Mycorrhizal colonization had little effect on growth of Carex thunbergii but inhibited its nitrogen uptake under deficit water supply

Abstract

Background and Aims

Plant nitrogen (N) acquisition via arbuscular mycorrhizal fungi (AMF) serves as a dominant pathway in the N nutrition of many plants, but the functional impact of AMF in acquisition of N by wetland plants has not been well quantified. Subtropical lake–wetland ecosystems are characterized by seasonal changes in the water table and low N availability in soil. Yet, it is unclear whether and how AMF alters the N acquisition pattern of plants for various forms of N and how this process is influenced by soil water conditions.

Methods

We performed a pot study with Carex thunbergii that were either colonized by AMF or not colonized and also subjected to different water conditions. We used ¹⁵N labelling to track plant N uptake.

Key Results

Colonization by AMF had little effect on the biomass components of C. thunbergii but did significantly affect the plant functional traits and N acquisition in ways that were dependent on the soil water conditions. The N uptake rate of AMF-colonized plants was significantly lower than that of the non-colonized plants in conditions of low soil water. A decreased NO₃⁻ uptake rate in AMF-colonized plants reduced the N:P ratio of the plants. Although C. thunbergii predominantly took up N in the form of NO₃⁻, higher water availability increased the proportion of N taken up as NH₄⁺, irrespective of the inoculation status.

Conclusions

These results emphasize the importance of AMF colonization in controlling the N uptake strategies of plants and can improve predictions of N budget under the changing water table conditions in this subtropical wetland ecosystem.

INTRODUCTION

Nitrogen (N) levels are typically the factor limiting plant growth in terrestrial ecosystems, because N concentrations in soil are generally low and fluctuating. The forms of N in soil solution available for plant growth include NO₃⁻, NH₄⁺, amino acids, peptides and proteins (Finzi and Berthrong, 2005). The ability of roots to take up available N forms is an essential function in regulating N acquisition and storage by plants and the ability of an ecosystem to retain N (Ma et al., 2020). N uptake can also determine the outcome of competition between species, affecting ecological succession and vegetative adaptation to climate change (Kahmen et al., 2008). Inorganic N sources are typically the pivotal forms available to plant roots in natural ecosystems, because organic forms of N are more difficult for plants to use directly (Kieloaho et al., 2016). Often plants adapt to make better use of the form of N in the soils of their native habitats. In sites whose soils are dominated by a specific form of N, plants can achieve higher biomass or accumulate more N of this dominant form over other forms, showing a N form preference. Some plants have a NO₃⁻ preference owing to its accessibility and mobility, and high concentrations of NH₄⁺ can be toxic to plants (Roosta and Schjoerring, 2007). However, NH₄⁺ can be used immediately to synthesize amino acids through the glutamine synthetase–glutamate synthase pathway, whereas NO₃⁻ requires prior reduction to NH₄⁺ by nitrate and nitrite reductase before being assimilated (Kronzucker et al., 1997). Consequently, some plants show a preference for NH₄⁺, because it requires less energy for assimilation compared with NO₃⁻ (Kronzucker et al., 1997; Liu et al., 2013).

Organic N is reported to be preferred over inorganic N in low-temperature, N-limited soils in alpine and arctic ecosystems owing to slow rates of N mineralization (Schimel and Bennett, 2004; Xu et al., 2006). However, other authors have indicated that inorganic N is still the dominant N source used by plants in such habitats, although uptake of organic N occurs (Persson et al., 2006; Kahmen et al., 2009). Variation in N preferences is mostly attributable to interactions of diverse abiotic and biotic influences, such as the species of plants, mycorrhizal symbionts, vegetation coverage, soil nutritional status and ambient conditions (e.g. soil moisture, temperature) (Britto and Kronzucker, 2002; Ashton et al., 2010; Liu et al., 2013; Zhang et al., 2019). Soil N availability, for example, is well known to affect plant N acquisition processes and the preferred form of N (Song et al., 2015). Among other factors, moisture can change the level of soil aeration and affect N transformation and movement, including key processes such as nitrification, denitrification and mineralization (Rex et al., 2021). Low soil moisture can lead to higher nitrification, resulting in enhanced NO₃⁻ levels (Girsang et al., 2020). Conversely, wetter soils inhibit nitrification, resulting in more NH₄⁺ (Sun et al., 2020). The N uptake and form preference of plants can be affected by these changes (Pang et al., 2019; Pokharel and Chang, 2021). For instance, flooding inhibits the NO₃⁻ uptake rate by plants by reducing its supply in the soil (Tang et al., 2020).

