A novel technique utilizing enriched 15N2 to trace nitrogen transfer in grass and legume mixtures

Abstract

Legumes are a potentially important N source in pasture systems, but quantifying the transfer of biologically fixed N from the legume to the grass component is difficult. A greenhouse H-pot system was developed to directly estimate belowground N transfer from biological N₂ fixation (BNF) using ¹⁵N₂. The system was tested with ‘Prine’ annual ryegrass (Lolium multiflorum L.) and ‘Dixie’ crimson clover (Trifolium incarnatum L.). Legume and grass root systems growing in either individual or H pots were exposed to ¹⁵N₂. Control H pots were separated by mesh to prevent contact between roots from each side of the pot. To reduce enriched gas volume demand and avoid cross-contamination in the greenhouse, the gas was supplied through underground tubes in the root zone. Ryegrass and clover exhibited an enrichment of ¹⁵N₂ when their respective root systems were supplied with ¹⁵N₂. Additionally, ryegrass also showed enrichment when clover roots received the gas, provided there was direct contact between the root systems on both sides of the H pot; however, this enrichment did not occur when such contact was prevented. Plants cultivated in monoculture without the application of ¹⁵N₂ did not present enrichment. The H-pot facilitates the evaluation of belowground transmission, an essential mechanism for N transfer. The technique of gradually supplying ¹⁵N₂ directly to the root system may serve as a valuable labeling method for tracking nitrogen transfer. The absence of enrichment when plants were not directly supplied indicates negligible atmospheric enrichment. However, the enrichment observed in ryegrass when supplied with the gas suggests BNF through alternative pathways.

Introduction

Legumes enhance the provision of ecosystem services when included in mixtures with grasses in forage systems due to their association with rhizobia and consequent biological N₂ fixation (BNF)¹. The extent of BNF in grasslands is dependent on forage accumulation, the proportion of legume in the botanical composition, legume N concentration, and the proportion of N₂ that is derived from the atmosphere in the legume biomass². In grass-legume swards, N transfer from legume to companion species can occur both through belowground and aboveground processes, with plant litter and animal excreta generally considered the main pathways for this transfer^{3, 4, 5, 6–7}. Lesser-known transfer pathways include plant-associated arbuscular-mycorrhizal fungi; exudation and leakage from living nodules and roots; root cell and nodule sloughing in response to defoliation, and through root and nodule death and decay^{8, 9, 10, 11–12}.

Tracking both N₂ fixation and N transfer from legume to grasses in mixed swards has been challenging, and several techniques have been suggested. The heavy isotope technique using ¹⁵N has been used since the 1950s. The most common method for labeling is the addition of ¹⁵N-enriched fertilizer. The use of ¹⁵N-enriched N₂ gas is less common due to its cost, but its use avoids some of the well-known problems linked to using ¹⁵N-enriched fertilizers¹³. An advantage of gas is its use in an enclosed and sealed chamber, which ensures that the fixed N is coming from the labeled ¹⁵N₂ gas via BNF¹⁴. However, researchers have pointed out several issues with using ¹⁵N₂ gas, including the purity of the sealed ¹⁵N₂ gas cylinder, the costs involved in obtaining high-purity gas, and the necessity of monitoring factors such as temperature and humidity in closed systems. Consequently, most studies that utilize ¹⁵N gas as a tracer are conducted in controlled environment chambers with pots, where gas injections occur over several hours^{15, 16, 17–18}.As a result, these studies are often expensive to conduct.

To date, there have been no reported studies investigating the direct infusion of¹⁵N-enriched N₂ gas into the root zones of grasses and legumes, whether cultivated in interconnected pots (such as H-shaped configurations) or in monoculture systems. This study proposes a new strategy to address the limitations of closed chamber environments for tracking N transfer. By using pulse infusion of¹⁵N₂ directly into the plant root zone, this approach offers a more affordable alternative that can be widely applied across various grass-legume swards. The overall objective of this study was to quantify and track N transfer and validate the technique of infusing enriched¹⁵N₂ into grass and legume root zones using H pots. The hypotheses tested were: (1) N resources can be shared between grasses and legumes directly through root contact; (2) ‘Prine’ annual ryegrass (Lolium multiflorum Lam.) may associate with microorganisms capable of fixing atmospheric N₂; (3) it is possible to enrich plants by injecting¹⁵N₂ gas in the root zone; and (4) biologically fixed N exchange can be tracked in grass-legume mixtures using H pots.

