Establishment Of Calibration Models Of Nirs Analysis For Cottonseed Oil And Protein Contents、Molecular Tagging、Location Of Qtls In Upland Cotton

Cottonseed is a by-product of cotton production, and the seed kernel accounts for about60%of cottonseed weight, with abundant protein and oil content. It is an important edible blend oil in plant oil industry. But for a long time, cotton breeders pay little attention on improvement of cottonseed quality, cotton breeding over the past decades is just a natural selection process for quality traits of cottonseed. Conducting research on the comprehensive utilization of cotton by-product and developing new cultivars used as fiber and edible oil is an important means to chang cotton production from pure fiber production to comprehensive utilization of cotton production.Cottonseed nutritional quality traits are controlled by many genes, so it is restricted by the traditional breeding method which is through the individual phenotypic traits selection. Because of complex relationship among cotton fiber yield, fiber quality and seed nutritional quality, it is difficult to combine all these traits together in cotton breeding programm. In this study, we conducted a series of research on cottonseed nutritional quality traits and laid some foundation of cotton nutritional quality traits improvement.1. Analysis of correlation on seed nutrient quality traits and main yield traits, fiber quality traits in upland cotton.One hundred and one upland cotton germplasms were selected to measure their protein and oil content of cottonseed, lint percentage, boll weight and fiber quality traits. Correlation analysis were conducted on some major agronomic traits, seed nutritional quality and fiber quality traits by SPSS11.0computer programm and Microsoft Excel2003The results showed that protein and oil content of upland cotton had great genetic variation, but the total amount of protein and oil showed a lower variability, and both are highly negative correlation. So it is difficult to improve both oil and protein content simultaneously. Oil content was negativelv related to uniformity, fiber lensth, fiber strength, micronaire and fiber elongation; Protein content was positively related to fiber length, fiber uniformity, micronaire, fiber elongation and strength.2. Establishment of calibration models of NIRS analysis for cottonseed oil and protein contentsA total of112cotton lines were analyzed on oil content by soxtec method. Eighty seven samples were selected to establish math model of oil content, and the remained25samples were used to test this model. NIRS predicted values were consistent significantly with chemical analysis results which was supported by a value of R=0.9656and SEP=3.55%. A total of103samples were analyzed on protein content by Micro-Kjeldaha method. Eighty samples were chose to construct math model of protein content, and the remained23samples were used to test the model. NIRS predicted values were consistent significantly with chemical analysis results which was supported by a value of R=0.9727and SEP=3.56%. These results showed that the NIRS methods established in this study have relatively high accuracy in prediction test for cottonseed protein and oil content.3. Molecular mapping of QTLs for cottonseed oil and protein contents in upland cottonTwo mapping populations were developed by using two cultivars with high oil and low protein content (DPLSP3and Humupedixil4wstr5) and two cultivars with low oil and high protein content(Yancheng115and Yancheng90-110). The population from DPLSR3and Yancheng115was named as PopA, and the one from Humcopedixil4wtr5and Yancheng90-110as PopB.. A total of1221SSR markers covering the whole cotton genome were selected to screen polymorphism between parents. In Pop A,52loci showed polymorphic, among which27loci were mapped into9linkage groups (8chromosomes). The genetic maps spanned364.2cM, approximately7.2%of the total recombinational length with an average distance of12.7cM between adjacent markers. In Pop B,49loci showed polymorphic, among which18loci were assigned to8linkage groups (8chromosomes), spanning192.8cM, approximately3.8%of the total recombinational length with an average genetic distance of10.7cM between adjacent markers.By composite interval mapping,.4QTLs for cottonseed nutrient quality,2for oil content and2for protein content, were identified, qOC-13-1(Pop A)、qPC-19-1(Pop A). qOC-19-1(Pop B)、qPC-19-1(Pop B).Among them, qOC-13-1(Pop A) showed good stability, which was identified between the SSR maker DC20120and NAU2893on Chr.13 in both F2and F3generations, explaining9.21%and12.01%of the phenotypic variations, respectively.By single marker analysis, other4QTL were identified on chromosome7(protein content QTL in PopA), chromosome12(oil content QTL in PopB), chromosome12(protein content QTL in PopB), and chromosome15(protein content QTL in PopB).