Another strategy that plants use for N acquisition is association with arbuscular mycorrhizal fungi (AMF) (Brundrett and Tedersoo, 2018). Although the importance of AMF for plant phosphorus (P) nutrition is well established (Gamalero et al., 2004), an increasing number of studies have demonstrated that AMF can have an important role in plant N acquisition (Govindarajulu et al., 2005; Han et al., 2020), either directly, by root epidermal cells and root hairs, or indirectly, through AMF hyphae (Zhang et al., 2019). AMF can increase the active absorbing area of roots, increasing the ability of a plant to compete for inorganic N (Kuzyakov and Xu, 2013), and in some cases ≤70 % of total N acquired can be attributable to influences of AMF (Shi et al., 2016). The external mycelium of AMF appears to have a direct role in the transport of NO₃⁻ to roots (Subramanian and Charest, 1999) and in enhancement of hydraulic conductivity to water flow (Khalvati et al., 2005). AMF can acquire both NH₄⁺ and NO₃⁻ from soil (Hodge and Storer, 2015), with absorption and transportation of NH₄⁺ being higher owing to greater acquisition efficiency (Govindarajulu et al., 2005), although converse findings have been reported (Thirkell et al., 2019). Plants with AMF association can show N form preferences (Hodge and Storer, 2015; Thirkell et al., 2019), but it is unclear whether the N preference of an AMF-associated plant is directly attributable to its AMF partner and, if so, how the AMF partner might induce this N preference.

Seasonal wetlands intrinsically experience changes in their water tables over the course of a year. Water table fluctuations (WTFs), owing to unbalanced water budgets, are common abiotic stresses affecting the structure and function of riparian wetlands (Gownaris et al., 2018). Shallow freshwater wetlands are more sensitive to drying/wetting events attributable to WTFs because of aquatic–terrestrial interface phenomena (Leira and Cantonati, 2008). WTFs can profoundly affect soil properties, nutrient cycling, plant performance and plant physiological processes (Luo et al., 2018; Cao et al., 2022). It is, therefore, important to understand the impacts of WTFs on aquatic ecosystem structure and function (Akatsuka and Mitamura, 2011). Assessments of N uptake strategies by wetland plants have documented that plant species can differ in their N form preferences (Tian et al., 2020). AMF are commonly associated with wetland plants (Hu et al., 2020). However, it is not known whether plant species in subtropical wetland ecosystems can shift their N form preferences under different WTF regimes and how AMF participate in these processes. Carex thunbergii, a rhizomatous clonal sedge hygrophyte, accounts for >70 % of total coverage across the riparian zone and has been suggested previously to form functional mycorrhizal symbiosis (Wang et al., 2022). To that end, we performed a pot experiment with C. thunbergii either inoculated with AMF or not and grown in different soil water conditions. We test the hypotheses that AMF colonization could increase plant growth by promoting N uptake and that such promotion will vary with soil water availability.

MATERIALS AND METHODS

Plant and AMF sources

Carex thunbergii rhizomes and fresh soils were collected in February 2022 in the riparian zone of Yang’etou Conservation Station of Shengjin Lake (30°15ʹ–30°30ʹN, 116°55ʹ–117°15ʹE). Non-mycorrhizal rhizomes of C. thunbergii were obtained via vegetative reproduction from maternal plants, from which new unrooted rhizomes ~10 cm long were used. Fresh soil was collected from 0–20 cm soil depth, mixed thoroughly with river sand (1:1 v/v), and after passing the mixture through a 5 mm sieve, it was used as potting soil. To control the influence of soil biota, the potting soil was sterilized using γ-irradiation at 20 kGy for 120 min (Hefei Jieneng Irradiation Technology Co., Ltd). The mycorrhizal inoculum used in this study was Rhizophagus irregularis, purchased from the Bank of Glomeromycota in China (BGC, Beijing, China). The purchased material consisted of a mixture of spores, hyphae, soil and infected root fragments of Sorghum sudanense. Half of the inoculum was sterilized in same way as for the potting soil, and the sterilized inoculum was then applied as the non-AMF treatment. Pots without drain holes (16 cm in inner diameter and 13 cm in depth) were used. Before being planted with rhizomes, all pots were surface sterilized via a 15 min bath in 1 % sodium hypochlorite (NaClO), then rinsed three times in sterile water.