Results

Experiment 1 – Single pot experiment

δ ¹⁵N, biomass, N, and N accumulation

There was an enrichment effect for δ ¹⁵N (P < 0.0001; SE = 4.78), with both species presenting greater δ ¹⁵N for the enriched treatment (42.5‰) than for the non-enriched (0.7‰) (Fig. 1). Both species exhibited greater δ ¹⁵N values when they received the ¹⁵N₂ (P < 0.001; SE = 7.10) (Table 1). In the absence of the enrichment, there was no difference in their δ ¹⁵N signature (P = 0.71; SE = 7.10) (Table 1). However, when both species received the ¹⁵N₂ injection, ryegrass showed a greater δ ¹⁵N enrichment (50.86‰) compared to crimson clover (34.24‰) (P = 0.029; SE = 7.10) (Table 1). There was a harvest × species interaction for biomass (P = 0.056; SE = 0.56). Crimson clover biomass was greater for the second harvest compared to the first (11 vs. 7 g pot^− 1). In contrast there was no difference between the two harvests for annual ryegrass (1.4 and 2.0 g pot^− 1, for 1st and 2nd harvests) (Table 3). Nitrogen concentration was affected by species (P < 0.0001; SE = 0.8) and harvest (P = 0.0026; SE = 0.8) (Table 2). Crimson clover had greater N concentration (34 g kg^− 1) than annual ryegrass (12 g kg^− 1). There was a harvest × species interaction (P = 0.0137; SE = 21.80) for N accumulation, with greater N accumulation in the second harvest for crimson clover (355 vs. 261 mg pot^− 1) but no difference for annual ryegrass between Harvest 2 and Harvest 1 (22 and 21 mg pot^− 1, respectively) (Table 3).

[See PDF for image]

Fig. 1

Enrichment effect (P < 0 0.0001) for the single pot experiment averaged across species, harvests, and replications.

Table 1. Enrichment effect for the single pot experiment within species and across species.

	Enriched, δ15N (‰)	Non-enriched, δ15N (‰)	SE	P value
Annual ryegrass	50.86	2.04	7.10	< 0.001
Crimson clover	34.24	-0.63	7.10	< 0.001
SE	7.10	7.10
P value for species	0.029	0.71

Table 2. Species effect and harvest effect for N concentration in the single pot experiment.

Species	N (g kg− 1)
Crimson clover	34
Annual ryegrass	12
SE	0.8
P value	< 0.0001
Harvest	N (g kg− 1)
1	25
2	22
SE	0.8
P value	0.0026

Table 3. Harvest × species interaction for biomass in the single pot experiment.

Harvest	Species	Biomass (g pot− 1)	N accumulation (mg pot− 1)
1	Crimson clover	7 B*	261B*
	Annual ryegrass	1.4 C	21C
2	Crimson clover	11 A	355A
	Annual ryegrass	2 C	22C
SE		0.56	21.80
P value		0.0056	0.0137

*Means within a column followed by a common uppercase letter do not differ (P ≥ 0.05) according to LSD.

Experiment 2 – H-pot experiment

Herbage δ ¹⁵N

The ¹⁵N₂ supply to roots induced (P < 0.0001; SE = 2.03) δ ¹⁵N enrichment for annual ryegrass, with the most enriched δ ¹⁵N in treatments HCE and HGE (23.0 and 24.4‰, respectively) and the most depleted for treatments H (1.4‰) and HMCE (2.4‰) (Fig. 2A). There was also a membrane effect (P < 0.0001) when contrasting treatment HCE versus HMCE.

[See PDF for image]

Fig. 2

Treatment effect (P < 0.0001) for δ ¹⁵N in the ryegrass (a) and crimson clover (b) in H-pots. Treatments were: H-pot with the clover side enriched (HCE); H-pot with the grass side enriched (HGE); H-pot without enrichment (H); H-pot with the clover side enriched containing a membrane (HMCE). Means followed by a common uppercase letter do not differ (P ≥ 0.05) according to LSD.

There was a treatment effect for δ ¹⁵N in the crimson clover (P < 0.0001; SE = 1.54). Treatments HCE (21.2‰) and HMCE (26.8‰) were the most enriched followed by treatments HGE (4.2‰) and H (-0.4‰) (Fig. 2B). There was also an injection side effect (P < 0.0001) when comparing treatments HCE and HGE, but as expected, there was no effect of the membrane on the clover δ ¹⁵N in the HMCE treatment.

Biomass, N, N accumulation, and non-legume N derived from the transfer of BNF

There was no treatment effect for any of the response variables, but total biomass was greater in the second than the first harvest for both annual ryegrass (P = 0.0035; SE = 0.4) and crimson clover (P < 0.0001; SE = 1.2). Annual ryegrass had a harvest effect for N concentration (P < 0.0001; SE = 1) and N accumulation (P = 0.0405; SE = 9), while N concentration was greater in the first harvest (19 g kg^− 1) than in the second (16 g kg^− 1; Table 4). The greater biomass in the second harvest resulted in greater N accumulation (73 mg pot^− 1) than in the first (46 mg pot^− 1). There was no effect of harvest on crimson clover N concentration, but the greater biomass in the second harvest led to greater N accumulation in the second harvest (Table 4). In the HCE and HGE treatments demonstrated greater proportions of non-legume nitrogen transfer originating from soil BNF at 3.22% and 4.11%, respectively (P = 0.0012; SE = 0.69). However, in the presence of the membrane, the transfer reduced to 0.32% (Fig. 3).

