Rhizosphere priming effects of Lolium perenne and Trifolium repens depend on phosphorus fertilization and biological nitrogen fixation

Live roots can stimulate microbial soil organic matter (SOM) decomposition and nutrient cycling, which is termed as the rhizosphere priming effect (RPE). Compared to nitrogen (N) availability, fewer studies have focused on the effect of phosphorus (P) availability on the RPE. Here we investigated the RPEs of ryegrass ( Lolium perenne ) and clover ( Trifolium repens ) with and without P fertilization (4 g P m −2 ) at three sampling times (Day 30, Day 44, and Day 58 after planting). A continuous 13 C–CO 2 labeling method was used to separate soil-derived CO 2 from root-derived CO 2 . A nutrient budget method was applied to evaluate the rhizosphere effect on net soil N and P release for plant uptake. We found that ryegrass and clover induced positive RPEs in most plant-soil combinations, ranging from −1% to 134%. Ryegrass exhibited a larger RPE than clover by Day 30, but clover exhibited a larger RPE than ryegrass by Day 44 and Day 58, possibly due to larger shoot biomass regrowth rates, root activity, and rhizodeposition during the later stages. P fertilization significantly decreased the RPE of ryegrass by Day 44 and Day 58, but did not change the RPE of ryegrass by Day 30 and clover at all three sampling times. The reduced RPE of ryegrass with P fertilization was associated with increased microbial biomass N, more root-derived microbial C, and less shoot biomass and root-derived CO 2 . These findings suggest that P fertilization coupled with C supply from root exudates induced more microbial N immobilization, which reduced the RPE of ryegrass during later stages when soil N limitation negatively impacted plant growth. However, P-induced microbial N immobilization did not affect clover as much because its biological N fixation, on average 37% of total plant N, may have alleviated soil N limitation. We further observed significant positive relationships between excess net soil N and P release and the RPE by Day 58 across all planted treatments, indicating that soil N and P release by plants can be directly linked to rhizosphere C mineralization. Overall, our results demonstrate the importance of C–N–P interactions for understanding the RPE, which have significant implications for P cycling in plant-soil systems. • Without P fertilization, rhizosphere priming effect increased soil C, N and P release. • P fertilization decreased the rhizosphere priming of grass on soil C and N release when N was limiting, but not for legume. • In the presence of plants, P fertilization plus rhizodeposits enhanced microbial N immobilization. • Biological N fixation by clover nodules alleviated P-induced soil N limitation. • C–N–P interactions regulated the rhizosphere priming effect.