Experimental design

We used a 3 × 3 × 2 factorial design to explore N uptake patterns of plants: three N treatments (3) × three soil water content (SWC) levels (3) × two AMF treatments (2). There were five replicates (pots) of each treatment, for a total of 90 pots. Work was done in a screenhouse at Anhui University (31°76ʹN, 117°18ʹE), where the test materials experienced climatic conditions during the experimental period similar to those at the natural wetland where sampling was done, except that in the screenhouse the soil water regime was determined by us in the absence of natural rainfall (see below about soil water treatment). On 2 March 2022, we filled each pot with 1000 g dry weight of potting soil, combined with AMF inoculum (sterilized or not). First, we added 500 g of potting soil at the bottom of the pot, then we scattered 2 g of live or sterilized AMF inoculum evenly over the potting soil, and we added another 500 g of potting soil on top (Li et al., 2019). The rhizomes we produced, which were not infected with mycorrhizae, were carefully washed with tap water to remove adherent material and transplanted into the pots, with two rhizomes per pot. After a 2 week acclimation period with daily watering to full field capacity, we imposed the soil water treatment by watering pots with tap water until SWC was achieved (in metres cubed per metre cubed, expressed as a percentage), equal to 10, 20 or 35 %. This process was repeated every other day (Li et al., 2019). Plants were grown under ambient light and temperature, with mean daily minimum and maximum temperatures of 8.0 °C and 21.0 °C, relative humidity levels of 92.0 % and 37.0 %, and a light–dark cycle of 14 h–10 h, respectively. Every week, we switched the position of the pots randomly to eliminate any effect of position.

After 50 days of acclimation, we added the ¹⁵N labelling treatments at 10:00 h on 7 May 2022. We used ¹⁵NH₄Cl and K¹⁵NO₃ (both with 10 % atomic enrichment) as our labelled N. The test solution applied was a mixture of NH₄Cl and KNO₃, in which only one N form was ¹⁵N labelled for a given treatment, and the amount of labelled ¹⁵N was 0.2 mg (kg soil)⁻¹. An unlabelled N solution was used as the control, with 3.76 mg N (kg soil)⁻¹ in total, equivalent to the labelled treatment. Before adding the labelled N source, we drilled three similar 6-cm-deep holes in the medium in each pot. We then injected slightly <2 mL of N solution with a syringe into each hole, adding a total of 5 mL of the solution to each pot. The amount of N added per pot was small in order to minimize any potential fertilization effects [background N concentration was 747 mg N (kg soil)⁻¹]. Four hours after introduction of labelled N, we harvested the above-ground plant parts destructively by cutting the plants at the soil surface level. At the same time, the underground plant parts were separated carefully from the soil and rinsed with a 0.05 mmol L⁻¹ solution of CaCl₂ and distilled water to remove any labelled ¹⁵N from the surface of the roots. All plant materials were oven dried (at 70 °C) for 48 h. A 4 h incubation time minimized the potential for microbial transformation of the added inorganic N (especially NH₄⁺), while promoting the distribution of the N tracers over the whole plant, a process consistent with that of other studies using field crops, grasses and woody species (Liu et al., 2020, 2022; Wu et al., 2023), enabling comparison with that work.

Measurement and analysis

Before harvest, we used a portable laser leaf-area meter (V2-YMJ-B; Zhejiang Top Instrument Co. Ltd, Hangzhou, China) and measured the areas of the top four to six leaves. The selected leaves were then oven dried (70 °C), and we measured the specific leaf area (SLA; in centimetres squared per gram) by dividing leaf area (LA) by dry leaf weight (M). We measured the summed lengths of all fine roots (L; in metres) and divided L by their oven-dried mass (in grams), to obtain the specific root length (SRL; in metres per gram). Dried plant samples were ground with a ball mill (MM400; Retsch, Germany), then 2 mg was placed in tin capsules for measurement of the N concentration and the ¹⁵N/¹⁴N proportion using a Flash 2000 EA-HT elemental analyzer (Thermo Fisher Scientific, Inc., Waltham, MA, USA) and a DELTAV Advantage stable isotope ratio mass spectrometer.

