**4. Discussion**

### *4.1. Effect of AMF on Plant C:N:P Stoichiometry*

Previous studies demonstrated that terrestrial carbon and nitrogen pools can be significantly stimulated by experimental N input [43,44], which could be partly explained by warming-induced increases in net soil N mineralization and nitrification rates [45]. The significant effect of N input on plant C:N and N:P ratios may be attributed to higher soil N availability, which stimulates plant growth [23]. However, our results indicated that AMF alter different plant C:N:P stoichiometries under warming and nitrogen input in a semiarid meadow.

Our previous study has shown that under warming treatment, the mycorrhizal benefits increased by 374.4% for the aboveground biomass of *S. viridis* [35]. In the present study, under warming treatment, AMF significantly increased the P contents and the C:N ratio of *S. viridis*. AMF may increase plant biomass by promoting nutrient cycling [31].

In the N input treatment of our previous study [35], the mycorrhizal benefits conferred to the aboveground biomass of *S. viridis* and *L. chinensis* were increased by 51.1% and 47.4%, respectively, whereas the aboveground biomass of *S. corniculata* decreased significantly under both treatments. These results are generally in agreemen<sup>t</sup> with those of the present study. In the present study, under N input and a combination of warming and N input, AMF significantly enhanced the C and P contents and the C:N ratio of the C4 grass *S. viridis*. Under N input, AMF significantly increased the C, N, and P contents and the C:N ratio of the C3 grass *L. chinensis.* However, AMF significantly decreased the C, N, and P contents of *S. corniculata* under the combination of warming and N input. The mycorrhizal benefits conferred to C4 grasses are greater than those conferred to C3 grasses and should therefore result in higher carbon production and increased AMF development [41,46], an expectation consistent with the findings of the current study. However, in *S. corniculata*, AMF symbiosis was antagonistic. The results sugges<sup>t</sup> that the contribution of AMF to the three dominant species in the Songnen meadow steppe varies under global change. These results support our hypothesis and further sugges<sup>t</sup> that plant stoichiometric responses to global change and ecosystem stability can be adjusted by AMF.

N and P are the most common limiting elements for plant growth and have profound impacts on plant functions [30]. According to some studies, AMF can transfer a considerable amount of N from the soil to host plants [47,48], but other studies have found no evidence that AMF symbioses increase N uptake [49,50]. Phosphorus is the most readily immobilized element in the soil, and its availability is very low [51]; therefore, mycorrhizal P uptake is the dominant pathway [52]. AMF form extensive hyphal networks in the soil and forage effectively for nutrients, especially P, which is supplied to their host plants [10,53]. Fungal nutrient allocation is adjusted through the carbon source strength of individual host plants [54], and plant species affect the AMF response to resource stoichiometry [55].

Furthermore, under all treatments, AMF significantly decreased the N:P ratio of *S. viridis* under N input and a combination of warming and N input, and AMF significantly decreased the N:P ratios of *L. chinensis* and *S. corniculata*. Several studies have suggested that a plant N:P ratio < 14 indicates N limitation and that a plant N:P ratio > 16 indicates P limitation [56,57]. Numerous studies have shown that N input induces an imbalance in the N:P ratio and an increase in P limitation in grasslands [58,59]. In the present study, N input and the combination of warming and N input significantly enhanced the plant N:P ratios of the three species, which may result in an altered balance between N and P. Therefore, the Songnen meadow steppe ecosystem changed from being N limited to being P limited, which is in accordance with an experiment in a temperate steppe ecosystem [60]. However, AMF significantly reduced the N:P ratios of the three species under N input and the combination of warming and N input, which agrees with a previous result [61]. The trade balance model predicts that N enrichment of a P-limited soil will exacerbate the P limitation and increase the amount of P obtained through symbiosis [30]; thus, the plant N:P ratio will decrease in the presence of AMF. The results sugges<sup>t</sup> that AMF might slow the increase in P limitation caused by global change in Songnen meadows. Rational

managemen<sup>t</sup> of soil nutrients in these meadows is critically important to increase plant productivity and to improve the sustainable utilization of grassland ecosystems.

### *4.2. Effects of AMF on Soil C:N:P Stoichiometry*

Stoichiometry is a vital indicator of biogeochemical cycles in terrestrial ecosystems [62]. Soil C:N:P stoichiometry provides a crucial potential diagnostic value for nutrient mineralization and organic matter decomposition [63]. Studies have revealed that C:N:P stoichiometry in soil and plants is tightly linked [64,65]. Soil C:N:P stoichiometry not only regulates microbial activity but also plant N and P uptake [64], while plant C:N:P stoichiometry directly reflects the availability of soil N and P. Many studies have suggested that AMF may transport large numbers of limiting nutrients (N, P) to their host plants from the soil [48,66] and may promote nutrient use efficiency by accelerating the decomposition of organic matter [48,67].

Yue et al. [23] found that high stoichiometric homeostasis, measured as the soil C:N ratio, decreased significantly under N addition. In the present study, under N input, the soil N content and the N:P ratio were decreased significantly in the presence of AMF, but the soil C:N ratio under N input and AMF increased, suggesting that AMF play vital roles in soil nutrient cycling [68], including C and N cycling, in grassland ecosystems [45]. The results indicate that AMF might improve the soil stability and sustainability of plant-soil systems.
