To study the relationships between C4 enzyme activities and yield, C4 enzyme activities (phosphoenolpyruvate carboxylase (PEPCase), NADP-malate dehydrogenase (NADP-MDH), NADP-malic enzyme (NADP-ME), and pyruvate phosphate dikinase (PPDK)) in different organs of ten soybean cultivars with different yields were measured at different growth stages in China. The result showed that four enzyme activities in C4 pathway were obviously different among cultivars, especially PPDK activity was not detected in the leaves of Dongnong 1567 and Dongnong 1068 and the young leaves of Gongjiao 9107-1 and Dongnong 97-172, but there were weak activities in pod coats. The order of C4 enzyme activities is young leaves < old leaves < pod coats. The correlation coefficients between PEPCase activity and yield and between NADP-MDH activity at blooming stage and yield were 0.6979 and 0.6565, respectively, and both reached the significant level (5%), and PEPCase activity kept significant positive correlation with plant photosynthetic rate. There was a negative correlation between NADP-ME activity and yield, and no correlation was found between PPDK activity and yield.

Soyasaponins are valuable compounds in certain drugs, industry, food additives and surfactants. Selecting cultivars with higher-soyasaponin content along with agronomic traits is a main goal for many soybean breeders. The aim of the present study was to identify the quantitative trait loci (QTLs) associated with total soyasaponin content through a F2 population, which was derived from a cross between Ha 91016 (higher soyasaponin content cultivar, 16.8 mg g-1) and N98-9445A (lower soyasaponin content, only 5.7 mg g-1). A genetic linkage map including a total of 162 simple sequence repeat markers was constructed, which covered the total length 2 735.5 cM, and the average distance between markers was 16.96 cM. Two QTLs associated with total soyasaponin content were identified. One, qSAP_1 (located in sat_044-satt102 of linkage group (LG) K), could explain 12.6% of phenotypic variance. The other, qSAP_2, was located between satt368 and sat_413 of LG D1a, which could explain 15.8% of phenotypic variance. It was concluded that the two QTLs would have some potential value for marker-assisted selection for high-soyasaponin content breeding in soybeans.