For the fresh soil, the exchangeable NO₃⁻-N and NH₄⁺-N were extracted with 0.05 m K₂SO₄, then the levels of both N forms were determined using an auto analyser (AA3; Bran–Luebbe GmbH, Hamburg, Germany). To determine levels of P in plant components, 2 mg of the ground powder was weighed, then digested using the sulphuric acid–hydrogen peroxide digestion procedure (Zhang et al., 2015). The plant P concentrations in the digest were measured using the molybdate colorimetry method (Murphy and Riley, 1962). To measure the percentage of AMF root colonization of test plants, fresh root samples were fixed in a formalin–acetic acid–alcohol solution, then cleaned with 10 % KOH (w/v) at 90 °C for 1 h, followed by acidification with HCl for 5 min. Samples were then stained with 0.05 % Trypan Blue (w/v) at 90 °C for 30 min and incubated in decolouring solution (lactic acid:glycerol = 1:1, v/v) (Phillips and Hayman, 1970). Mycorrhizal colonization status was determined using the gridline intersect method under ×200 magnification with a microscope (Leica, DM2500; CMS GmbH, Wetzlar, Germany). Only arbuscules within the roots were considered as evidence of AMF colonization. The percentage of AMF root colonization was calculated as the number of roots colonized by AMF divided by the total number of roots examined.

Calculations and statistics

For plant tissues, the ¹⁵N atomic percentage excess (APE) was calculated as the difference in atomic percentage of ¹⁵N between ¹⁵N-labelled and unlabelled samples, as follows:

$$\begin{array}{l}{\displaystyle \mathrm{A}\mathrm{P}\mathrm{E}(\mathrm{\%})=\mathrm{A}\mathrm{t}\mathrm{o}\mathrm{m}{\mathrm{\%}}^{15}{\mathrm{N}}_{\mathrm{l}\mathrm{a}\mathrm{b}\mathrm{e}\mathrm{l}\mathrm{l}\mathrm{e}\mathrm{d}}-\mathrm{A}\mathrm{t}\mathrm{o}\mathrm{m}{\mathrm{\%}}^{15}{\mathrm{N}}_{\mathrm{u}\mathrm{n}\mathrm{l}\mathrm{a}\mathrm{b}\mathrm{e}\mathrm{l}\mathrm{l}\mathrm{e}\mathrm{d}}}\end{array}$$ (1)

For plant tissues, we measured the N concentrations and the atomic percentage of ¹⁵N of roots and shoots. We calculated the ¹⁵N level in roots or shoot tissues (in micrograms) by multiplying the root or shoot N concentration (in micromoles of N per gram) by the corresponding (APE/100), biomass (in grams) and the relative molecular mass of ¹⁵N, as follows:

$$\begin{array}{l}{\displaystyle {}^{15}\mathrm{N}\mathrm{a}\mathrm{m}\mathrm{o}\mathrm{u}\mathrm{n}\mathrm{t}(\mu \mathrm{g})=\mathrm{N}\mathrm{c}\mathrm{o}\mathrm{n}\mathrm{c}\mathrm{e}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}(\mu \mathrm{m}\mathrm{o}\mathrm{l}{\mathrm{g}}^{-1})\times \frac{\mathrm{A}\mathrm{P}\mathrm{E}}{100}\times \mathrm{b}\mathrm{i}\mathrm{o}\mathrm{m}\mathrm{a}\mathrm{s}\mathrm{s}(\mathrm{g})\times 15(\mathrm{g}\mathrm{m}\mathrm{o}{\mathrm{l}}^{-1})}\end{array}$$ (2)

The plant ¹⁵N uptake rate (¹⁵NUR; in micrograms per gram) was calculated by summing the root and shoot ¹⁵N values, then dividing by the root biomass (in grams), as follows:

$$\begin{array}{l}{\displaystyle {}^{15}\mathrm{N}\mathrm{U}\mathrm{R}(\mu \mathrm{g}{\mathrm{g}}^{-1})=\frac{\mathrm{R}\mathrm{o}\mathrm{o}\mathrm{t}{}^{15}\mathrm{N}\mathrm{a}\mathrm{m}\mathrm{o}\mathrm{u}\mathrm{n}\mathrm{t}(\mu \mathrm{g})+\mathrm{S}\mathrm{h}\mathrm{o}\mathrm{o}\mathrm{t}{}^{15}\mathrm{N}\mathrm{a}\mathrm{m}\mathrm{o}\mathrm{u}\mathrm{n}\mathrm{t}(\mu \mathrm{g})}{\mathrm{R}\mathrm{o}\mathrm{o}\mathrm{t}\mathrm{b}\mathrm{i}\mathrm{o}\mathrm{m}\mathrm{a}\mathrm{s}\mathrm{s}(\mathrm{g})}}\end{array}$$ (3)

Before the ANOVA, all the data were tested for normality (Kolmogorov–Smirnov test) and homogeneity of variance (Levene’s test). No data transformation was performed, because the assumptions of normality and homogeneity of variance were satisfied. The effects of AMF inoculation, SWC and N form on plant ¹⁵N uptake rate were analysed with a three-way ANOVA. Comparisons between means were tested using Tukey’s honestly significant difference (HSD) at the P = 0.05 level. Principal component analysis (PCA) was performed to explore the effects of different biotic and abiotic factors on NO₃⁻-/NH₄⁺ acquisition. All analyses were performed using SPSS v.25.0 (SPSS, Chicago, IL, USA).