Table 4. Nitrogen concentration and N accumulation of annual ryegrass and Crimson in two harvests in H pots.

Harvest	Annual ryegrass N, g kg− 1	Crimson clover N, g kg− 1	Annual ryegrass N accumulation, mg pot− 1	Crimson clover N accumulation, mg pot− 1
1	19	37	46	406
2	16	36	73	586
SE	1	1.3	9	41
P value	< 0.0001	0.2635	0.0405	0.0002

[See PDF for image]

Fig. 3

Treatment effect (P = 0.0012) for the proportion of non-leg N transfer derived from BNF (%). Treatments were: H-pot with the clover side enriched (HCE); H-pot with the grass side enriched (HGE); H-pot with the clover side enriched containing a membrane (HMCE). Means followed by a common uppercase letter do not differ (P ≥ 0.05) according to LSD.

Discussion

Experiment 1 – Single pot

The enrichment in δ ¹⁵N observed in the aboveground biomass for both species when the respective root systems received ¹⁵N₂ gradually in the root zone confirms the effectiveness of this technique to track N even when the legume and the grass were growing separately. Dilution was observed when infusing the ¹⁵N₂ in crimson clover root zone and therefore δ ¹⁵N present in the biomass was more depleted than the ryegrasss. Chalk and Craswell¹⁹ discuss challenges related with costs of ¹⁵N_2, as well problems related with leakage in closed chamber systems, for studies evaluating symbiotic nitrogen fixation. Among others, the main challenges related to closed chamber systems are related to the difficulties in keeping plants alive for long periods of time¹⁹. The pulse injection of ¹⁵N₂ directly into the root zone, as adopted in this experiment, provides an alternative method to closed chamber systems for tracking N transfer, which is beneficial for longer-term studies.

In this study, it is possible that certain bacteria associated with annual ryegrass may have a correlation with the presence of δ ¹⁵N in the biomass²⁰. Labeled N₂ must first undergo reduction before plant assimilation. Additionally, enrichment is achieved at a significantly lower cost due to the much-reduced ¹⁵N₂ demand in comparison with atmospheric liberation of this gas. Although supplying ¹⁵N₂ to the plant would generally be considered the gold standard for BNF estimates, this is usually unfeasible due to cost and logistical requirements¹⁴ which this experimental setup minimizes. At the same time, the lack of enrichment in crimson clover not supplied with the gas also indicates that atmosphere enrichment in the greenhouse was likely minimal, even though the experimental setup did not block gas transfer between soil and atmosphere.

Experiment 2 – H-pot experiment

When ¹⁵N₂ was supplied to crimson clover in treatments HCE and HMCE, annual ryegrass accumulated ¹⁵N in its biomass in HCE but not on HMCE, indicating that the 60-µm mesh membrane significantly reduced ¹⁵N transfer between the compartments, likely because it prevented root contact between the two forage species.

HCME significantly reduces the transfer of non-legume N derived from soil BNF. This suggests a mechanism of transfer associated with roots of the two species growing in proximity to each other. This could be explained by direct connection²¹bridging by mycorrhizae hyphae^{22, 23–24}through root exudates, especially at early growth stages²⁵ or the transfer of ¹⁵N via free-living soil microorganisms^9,26. Trannin et al.²⁷ quantified that half of the N in signalgrass (Urochloa decumbens Stapf. R) originated from the legume stylosanthes [Stylosanthes guianensis (Aubl.) Sw.], occurring mainly through root decomposition, with mycorrhiza playing a key role. The enrichment observed for HGE, where the gas was directly supplied to annual ryegrass roots, indicates BNF either by free-living diazotrophs²⁸ or by associative microorganisms already found in annual ryegrass²⁰. Biological N₂ fixation has been reported in other grass and cereal species²⁹. Dobereiner³⁰ initially reported BNF within bahiagrass (Paspalum notatum Flügge) and estimated that its association with Bradyrhizobium paspali was responsible for ~ 6 kg N ha^− 1 yr^− 1 across bahiagrass pastures. Santos et al.³¹ further supported these results across several bahiagrass cultivars, with quantities fixed of ~ 9 kg N ha^− 1 yr^− 1. The greater N accumulation for the second harvest is likely due to plants being better established, as also observed when harvesting annual ryegrass for silage production³².

