Stoichiometric homeostasis of N:P ratio drives species-specific symbiotic N fixation inhibition under N addition

Abstract

Introduction: Symbiotic N fixation inhibition induced by N supply to legumes is potentially regulated by the relative N and P availability in soil. However, the specific responses of different legume species to changes in N:P availability remain unclear, and must be better understood to optimize symbiotic N fixation inputs under N enrichment. This study investigated mechanisms by which soil N and P supply influence the symbiotic N fixation of eight legume species, to quantify the inter-specific differences, and to demonstrate how these differences can be determined by the stoichiometric homeostasis in N:P ratios (H_N:P).

Methods: Eight herbaceous legume species were grown separately in outdoor pots and treated with either no fertilizer (control), N fertilizer (14 g N m^-2), P fertilizer (3.5 g P m^-2) or both N and P fertilizer. Plant nutrients, stoichiometric characteristics, root biomass, non-structural carbohydrates (NSC), rhizosphere chemistry, P mobilization, root nodulation and symbiotic N fixation were measured.

Results: N addition enhanced rhizosphere P mobilization but drove a loss of root biomass and root NSC via exudation of P mobilization compound (organic acid), especially so in treatments without P addition. N addition also induced a 2-14% or 14-36% decline in symbiotic N fixation per plant biomass by legumes in treatments with or without P addition, as a result of decreasing root biomass and root NSC. The changes in symbiotic N fixation were positively correlated with stoichiometric homeostasis of N:P ratios in intact plants without root nodules, regardless of P additions.

Discussion: This study indicates that N addition can induce relative P limitations for growth, which can stimulate rhizosphere P mobilization at the expense of root biomass and carbohydrate concentrations, reducing symbiotic N fixation in legumes. Legume species that had less changes in plant N:P ratio, such as Lespedeza daurica and Medicago varia maintained symbiotic N fixation to a greater extent under N addition.

Keywords: N:P ratio; herbaceous legume; phosphorus mobilization; rhizosphere; stoichiometry; symbiotic N fixation.

Figure 1

Available nitrogen (AN, A) and available phosphorus (AP, B) concentrations and AN : AP (C) in bulk soil under control (no fertilizer addition), N (N addition), P (P addition) and NP (addition of both N and P) for eight legume species. MS, Medicago sativa; MV, Medicago varia; MF, Medicago falcata; MR, Medicago ruthenica; LC, Lotus corniculatus; MO, Melilotus officinalis; LD, Lespedeza daurica; GS, Glycine soja. Letters (N, P, S, N×P, N×S, P×S, N×P×S) indicate significance overall from N addition, P addition, legume species (S) and their interaction effects according to general linear model P-values, the significance level are reported as ***(P<0.001). Inside the embedded figures, the bars represent the means of each independent factor and interacted treatment with significant effect.

Figure 2

P mobilization effect in rhizosphere (ratio of AP in the rhizosphere to that in the bulk soil) (A) and available P (AP) concentration in rhizosphere soil (B) under control (no fertilizer addition), N (N addition), P (P addition) and NP (addition of both N and P) for eight legume species. MS, Medicago sativa; MV, Medicago varia; MF, Medicago falcata; MR, Medicago ruthenica; LC, Lotus corniculatus; MO, Melilotus officinalis; LD, Lespedeza daurica; GS, Glycine soja. Letters (N, P, S, N×P, N×S, P×S, N×P×S) indicate significance overall from N addition, P addition, legume species (S) and their interaction effects according to general linear model P-values, the significance level are reported as ***(P<0.001), **(P<0.01), *(P<0.05). Inside the embedded figures, the bars represent the means of each independent factor and interacted treatment with significant effect.

