| Soybean,whose annual consumption exceeds 100 million tons,is one of the most important oil crops in China.In 2017,China imported more than 95 million tons of soybeans,accounting for more than 80%of the total consumption.In recent years,a serious imbalance in the supply and demand of soybeans has caused a great threat to China’s food security.In addition,to obtain high yields in production,excessive fertilization is applied,which not only increases the production costs,but also causes many problems such as environmental pollution.Nitrogen fertilizer is the largest fertilizer used in agricultural production,and it is also a major factor causing pollution to the agricultural environment.Therefore,how to reduce the nitrogen input and maintain the output with the reducing fertilization have become the hotspots and difficulties in the study of plant nutrition.Biological nitrogen fixation is the process by which nitrogen-fixing microorganisms reduce nitrogen in the atmosphere to form ammonia.The nitrogen fixed by microorganisms is a clean and environmentally friendly nitrogen source.Soybean is a leguminous crop with high nitrogen fixation efficiency and plays an irreplaceable role in the agricultural nitrogen cycle system.Therefore,excavating high-yielding loci/genes and increasing soybean yields are of great significance in reducing China’s dependence on imported soybeans.At the same time,increasing the biological nitrogen fixation efficiency of soybeans and reducing the amount of nitrogen fertilizer application can provide new ideas for increasing the agricultural production efficiency.In this study,we use cultivars Jidou 12,Ji nf58 and their 175 derived F9:11 recombinant inbred line（the RIL）population as materials for the construction of new soybean genetic map.Through the treatments with or without inoculation,the effects of rhizobium on the agronomic characters and soybean yields were investigated and the QTLs controlling these agronomic characters were identified.The key results are showed in the following:（1）Genetic analysis of 19 agronomic characters revealed that:Except for stem weight,the heritabilities of all agronomic characters was high,ranging from 0.46 to0.98.With inoculation treatments,the highest heritability is found in plant height,which is 0.98,while the lowest heritability is found in stem weight,which is only 0.46.The coefficient of variation of each trait was between 4 and 113%.The lowest coefficient of variation was found in the protein content,which was only4%in both treatments.The highest coefficient of variation was up to 113%,which was found in the dry weight of small root nodules under non-inoculation condition.The analysis of kurtosis and skewness for each trait revealed that except for stem weight with inoculation,the absolute values of kurtosis and skewness of other traits were all less than 1,which obey a normal distribution and could be used for subsequent QTL mapping analysis.（2）Correlation analysis showed that:A total of 15 traits were significantly and positively correlated with yield,with the correlation coefficient ranging from 0.31 to0.90.Among them,the correlation between grain number per plant and root dry weight was the highest.In addition,there was a significant negative correlation between protein content and yield,with a correlation coefficient of-0.17.There was a very significant negative correlation between protein content and oil content.The protein and oil contents were significantly associated with six and eight agronomic traits,respectively,with correlation coefficients between-0.41 and 0.27 and-0.38 and 0.29.The number of large root nodules and protein content was positively correlated and the correlation between them are the highest.In contrast,the number of large root nodules was negatively correlated with the oil content and the correlation between them are the lowest..Interestingly,in the respect of grain number,the highest negative correlation was found between grain number and protein content,but the highest positive correlation was found between grain number and oil content.（3）Rhizobium inoculation significantly affected major agronomic traits in soybean.The inoculation of high-efficiency rhizobia significantly increased the biomass of soybean plants,indicating by a significant increase in leaf dry weight,stem weight,and root dry weight after inoculation,with increases of 10.7%,21.3%,and 7.9%,respectively.Inoculation with highly efficient rhizobia significantly altered the soybean root architecture.After inoculation,the soybean roots became shallower,and the root dry weight and some root morphology indicators（root length and root surface area）increased significantly.（4）The present study build a new genetic map using 133 pairs of SSR markers with polymorphism between the parents.The total length of the map is 2069.77 cM,which covers more than 60%of known public soybean linkage maps on SoyBase（https://soybase.org）.Each linkage group covers a genome range of 28.7 cM to 163.8cM,and the average distance between each marker is 15.56 cM.Position and distance of most of the SSR in our genetic map were consistent with that reported in the public soybean genetic linkage map in SoyBase.（5）The results of QTLs analysis showed that 87 QTLs were identified,which distributed on 13 linkage groups.For each trait,an average of 4.35 QTLs were detected.Among them,a total of 23 QTLs were consistently detected under both inoculated and non-inoculated condition.Without inoculation,a total of 39 QTL loci was detected distributing on 10 linkage groups including D1a,N,C2,O,B1,F,B2,D2,L and I.Under inoculation condition,a total of48 QTL were detected,which are distributed on10 linkage groups including D1a,C2,M,O,D2,B1,N,H,E.The LOD values of all the QTL loci detected ranged from 2.5 to 18.86.The minimum LOD value was detected in pod number q-PN-O,which was 2.5,while the maximum LOD value was detected in root dry weight of q+RW-C2,,which was 18.86.Genetic contribution of all the QTL loci detected ranged from 3.39%to 28.98%.The minimum genetic contribution ratio was 3.39%,which was found in the plant height q+PH-1-N-1,,while the maximum genetic contribution ratio was28.98%,which was found in the branches number q+BN-C2,.Moreover,all the QTLs are unevenly clustered on linkage groups.Among them,82.75%of the QTLs are distributed on four linkage groups C2,O,B1,and L,which consist of 29,27,8 and 8 QTLs,respectively.The results suggested that one gene or tightly linked multigenes might be involved in controlling these traits.In summary,there are complex associations among soybean agronomic traits.Although inoculation of rhizobia had certain influences on the major agronomic traits of soybean,these agronomic traits still had high heritabilities,indicating that these agronomic traits were mainly controlled by genetics.Thus,the genetic characters of different agronomic traits could be used for the identification of QTLs,which controls soybean yields and could be applied to improve and breed soybean new varieties.At the same time,the present study identified some new QTLs,which not only provided genetic basis for subsequent fine mapping of related traits,but also provided theoretic basis for soybean genetic breeding. |