It is pertinent to note that the current technique still has room for refinement, and both advantages and disadvantages are associated with its use. The 10-day enrichment procedure was selected due to the dearth of studies that utilized ¹⁵N₂ to track belowground N transfer in mixed swards and because the gas was directly injected into the root zone to track its distribution. However, it is essential to mention that the use of ¹⁵N₂ is relatively costly, and there is an opportunity to improve the technique by perhaps reducing the number of days and the quantity of gas used, thereby optimizing the effective tracking of N while also being mindful of resources.

In conclusion, the infusion of enriched ¹⁵N₂ at 12-cm depth gradually and directly into the root zone utilizing H-format pots can be used as a technique to track N transfer in grass-legume mixtures. This novel methodology can also be used with other stable isotopes to improve our knowledge of belowground network and resource sharing with plant and microbial communities. The H-pot technique with ¹⁵N₂ infusion directly into the root zone indicated that root contact and/or mycorrhizae network should also be considered as important pathways for belowground N transfer, although often just aboveground N transfer pathways are considered¹³. The ¹⁵N was present in annual ryegrass biomass in both H and single pots, which indicates this grass species may fix atmospheric nitrogen. Further studies are necessary to identify the soil or endophytic microorganisms responsible for this fixation.

Materials and methods

Experiment 1 – Single-pot experiment

The experiment was conducted in a greenhouse environment arranged in a randomized complete block design in a 2 × 2 factorial arrangement (grass or legume species, with or without ¹⁵N₂ enrichment) (Fig. 7) with four replicates at the University of Florida - North Florida Research and Education Center (NFREC), Marianna, FL, USA (30°51’06’’N, 85°09′ 55’’W, 35 m asl). ‘Prine’ annual ryegrass (Lolium multiflorum L.; Fig. 4, b) and ‘Dixie’ crimson clover (Trifolium incarnatum L.; Fig. 4, c) were the forage species, with treatments based on the use (or not) of enriched¹⁵N gas (98% atom ¹⁵N - Sigma-Aldrich).

A soil mixture composed of 1/3 sand (all-purpose sand - Quikrete), 1/3 potting soil (potting mix – MiracleGro), and 1/3 field soil (Orangeburg fine-loamy sand) was made for inclusion in the pots. Each pot received 3.074 kg of the soil mixture to obtain a density of 0.725 g cm^-2. Average Mehlich- I extractable soil P, K, Mg, and Ca concentrations from air-dried soil sample were 68, 307, 336, 789 mg kg^{− 1}. Estimated cation exchange capacity was 10.3 cmol kg^{- 1}, soil organic matter was 25.6 g kg^{- 1}, and soil pH in water was 6.2. The NH₄-N and NO₃-N concentrations were 7.66 and 8.8 mg kg⁻¹ respectively. The ryegrass and crimson clover were planted in 4.2-liter polyvinyl chloride (PVC) pipes (Figs. 5 and b and 7) on January 13, 2020, with each species allocated to a distinct pipe. Four seeds of ryegrass were sown in one pipe, while four seedlings of crimson clover were transferred to another pipe. The bottom of the pots was filled with 2.4 kg of gravel to facilitate water drainage.

A 0.3-cm diameter hole was drilled 16 cm from the top of each pot, into which an 18-cm long copper pipe was inserted to allow access to the root zone in the treatments receiving ¹⁵N₂ (Fig. 5, b). Holes were drilled in the copper tubes ~ 1-cm apart on their upper side to allow gas to flow directly into the root zone(Fig. 4, a, and d). To this end, the pipes were fitted with a 2-way stopcock (Smith’s Medical) attached to a 10-mL syringe (Becton Dickinson) (Fig. 5, c).

Plants were watered every other day with 300 mL of water per pot. Daily temperature ranged from 12.7 to 15.5 °C with 15–20% relative humidity, and during the night, temperatures ranged from 7.2 to 10 °C with 25 to 30% relative humidity. Each pot was sprayed with 14.8 mL of a Bifenthrin 2.4 mix (9.25 mL to 3.78 L of water) for spider mite control in March and April 2020.

[See PDF for image]

Fig. 4

Single-pot treatments: (a) crimson clover cultivated in a single pot enriched with ¹⁵N₂; (b) ryegrass grown in a single pot without enrichment; (c) crimson clover grown in a single pot without enrichment; and (d) ryegrass cultivated in a single pot enriched with ¹⁵N₂.