RESULTS

Soil properties and mycorrhizal colonization

No root samples were colonized by mycorrhizae in plants treated with sterilized inoculum (Supplementary Data Fig. S1). High SWC (20–35 %) significantly reduced AMF colonization in plants treated with live AMF inoculum. The percentage of AMF root colonization was the highest (35.9 %) at low SWC (10 %), which was higher than that at medium (29.8 %) or high (27 %) soil water levels (P < 0.05). The percentages of AMF root colonization between medium and low SWC levels were not significantly different (Supplementary Data Fig. S1). In most cases, the concentrations of soil NH₄⁺ were higher than levels of NO₃⁻, except in the case of low SWC without AMF, for which NO₃⁻ concentration was high (Supplementary Data Fig. S2).

The growth performance of C. thunbergii

Although most biomass components were not affected by AMF colonization, SWC or their interactions (P > 0.05; Table 1; Supplementary Data Fig. S3), we found contrasting effects of AMF colonization on the root-to-shoot (R:S) ratio in C. thunbergii along the SWC gradient (P < 0.05; Table 2). In the presence of AMF, SWC had no effect on R:S, but significantly reduced the R:S in the absence of AMF (Fig. 1). This difference is largely attributable to variation of below-ground biomass allocations between AMF treatments. Meanwhile, R:S was higher with AMF inoculation than without inoculation, especially in medium and high SWC treatments (P < 0.05).

Table 1.

Multivariate ANOVA results (P-values) for the effects of AMF inoculation, soil water content and their interaction on plant biomass components, AMF colonization and soil inorganic N concentrations.

Source of variation		Above-ground biomass		Below-ground biomass		Total biomass		NH4-N		NO3-N		AMF colonization		Inorganic N
	d.f.	F	P-value	F	P-value	F	P-value	F	P-value	F	P-value	F	P-value	F	P-value
AMF inoculation (A)	1	2.462	0.122	0.085	0.772	0.026	0.873	8.451	0.013	1.651	0.223	2936.85	<0.001	4.708	0.051
Soil water content (S)	2	2.715	0.074	0.427	0.655	1.119	0.333	2.25	0.148	3.279	0.073	23.20	<0.001	3.065	0.084
A × S	2	0.989	0.378	1.627	0.205	1.891	0.16	0.562	0.584	0.796	0.474	20.47	<0.001	0.47	0.636

Significant values are shown in bold.

Table 2.

Multivariate ANOVA results (P-values) for the effects of AMF inoculation, soil water content and their interaction on root length, SRL, R/T, plant height, tiller number and SLA.

Source of variation		Root length		SRL		Root/Shoot		Plant height		Tiller number		SLA
	d.f.	F	P-value	F	P-value	F	P-value	F	P-value	F	P-value	F	P-value
AMF inoculation (A)	1	0	1	1.664	0.202	4.596	0.036	15.909	<0.001	4.581	0.035	0.27	0.605
Soil water content (S)	2	14.224	<0.001	1.768	0.179	4.735	0.012	5.672	0.004	2.443	0.093	4.491	0.015
A × S	2	0.437	0.648	0.968	0.386	0.626	0.538	1.111	0.332	0.626	0.537	5.278	0.008

Abbreviations: SLA, specific leaf area; SRL, specific root length. Significant values are shown in bold.

Fig. 1.

Specific leaf area (SLA), root-to-shoot (R:S) ratio, root length, specific root length (SRL), tiller number and plant height of Carex thunbergii colonized either with AMF (AMF⁺) or without AMF (AMF⁻) at 10, 20 and 35 % SWC at harvest. Different lowercase letters indicate significant differences among different SWC levels in the same inoculation conditions; different capital letters indicate significant differences between AMF⁺ and AMF⁻ at the same SWC level. Vertical bars indicate s.e.m. (n = 15).

