Genetic Basis Of Introduced Yugoslavia Maize Inbred Lines And Prediction Of Heterosis Using SSR Markers

Exotic germplasms have been playing an important role in corn breeding and production in our country. But to make good use of them, their genetic characteristics and germplasm relationship with Chinese inbred lines representing different heterotic groups should be studied. In this study, through the traditional methods, analysis of combining ability and heterosis, and modern SSR marker technique, the combining ability, heterosis, genetic distance and heterozygosity between introduced Yugoslavia inbreds and important lines representing different heterotic groups in our country were studied, which will greatly influence to make full use of those indroduced lines. Through all the parameters and their correlations with F1 grain yield, their abilities and results in heterosis prediction were also studied. Nine introduced Yugoslavia inbreds and 7 Chinese inbreds, representing different heterotic groups, were chosen to make 63 crosses using NCII genetic design. All the 63 crosses were evaluated in 3 replicated experiments with randomized completed block design at Xinxiang in 2000 and Changge in 2002, using Yuyu18 and Yuyu33 as check hybrids respectively. The combining ability and heterosis of 7 ear-kernel characters were analysed. The results showed that ZPL10 had the highest GCA and TCA, and its SCAs and the over-check heterosis with inbreds from Tangsipingtou, Lüdahonggu heterotic groups were the highest, so it had great value in direct utilization. The grain yield of ZPL10×S22 and ZPL10×Yu87-1 were greater than the two CKs by 41.32% and 24.26%, 5.54% and 23.74% respectively. For other lines,their GCAs, TCAs and over-check heterosis in ear-kernel weight were low and different in the two locations, so they might only be used to improve inbreds from the same or low heterotic groups.Genetic diversity of 27 elite maize inbred lines were evaluated by 113 SSR markers, which yielded 411 alleles. The average number of alleles per SSR locus was 3.64 with a range from 2 to 9, and the polymorphic index content (PIC) value 0.13-0.82 with an average of 0.57. The average genetic distance among 27 lines was 0.595 with a range from 0.095 to 0.719. A cluster analysis using SSR data classified the 27 inbred lines into 6 clusters, which was in agreement with pedigree analysis. All PA, PB and Tangsipingtou lines were assigned into their groups accordingly; Yugoslavia lines ZPL01, ZPL02, ZPL06, ZPL09 and ZPL10 were placed into cluster I, ZPL03, ZPL04, ZPL05 into cluster II with Mo17 and Dan232, ZPL07 into cluster III with Zong3 and Zong31. The GD and GH varied according to different Yugoslavia inbreds and inbreds from different heterotic groups in our country. Except ZPL07, the GD and GH between Yugoslavia lines and PA inbreds were all small. But the GD and GH between inbreds from cluster I and PB or Tangsipingtou, cluster II and Tangsipingtou, Lüdahonggu, PB, ZPL07 and lines from all our four heterotic groups were rather high.The correlation analysis among GD, GH, SCA, the over-check heterosis and F1 ear-kernel weight showed that there were significantly positive correlations among SCA, heterosis and F1 ear-kernel weight, but those GD, GH and F1 ear-kernel weight or over-check heterosis were significant only at Changge. So SCA and over-check heterosis could be used to predict F1 yield, but the ability of GD and GH in heterosis predication was limited and easily affected by environment. The correlation between GD and GH was significantly positive, but they had low correlation with SCA.One-way analysis of variance for 113 SSR markers showed that the positive markers of different ear-kernel characters were greatly different, which were 2.92-16.30% of the total markers. The correlations between SH and all the ear-kernel traits were almost identical, and was not significant with ear-kernel weight at two locations, indicating that only through SH heterosis could not be predicted correctly.