Two ¹⁵N₂ enrichment events took place. The first enrichment began on 16 April 2020 and the second on 7 May 2020. Every ¹⁵N₂ enrichment lasted 10 days, with 10 mL of the ¹⁵N₂ injected daily into the enriched treatments at a rate of 2 mL per hour from 10:00 to 14:00, for a total of five injections (Fig. 7, b). Both plant species were staged to a 10-cm stubble using scissors on 9 April 2020, seven days before the first enrichment started. The first harvest was on 30 April 2020, 14 days after the first enrichment, while the second harvest was on 21 May 2020, also 14 days after the second enrichment started. Herbage harvests occurred at 21-day intervals. Samples were dried in a forced-air oven at 55 °C for 72 h after the harvest, and dry weights were recorded for biomass. All samples were ground to pass through a 2-mm screen using a Wiley Mill (Model 4 Thomas-Wiley Laboratory Mill; Thomas Scientific, Swedesboro, NJ), and analyzed for δ ¹⁵N, N, and dry matter (DM) concentration.

Dry matter concentration was determined by weighing 1 g of ground forage sample into crucibles and drying it at 105 °C for 16 h. Nitrogen concentration and δ ¹⁵N were obtained through the Dumas dry combustion method using a Vario Micro Cube (Elementar, Hanau, Germany) coupled to an isotope ratio mass spectrometer (IsoPrime 100; Iso Prime, Manchester, UK) after samples were ball milled utilizing a Mixer Mill MM400 (Retsch, Haan, Germany) at 25 Hz for 9 min. Shoot N accumulation was quantified by multiplying the biomass of ryegrass or crimson clover by their corresponding N concentrations.

Experiment 2 – H-Pot experiment

The experiment was arranged in a randomized complete block design with four replicates in a greenhouse environment at the University of Florida - North Florida Research and Education Center (NFREC), Marianna, FL, USA. Sixteen-liter polyvinyl chloride (PVC) pipes connected in an H shape, referred to as H pots, were made using two PVC tees joined together with PVC coupling and medium blue PVC cement (Oatey Rain-R-Shine) to create an H shape, with PVC end caps on the bottom of each side (Fig. 5, a). A 6-cm diameter hole was drilled in the middle of the horizontal portion of each H pot to facilitate uniform watering throughout all sections of the pot (Fig. 5, a).

‘Prine’ annual ryegrass (Lolium multiflorum L.) and ‘Dixie’ crimson clover (Trifolium incarnatum L.) were the forage species, with treatments based on the use (or not) of enriched ¹⁵N gas (98% atom ¹⁵N - Sigma-Aldrich), the ¹⁵N gas infusion side (grass or legume), and the presence (or not) of a 60-µm mesh size synthetic membrane to reduce belowground interaction between grass and legume roots. The membrane was placedbehind the thin pipe used to connect both ends of the pot, giving it an H-shape format. The glued piece of membrane in the middle covers the connection piece, ensuring that both membranes were stretched and secure. These membranes covered the lateral section of the pots where the horizontal section connected, as depicted with the pink circles in Fig. 6.

[See PDF for image]

Fig. 5

H pot (a) and single-pot (b) diagram with dimensions and perforated copper pipe (c) used to deliver ¹⁵N₂ to the root zone. On each pot type, the brown dot represents the insertion point of the copper pipes.

A 0.3-cm diameter hole was also drilled, and the copper piece was perforated and inserted as described in Experiment 1 to allow access to the root zone in the treatments receiving ¹⁵N₂ (Fig. 5, a and c).

For the first treatment (HCE) ¹⁵, N₂ 98% atom was injected into the crimson clover side. In the second treatment (HGE), the ¹⁵N₂ gas was supplied to the annual ryegrass. In the third (H) treatment, none of the species received ¹⁵N₂. In the fourth (HMCE), crimson clover was supplied with ¹⁵N₂, while a 60-µm synthetic membrane was placed on the middle extremities of the pot to prevent root contact between the two species (HMCE) (Fig. 6).

[See PDF for image]

Fig. 6

H-pot treatment layout. H pot in which the crimson clover side recevied ¹⁵N₂ 98% atom (HCE); H pot in which the ryegrass side received ¹⁵N₂ (HGE); H pot in which none of the species received¹⁵N₂ (H); and H pot in which the crimson clover received ¹⁵N₂, but contained a 60-µm membrane separating the sides (HMCE).

Two enrichment events for ¹⁵N₂ injection were conducted. The first event commenced on April 16, 2020, followed by a second event beginning on May 7, 2020. Each ¹⁵N₂ enrichment lasted for 10 days, during which 10 mL of ¹⁵N₂ was injected daily into the enriched treatments at a rate of 2 mL per hour, from 10:00 to 14:00, amounting to a total of five injections (Fig. 7, b). On April 9, 2020, plants were trimmed to a 10-cm stubble using scissors, one week prior to the start of the first enrichment (Fig. 7, a). The first harvest was on 30 April 2020, 14 days after the first enrichment, while the second harvest was on 21 May 2020, also 14 days after the second enrichment started. Herbage harvests occurred at 21-day intervals. Samples were dried in a forced-air oven at 55 °C for 72 h after the harvest, and dry weights were recorded. All samples were ground to pass through a 2-mm screen using a Wiley Mill (Model 4 Thomas-Wiley Laboratory Mill; Thomas Scientific, Swedesboro, NJ), and analyzed for δ ¹⁵N, N, and dry matter (DM) concentration.

