selection for these traits was efficient and least likely to be influenced by the environment. However, selection efficiency for PAC and LLAC was more likely to be influenced by the environment because of their higher interaction heritabilities at different developmental times. Since KSW, OC and PAC had higher maternal and cytoplasmic general and interaction heritabilities, the selection can be done mainly according to the holistic perfrormance of the maternal plant for these three seed traits, while single seed selection in earlier generations could be more beneficial for improving the performance of other fatty acids due to their higher embryo general and interaction heritability at most development times.4.The results of heterosis analysis indicated that embryo main heterosis for KSW at 22d, 29d, 36d and 43d or PAC at 15d, 22d and 29d, maternal main heterosis for OC at 29d, 36d and 43d, and cytoplasmic main heterosis for EIAC at all developmental times was relevant. For the heterosis of other fatty acids considerable variation was observed among development times. Among GE interaction heterosis components, the embryo interaction heterosis for KSW at 15d and 29d, or EAC at 22d and 29d, cytoplasmic and maternal interaction heterosis at 43d for OC, embryo interaction heterosis at 22d and maternal interaction heterosis at 15d for OAC, embryo and maternal interaction heterosis at 22d, maternal interaction heterosis at 29d and cytoplasmic and maternal interaction heterosis at 43d for LAC, embryo interaction and cytoplasmic interaction heterosis at 22d for PAC, cytoplasmic interaction heterosis at 36d for EIAC, and/or embryo interaction heterosis at 22d and 29d and maternal interaction heterosis at 22d for LLAC, was found to be stable across environments.5.Correlation analysis of rapeseed traits showed that significant positive or negative relationships were existed between different developmental times. Among them, maternal additive correlation (r_Am) for KSW, maternal additive interaction correlation (r_AmE) for OC or EAC, maternal dominant interaction correlation (r_DmE) for OAC or LAC, embryo dominant correlation (r_Do) for PAC and embryo additive correlation (r_Ao) for EIAC were relevant as they were significant among most pairwise developmental times. 6.Correlation analysis between different seed traits for mature rapeseed showed that the relationship between traits could affect the selection of multiple traits in rapeseed breeding program. There was observed significant positive r_Ao and r_Am between KSW and OC, r_Dm between KSW and PAC, r_Do between KSW and OAC, r_Ao between KSW and LAC, r_DoE, between KSW and LLAC or KSW and EIAC and r_Am between KSW and EAC which indicated that selection for above traits was possible, but between KSW and LLAC or EIAC the selection could be influenced by the environment . Genetic correlations between OC and PAC, OC and OAC, OC and LAC and OC and LLAC were in most cases significant negative, while that between OC and EIAC or EAC was significant positive. There was observed negative significant r_Do, r_AoE and r_AmE between PAC and OAC or r_C and r_DoE between PAC, and EIAC, while r_C,r_Dm,r_AoEand r_AmE between PAC and LAC or r_Do, r_AoE and r_AmE between PAC and LLAC were significantly positive. Between PAC and EAC except for significant negative r_C, or positive r_AoE the others were not significant. Except for r_Ao,r_C and r_CE between OAC and LAC, the significant genetic correlation components between OAC and other fatty acids were mostly negative and this was especially true for that between OAC and EAC. LAC had significant positive r_Ao,r_Am,r_AoE and r_AmE with LLAC while the correlations between LAC and EIAC or EAC was mostly negative. For LLAC except for significant negative r_Aoor significant positive r_AmE with EIAC, and negative r_Ao and r_Do or positive r_Am and r_AoE with EAC the others were not significant. Between EIAC and EAC the r_Ao, r_C and r_Am values were significant positive which meant that the genes expression from cytoplasmic and embryo or maternal genetic system could improve the performance of both EIAC and EAC in mature rapeseeds.7.The prediction studies for 9 parents showed that there were large differences in genetic main effects and GE interaction effects at all 5 developmental times across environments. Great instability with regards to environment was noted among parents for almost all traits at 15d and for some traits at 22d. Regarding selection of the better parents, P4 (Gaoyou 605) and P8 (Tower) for KSW, P1 (Youcai 601) and P6 (Eyouchangjia) for OC and P2 (Double 20-4) for LAC, could be utilized in rape breeding programs because of their higher stable positive values at most development times. For improving OAC, except P1, P5 (Zhongyou 821) and P6, all other parents could be used while P1 for PAC, P3 (Huashuang 3) and P9 (Zheshang 72) for EIAC, P6 and P8 for LLAC and P3 for EAC could be used in rape breeding programs for developing more stable and nutritionally beneficial oils.