Plant functional traits and stoichiometry

The SLA was affected by SWC and its interaction with AMF inoculation (P < 0.05; Table 2). In the AMF⁺ treatment, SLA increased with increasing SWC, whereas this trend was reversed in AMF⁻ treatment. Root length was affected only by SWC (P < 0.05), with a similar increasing trend along the SWC gradient for both AMF treatments. However, neither SWC nor AMF treatment had a significant effect on the SRL of C. thunbergii. Likewise, although an AMF effect on tiller number was indicated by ANOVA, no significant difference was detected owing to its wide variation. Plant height was affected by both AMF inoculation and SWC, with consistently higher values found in the AMF⁻ treatment than that in the AMF⁺ treatment (all P < 0.05). Also, plant height increased with SWC, and to a significant level in the AMF⁻ treatment (P < 0.05). Consequently, the highest plant height of 45.95 cm was found at high SWC in the AMF⁻ treatment (Fig. 1).

Both the shoot and total N:P ratios in the AMF⁺ treatment were lower than those in the AMF⁻ treatment (P < 0.05; Supplementary Data Table S1). However, the SWC mainly influenced root and total N:P ratios, and the root N:P at the low SWC was significantly lower than that at medium or high SWC (Fig. 2). Overall, the N:P ratio of C. thunbergii was <14, signifying N-limited growth, and AMF inoculation reinforced this limitation.

Fig. 2.

Plant N:P partitioning by Carex thunbergii either with or without AMF colonization, at 10, 20 and 35 % SWC at harvest. Different lowercase letters above the bars indicate significant differences for the N:P ratio among different soil water levels (SWC) in the same inoculation conditions. Asterisks indicate significant differences between different inoculation treatments within the same water level. Vertical bars indicate s.e.m. (n = 5). Abbreviations: AU, above-ground N:P; BG, below-ground N:P; T, total N:P.

Nitrogen uptake by C. thunbergii

The plant N uptake pattern was altered by AMF inoculation, SWC, N form and their interactions (all P < 0.05; Table 3). In the AMF⁺ treatment, C. thunbergii took up more NH₄⁺, NO₃⁻ and total inorganic N with increasing SWC. The uptake rate of NH₄⁺ and NO₃⁻ was highest at high SWC (0.67 and 0.78 μg g⁻¹ h⁻¹, respectively). In the AMF⁻ treatment, there was no significant difference in the uptake rate of NH₄⁺ with increases of SWC, but the uptake rate of NO₃⁻ at medium SWC was significantly higher (1.62 μg g⁻¹ h⁻¹) than at either low or high SWC. The uptake rate of NH₄⁺ was lower than that of NO₃⁻ irrespective of inoculation status, especially at low SWC (P < 0.05; Fig. 3). Finally, at low SWC, the NH₄⁺, NO₃⁻ and total inorganic N uptake rates in the AMF⁻ treatment were higher than those in AMF⁺ treatment (P < 0.05).

Table 3.

Multivariate ANOVA results (P-values) for the effects of AMF inoculation, soil water content, N form and their interaction on plant ¹⁵N uptake rate.

	d.f.	F	P-value
AMF inoculation (A)	1	11.877	0.001
Soil water content (S)	2	35.09	<0.001
N form (N)	1	58.394	<0.001
A × S	2	6.05	0.005
A × N	1	11.883	0.001
S × N	2	17.069	<0.001
A × S × N	2	5.292	0.009

Significant values are shown in bold.

Fig. 3.

Plant ¹⁵N uptake rate of Carex thunbergia, either with (AMF⁺) or without AMF (AMF⁻), at 10, 20 and 35 % SWC at harvest. Different lowercase letters indicate significant differences among different SWC in the same inoculation conditions for the same N form; different capital letters indicate significant differences between AMF⁺ and AMF⁻ at the same SWC for the same N form. The different letters within the column indicate significant differences in total inorganic N uptake rates. Asterisks indicate significant differences between different N forms within the same water level and inoculation treatment. Vertical bars indicate s.e.m. (n = 5).

Effects of biotic and abiotic factors on plant ¹⁵N uptake

The relationships between the AMF⁻ treatment and other factors were obviously different from that of the AMF⁺ treatment with different biotic and abiotic factors, as suggested by PCAs (Fig. 4). In the AMF⁺ treatment, the first axis (PC1) explained 34.1 % of the variance and was dominated by above-ground biomass, root length, N uptake rate and SWC, the second axis (PC2) explained 17.9 % of the variance and was dominated by below-ground biomass, R:S and SRL for NH₄⁺ uptake; while PC1 explained 29.4 % of the variance and was dominated by R:S, SRL, plant height, N uptake rate and N:P, PC2 explained 21.4 % of the variance and was dominated by both above- and below-ground biomass, root length and SWC for NO₃⁻ uptake (P < 0.05; Fig. 4; Supplementary Data Table S2). In the AMF⁻ treatment, PC1 explained 34.1 % of the variance and was dominated by plant height, root length, N uptake rate, N:P and SWC, PC2 explained 22.3 % of the variance and was dominated by above-ground biomass, R:S and SLA for NH₄⁺ uptake; while PC1 explained 43.2 % of the variance and was dominated by below-ground biomass, R:S, N uptake rate and SWC, PC2 explained 24.5 % of the variance but there was no variable associated with this axis for NO₃⁻ uptake (P < 0.05; Supplementary Data Table S3).