The experiment was conducted in the same greenhouse environment under the same soil, temperature, humidity, planting, and enrichment conditions previously described for the first experiment. All plants were watered every other day, on both sides of the H pot and in the middle (H connection), with a total of 750 mL of water per pot. Biomass, δ ¹⁵N, N, N accumulation, and DM were obtained similarly as described in Experiment 1.

To estimate the proportion of non-legume nitrogen derived from the transfer of BNF, a yield-dependent approach was used, involving a ¹⁵N mass balance equation (Eq. 1) as detailed by Chalk et al.³³. In this equation, ‘N’ represents the nitrogen yield of each plant species, ‘P_{non−leg (⇦atm)}’ denotes the proportion of non-legume nitrogen derived from BNF transfer, and ‘E’ signifies the ¹⁵N enrichment in atom % excess from the legume, non-legume, or the soil. This response variable specifically focuses on treatments HCE, HGE, and HMCE.

1

[See PDF for image]

Fig. 7

H-pot treatments. (a) Annual ryegrass and crimson clover plants after staging to 10 cm, one week before the start of the enrichment. (b) The ¹⁵N₂ was injected into the root zone with a syringe through a copper tube containing a valve. (c) Annual ryegrass and crimson clover plants before the harvest, 14 days after enrichment.

Statistical analysis

In both experiments, all response variables were analyzed using linear mixed model procedures as implemented in SAS PROC GLIMMIX (SAS/STAT 15.1, SAS Institute). Data from the two experiments were analyzed separately. Harvest, species, enrichment, and their interaction were considered fixed effects. Block was considered a random effect. The best covariance structure was selected based on the AICC fit statistic. Least square means were compared through pairwise t test using the PDIFF option of the LSMEANS statement. Differences were considered significant at P ≤ 0.05. For the H pot experiment, orthogonal contrasts were also performed to test for ‘membrane effect’ and ‘injection side effect’ using contrast statements.

Acknowledgements

The authors would like to extend their gratitude to the visiting scholars and all contributors who made this study possible. They also acknowledge the funding received from the United States Department of Agriculture - National Institute of Food and Agriculture, grant No. 2019- 67013-29107.

Author contributions

L.M.D.Q., J.C.B.D., M.R.M, and D.R.C. designed the experiment. L.M.D.Q., J.C.B.D., and M.R.M. built the experiment. J.C.B.D., H.L.L., and C.L.M. were responsible for the funding acquisition. J.C.B.D., J.M.B.V., L.E.S., C.L.M., M. A. L. Jr., and H.L.L provided supervision and contributed to the critical revision of the manuscript. L.M.D.Q., D.M.J., D.S.A., E.R.S.S., and M.R.M. conducted data curation and processed samples. L.M.D.Q. was responsible for preparing the initial draft and conducting the statistical analysis. All authors read and approved the final manuscript.

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding authors on areasonable request.

Declarations

Competing interests

The authors declare no competing interests.

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

1. Sollenberger, LE; Kohmann, MM; Dubeux, JCB; Silveira, ML. Grassland management affects delivery of regulating and supporting ecosystem services. Crop Sci.; 2019; 59, pp. 441-459.1:CAS:528:DC%2BC1MXhs1OrsbbP [DOI: https://dx.doi.org/10.2135/cropsci2018.09.0594]

2. Boller, BC; Nösberger, J. Symbiotically fixed nitrogen from field- grown white and red clover mixed with ryegrasses at low levels of15N-fertilization. Plant. Soil.; 1987; 104, pp. 219-226.1:CAS:528:DyaL1cXptlKhsQ%3D%3D [DOI: https://dx.doi.org/10.1007/BF02372535]

3. Dubeux, JCB; Sollenberger, LE; Mathews, BW; Scholberg, JM; Santos, HQ. Nutrient cycling in Warm-Climate grasslands. Crop Sci.; 2007; 47, pp. 915-928.1:CAS:528:DC%2BD2sXotFyru7o%3D [DOI: https://dx.doi.org/10.2135/cropsci2006.09.0581]

4. Ledgard, SF. Transfer of fixed nitrogen from white clover to associated grasses in Swards grazed by dairy cows, estimated using 15 N methods. Plant. Soil.; 1991; 131, pp. 215-223.1:CAS:528:DyaK3MXhsVaiur4%3D [DOI: https://dx.doi.org/10.1007/BF00009451]

5. Rouquette, FM; Smith, GR; Review,. Effects of biological nitrogen fixation and nutrient cycling on stocking strategies for Cow-Calf and stocker programs. Prof. Anim. Sci.; 2010; 26, pp. 131-141. [DOI: https://dx.doi.org/10.15232/S1080-7446(15)30572-6]

6. Rotz, CA et al. Whole-Farm perspectives of nutrient flows in grassland agriculture. Crop Sci.; 2005; 45, pp. 2139-2159.1:CAS:528:DC%2BD2MXht1OktbzE [DOI: https://dx.doi.org/10.2135/cropsci2004.0523]

7. Hughes, R. The Role of Nitrogen in Grassland Productivity By D. C. Whitehead Farnham Royal: Commonwealth Agricultural Bureaux pp. 202, £2.25. Exp. Agric.7, 189–189 (1970).