Figure 3

pH (A), citric acid concentration (B), malic acid concentration (C) and ${\text{NO}}_3^{-}$ / ${\text{NH}}_4^{+}$ ratio (D) in rhizosphere soil under control (no fertilizer addition), N (N addition), P (P addition) and NP (coupling addition of N and P) for eight legume species. MS, Medicago sativa; MV, Medicago varia; MF, Medicago falcata; MR, Medicago ruthenica; LC, Lotus corniculatus; MO, Melilotus officinalis; LD, Lespedeza daurica; GS, Glycine soja. Letters (N, P, S, N×P, N×S, P×S, N×P×S) indicate significance overall from N addition, P addition, legume species (S) and their interaction effects according to general linear model P-values, the significance level are reported as ***(P<0.001), **(P<0.01), *(P<0.05). Inside the embedded figures, the bars represent the means of each independent factor and interacted treatment with significant effect.

Figure 4

Leaf photosynthetic rate (A), total biomass (B), shoot biomass (C), root biomass (D), shoot (E) and root non-structural carbohydrate (NSC) concentration (F) under control (no fertilizer addition), N (N addition), P (P addition) and NP (addition of both N and P) for eight legume species. MS, Medicago sativa; MV, Medicago varia; MF, Medicago falcata; MR, Medicago ruthenica; LC, Lotus corniculatus; MO, Melilotus officinalis; LD, Lespedeza daurica; GS, Glycine soja. Letters (N, P, S, N×P, N×S, P×S, N×P×S) indicate significance overall from N addition, P addition, legume species (S) and their interaction effects according to general linear model P-values, the significance level are reported as ***(P<0.001), **(P<0.01), *(P<0.05). Inside the embedded figures, the bars represent the means of each independent factor and interacted treatment with significant effect.

Figure 5

Plant N concentration (A), plant P concentration (B), plant N:P ratio (C) under control (no fertilizer addition), N (N addition), P (P addition) and NP (addition of both N and P), and stoichiometric homeostasis of plant N:P (H_N:P, d) for eight legume species. MS, Medicago sativa; MV, Medicago varia; MF, Medicago falcata; MR, Medicago ruthenica; LC, Lotus corniculatus; MO, Melilotus officinalis; LD, Lespedeza daurica; GS, Glycine soja. Letters (N, P, S, N×P, N×S, P×S, N×P×S) indicate significance overall from N addition, P addition, legume species (S) and their interaction effects according to general linear model P-values, the significance level are reported as ***(P<0.001), **(P<0.01), *(P<0.05). Inside the embedded figures, the bars represent the means of each independent factor and interacted treatment with significant effect. Solid triangles without a common lowercase letter differed according to Duncan’s multiple comparison test (P<0.05).

Figure 6

Root nodule number (A), root nodule biomass (B), root nodule P concentration (C) and specific symbiotic N fixation (D) under control (no fertilizer addition), N (N addition), P (P addition) and NP (addition of both N and P) for eight legume species. MS, Medicago sativa; MV, Medicago varia; MF, Medicago falcata; MR, Medicago ruthenica; LC, Lotus corniculatus; MO, Melilotus officinalis; LD, Lespedeza daurica; GS, Glycine soja. Letters (N, P, S, N×P, N×S, P×S, N×P×S) indicate significance overall from N addition, P addition, legume species (S) and their interaction effects according to general linear model P-values, the significance level are reported as ***(P<0.001), **(P<0.01), *(P<0.05). Inside the embedded figures, the bars represent the means of each independent factor and interacted treatment with significant effect.

Figure 7

Correlation relationships between stoichiometric homeostasis of plant N:P (H_N:P) of legume species and its change in pH (A), citric acid concentration (B), malic acid concentration (C) or ${\text{NO}}_3^{-}$ / ${\text{NH}}_4^{+}$ ratio (D) in rhizosphere under N addition with or without P addition.

Figure 8

Correlation relationships between stoichiometric homeostasis of plant N:P (H_N:P) of legume species and its change in leaf photosynthetic rate (A), root biomass (B), root non-structural carbohydrate (NSC) concentration (C), shoot biomass (D), shoot non-structural carbohydrate (NSC) concentration (E) and %Ndfa (F) under N addition with or without P addition.