Fig. 4.

Principal component analysis correlation biplots depicting the relationship between plant N uptake rates and biotic and abiotic factors. More than 70 % of the variability was comprised by the first four components (for component loading values, see Supplementary Data Table S4).

DISCUSSION

Colonization by AMF in response to SWC

Many wetland species form associations with AMF, in both experimental (Wolfe et al., 2006) and natural conditions (Muthukumar et al., 2004; Fraccaro de Marins et al., 2009). SWC can have different effects on root AMF colonization (Michalis et al., 2013), and lower percentages of AMF root colonization have been reported in flooded conditions compared with non-flooded ones (e.g. Deepika and Kothamasi, 2015). In this study, we found a decreasing percentage of AMF root colonization in association with increasing water availability, which is consistent with previous studies. The percentages of AMF root colonization might be lower in flooded soil compared with dry soil; however, in our flooded treatments C. thunbergii was colonized by the AMF inoculum at 27 %, which was a higher percentage than in other wetland plants in other studies (e.g. <6 % in Typha latifolia L., Ipsilantis and Sylvia, 2007). A low level of root colonization does not necessarily imply low functionality, because the established symbiosis might still be functional in transferring nutrients (e.g. P and N) to plants (Fitter, 2006).

Dominant effect of SWC on plant growth

As in terrestrial ecosystems, AMF colonization of plants in wetlands has effects that range from beneficial (e.g. Andersen and Andersen, 2006) to neutral (Deepika and Kothamasi, 2015; Bao et al., 2019) to negative (e.g. Wolfe et al., 2006). A meta-analysis of experimental pot studies by Ramírez-Viga et al. (2018) concluded that the overall degree of benefit provided by AMF to wetland plants in pot experiments depended on the identity of the plant and the P and water availability in the soil. Studies show that the effects of AMF infections on plant growth are plant specific (Zheng et al., 2018; Sun et al., 2022). Some plants have strong, positive performance responses to AMF, whereas others show no significant changes. Previously, AMF inoculation was reported to stimulate biomass accumulation of wetland plants but to reduce the R:S ratio (Fraser and Feinstein, 2005). However, in our study, we detected no significant effect of AMF on total biomass or its components for C. thunbergii, in contrast to our first hypothesis. Not all attributes of mycorrhiza-infected plants show improvement over non-inoculated plants [e.g. dry weight (Dhillion, 1992) and height (Solaiman and Hirata, 1996)], and this variability might depend on the symbiont species involved, the characteristics of the physicochemical substrate (i.e. nutrient availability) or even the duration of the pot assay. Also, the degree of benefit obtained from AMF association can differ according to the mycorrhizal dependence of the host plant (van der Heijden et al., 1998; Chandrasekaran et al., 2014). Wu et al. (2014), for instance, showed that R. irregularis did less to improve the performance of Phragmites australis (Cav.) Trin ex. Steudel seedlings than did another AMF, Funnelliformis mosseae, suggesting that variability among AMF species in stimulation of host plant growth might be common. The variables that have typically shown the greatest changes in response to AMF inoculation are those related to nutrient acquisition, followed photosynthetic activity and biomass production (Ramírez-Viga et al., 2018). Therefore, evaluation of nutrient acquisition responses, the core benefit to plants of AMF association (Smith and Read, 2008), might be the best approach for evaluation of plant responses to mycorrhizal inoculation.

In dry or nutrient-limited conditions, many plants allocate more biomass to roots. Such functional adjustments are assumed to improve the access of plants to a limiting resource (Freschet et al., 2018). In accordance with the functional equilibrium hypothesis (Freschet et al., 2015) and the optimal allocation theory (Shipley and Meziane, 2002), the lower SWC (10–20 %) induced a higher R:S ratio, presumably allowing the plant better to exploit a larger soil volume. These events occurred in the absence of AMF, but in the presence of AMF symbionts, plant roots are generally able to forage for resources more effectively, and AMF-colonized plants tend to invest fewer resources in root system development, resulting in a lower R:S ratio and reduced root biomass (Smith and Read, 2008). In our study, AMF inoculation was associated with a trend towards a lower R:S ratio in C. thunbergii, while we found no effects of inoculation on other functional traits (e.g. root length), as found by Stevens et al. (2002) with Lythrum salicaria.