8. Dhamala, NR; Rasmussen, J; Carlsson, G; Søegaard, K; Eriksen, J. N transfer in three-species grass-clover mixtures with chicory, ribwort plantain or Caraway. Plant. Soil.; 2017; 413, pp. 217-230.1:CAS:528:DC%2BC28XhslGru77M [DOI: https://dx.doi.org/10.1007/s11104-016-3088-6]

9. Fustec, J; Lesuffleur, F; Mahieu, S; Cliquet, JB. Nitrogen rhizodeposition of legumes. A review. Agron. Sustain. Dev.; 2010; 30, pp. 57-66.1:CAS:528:DC%2BC3cXjs12ks7w%3D [DOI: https://dx.doi.org/10.1051/agro/2009003]

10. He, X; Xu, M; Qiu, GY; Zhou, J. Use of 15 N stable isotope to quantify nitrogen transfer between mycorrhizal plants. J. Plant. Ecol.; 2009; 2, pp. 107-118. [DOI: https://dx.doi.org/10.1093/jpe/rtp015]

11. Paynel, F; Lesuffleur, F; Bigot, J; Diquélou, S; Cliquet,. J.-B. A study of 15 N transfer between legumes and grasses. Agron. Sustain. Dev.; 2008; 28, pp. 281-290.1:CAS:528:DC%2BD1cXosVars78%3D [DOI: https://dx.doi.org/10.1051/agro:2007061]

12. Thilakarathna, MS; McElroy, MS; Chapagain, T; Papadopoulos, YA; Raizada, MN. Belowground nitrogen transfer from legumes to non-legumes under managed herbaceous cropping systems. A review. Agron. Sustain. Dev.; 2016; 36, 58. [DOI: https://dx.doi.org/10.1007/s13593-016-0396-4]

13. Russelle, M. P. Biological Dinitrogen Fixation in Agriculture. in Agronomy Monographs (eds. Schepers, J. S. & Raun, W. R.) 281–359American Society of Agronomy, Crop Science Society of America, Soil Science Society of America, Madison, WI, USA, (2015). https://doi.org/10.2134/agronmonogr49.c9

14. Measuring Plant-Associated Nitrogen Fixation in Agricultural Systems. (ACIAR, Canberra, (2008).

15. McNEILL, AM; Wood, M; Gates, RP. The use of a closed system Flow-Through enclosure apparatus for studying the effects of partial pressure of dinitrogen in the atmosphere on growth of Trifolium repens L. and Lolium perenne L. J. Exp. Bot.; 1993; 44, pp. 1021-1028.1:CAS:528:DyaK3sXltFemt78%3D [DOI: https://dx.doi.org/10.1093/jxb/44.6.1021]

16. Nitrogen Isotope Techniques. (Academic, San Diego, 1993).

17. Boddey, RM et al.et al. Dakora, FD; Chimphango, SBM; Valentine, AJ; Elmerich, C; Newton, WE et al.et al. Techniques for the quantification of Plant-Associated biological nitrogen fixation. Biological Nitrogen Fixation: Towards Poverty Alleviation Through Sustainable Agriculture; 2008; Dordrecht, Springer Netherlands: pp. 37-41. [DOI: https://dx.doi.org/10.1007/978-1-4020-8252-8_12] 42

18. Dabundo, R et al. The contamination of commercial 15N2 gas stocks with 15 N–Labeled nitrate and ammonium and consequences for nitrogen fixation measurements. PLoS ONE; 2014; 9, e110335.2014PLoSO..9k0335D [DOI: https://dx.doi.org/10.1371/journal.pone.0110335] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/25329300][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4201487]

19. Chalk, PM; Craswell, ET. An overview of the role and significance of 15 N methodologies in quantifying biological N2 fixation (BNF) and BNF dynamics in agro-ecosystems. Symbiosis; 2018; 75, pp. 1-16.1:CAS:528:DC%2BC2sXhvFekt7bL [DOI: https://dx.doi.org/10.1007/s13199-017-0526-z]

20. Castanheira, N et al. Annual ryegrass-associated bacteria with potential for plant growth promotion. Microbiol. Res.; 2014; 169, pp. 768-779.1:CAS:528:DC%2BC2cXhsVyksr4%3D [DOI: https://dx.doi.org/10.1016/j.micres.2013.12.010] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/24485300]

21. McKell, CM; Duncan, C. Competitive relationships of annual ryegrass (Lolium multiflorum Lam). Ecology; 1969; 50, pp. 653-657. [DOI: https://dx.doi.org/10.2307/1936254]

22. Hamel, C; Barrantes-Cartin, U; Furlan, V; Smith, DL. Endomycorrhizal fungi in nitrogen transfer from soybean to maize. Plant. Soil.; 1991; 138, pp. 33-40.1:CAS:528:DyaK38Xlslyl [DOI: https://dx.doi.org/10.1007/BF00011805]

23. Hamel, C; Furlan, V; Smith, DL. N2-fixation and transfer in a field grown mycorrhizal corn and soybean intercrop. Plant. Soil.; 1991; 133, pp. 177-185.1:CAS:528:DyaK3MXkslSntL4%3D [DOI: https://dx.doi.org/10.1007/BF00009190]

24. Hamel, C; Furlan, V; Smith, DL. Mycorrhizal effects on interspecific plant competition and nitrogen transfer in Legume-Grass mixtures. Crop Sci.; 1992; 32, pp. 991-996.1:CAS:528:DyaK3sXitFKhug%3D%3D [DOI: https://dx.doi.org/10.2135/cropsci1992.0011183X003200040032x]

25. Lesuffleur, F; Salon, C; Jeudy, C; Cliquet, JB. Use of a 15N2 labelling technique to estimate exudation by white clover and transfer to companion ryegrass of symbiotically fixed N. Plant. Soil.; 2013; 369, pp. 187-197.1:CAS:528:DC%2BC3sXhtVyqsbfJ [DOI: https://dx.doi.org/10.1007/s11104-012-1562-3]

26. Jalonen, R; Nygren, P; Sierra, J. Transfer of nitrogen from a tropical legume tree to an associated fodder grass via root exudation and common mycelial networks. Plant. Cell. Environ.; 2009; 32, pp. 1366-1376.1:CAS:528:DC%2BD1MXhtlWhs7zP [DOI: https://dx.doi.org/10.1111/j.1365-3040.2009.02004.x] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/19552666]

27. Trannin, WS; Urquiaga, S; Guerra, G; Ibijbijen, J; Cadisch, G. Interspecies competition and N transfer in a tropical grass-legume mixture. Biol. Fertil. Soils; 2000; 32, pp. 441-448.1:CAS:528:DC%2BD3cXoslWjt7w%3D [DOI: https://dx.doi.org/10.1007/s003740000271]

28. Fracetto, FJC; Fracetto, GGM; Barros, FMDR; Junior, L. Siqueira neto, M. Free-living diazotrophs drive castor bean nitrogen input in tropical semiarid soils. Ciênc Rural; 2019; 49, e20190446. [DOI: https://dx.doi.org/10.1590/0103-8478cr20190446]

29. Boddey, RM; Dobereiner, J. Nitrogen fixation associated with grasses and cereals: recent progress and perspectives for the future. Fertil. Res.; 1995; 42, pp. 241-250.1:CAS:528:DyaK28XivFWlsrw%3D [DOI: https://dx.doi.org/10.1007/BF00750518]

30. Dôbereiner, J. Azotobacter paspali Sp. N., UMA BACTÉRIA FIXADORA DE NITROGÊNIO NA RIZOSFERA DE Paspalum’.

31. Santos, ERS et al. Sward responses of Bahiagrass cultivars under no nitrogen fertilization. Crop Sci.; 2019; 59, pp. 2893-2902.1:CAS:528:DC%2BB3cXhtFGkurrM [DOI: https://dx.doi.org/10.2135/cropsci2019.06.0387]

32. Anaya-Ortega, JP; Garduño-Castro, G; Espinoza-Ortega, A; Rojo-Rubio, R; Arriaga-Jordán, CM. Silage from maize (Zea mays), annual ryegrass (Lolium multiflorum) or their mixture in the dry season feeding of grazing dairy cows in small-scale dairy production systems in the highlands of Mexico. Trop. Anim. Health Prod.; 2009; 41, pp. 607-616.1:STN:280:DC%2BD1M3htlGltg%3D%3D [DOI: https://dx.doi.org/10.1007/s11250-008-9231-5] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/18787970]

33. Chalk, PM et al. Methodologies for estimating nitrogen transfer between legumes and companion species in agro-ecosystems: A review of 15 N-enriched techniques. Soil. Biol. Biochem.; 2014; 73, pp. 10-21.1:CAS:528:DC%2BC2cXmvVWms7Y%3D [DOI: https://dx.doi.org/10.1016/j.soilbio.2014.02.005]