Nitrogen uptake as regulated by SWC and AMF colonization

The relative roles of plant roots and AMF in plant N acquisition are affected by the mycorrhizal taxon, soil N availability and soil conditions (e.g. Wallander, 2002). Although AMF can take up both NO₃⁻ and NH₄⁺, they exhibit a clear preference for NH₄⁺ (Toussaint et al., 2004). NH₄⁺ is generally bound to soil particles and not easily moved, especially in dry conditions, compared with NO₃⁻. In our study, we found that the preference of C. thunbergii for NO₃⁻ decreased whereas its preference for NH₄⁺ increased in high-water compared with low-water conditions, regardless of AMF colonization status (Fig. 3). These results reinforced the idea that plants usually absorb more NO₃⁻ in drier environments and more NH₄⁺ in wetter environments, a finding that has been validated in other natural ecosystems (Houlton et al., 2007; Wang and Macko, 2011). In our study, both high SWC (35 %) and AMF inoculation promoted N uptake, especially in the form of NH₄⁺, showing that adequate water supply could amplify the benefits of AMF symbionts on plant growth (Wang et al., 2022). Importantly, in support of our second hypothesis, water availability might override AMF colonization status and dominate the N form preference of a plant, enabling C. thunbergii in our study preferentially to absorb NH₄⁺, the relatively more abundant N form in the growth medium.

As in other studies (Michalis et al., 2013), we found that low water availability reduced plant N uptake by C. thunbergii. Low water availability might reduce N mineralization, consequently lowering N availability in soils. However, no significant difference was detected in N availability among the three soil water treatments here. P is another key nutrient potentially enhanced via AMF association. We quantified the P concentration and found that P levels in the mycorrhizal treatment were consistently higher than in the non-mycorrhizal treatment, indicating that AMF can aid plants in P uptake (Mei et al., 2019), especially in drier soils. P uptake by mycorrhizae-infested roots can be 2- to 5-fold greater than by non-AMF roots (Sanders and Sheirh, 1983). The exchange balance model (Johnson, 2010) posits that the function in terrestrial systems of mycorrhizal symbiosis depends especially on ratios of available N and P in the soil. The fact that N:P ratios were lower in the mycorrhizal treatment than in the non-mycorrhizal treatment supports this view (Fig. 2). Although mycorrhizal associations can reduce plant biomass, plants need AMF symbionts to increase P uptake (Chu et al., 2020). Recently, Marro et al. (2022) have also indicated that AMF can have stronger positive effects on P nutrition than on plant growth or N nutrition. Thus, P availability in soil might determine whether AMF symbiosis is beneficial or parasitic. Considering the high plant N:P ratio observed in our study (especially in shoots) in low-water conditions, AMF association might have decreased plant N:P ratios by supplying extra P, leading to N limitation (Zhao et al., 2015; Mei et al., 2019), supporting the growth rate hypothesis. Conversely, with the higher soil water levels in our study, the interaction between host plants and AMF might have become neutral, partly confirming the view that plant–AMF relationships lie along a mutualism-to-parasitism continuum (Johnson et al., 1997; Wipf et al., 2019). Another possible mechanism explaining a neutral interaction between AMF and plants in medium- or high-SWC conditions could be that adequate soil water alleviates N limitation by increasing plant N:P ratios (albeit not to a significant level; Fig. 2). Ramírez-Viga et al. (2018) inferred that, in wetland habitats, mutualistic functions are more likely in P-limited systems, and commensalism or parasitism is more likely in N-limited systems.

Also, the degree of benefit from AMF association can be host-plant specific (Ramírez-Viga et al., 2018). Furthermore, given that AMF species might require considerable N to support their own growth (Corrêa et al., 2015), the AMF associate might compete with its host plant or N (Eisenhauer et al., 2009), especially in low-SWC conditions. Our results also suggest that not all AMF symbiosis is beneficial for the host plant, as mentioned above. Different AMF and plant species differ in their ability to acquire and deliver nutrients and carbon to their partner (Johnson, 2010). Taken together, these results imply that symbiotic AMF might be competitors that worsen N limitations on plant growth by reducing plant uptake of N under deficit water supply (Johnson et al., 1997, 2015).

Functional traits contributing to N uptake by C. thunbergii are N source specific

Plants have evolved specific morphological and physiological plasticity in root traits to capture N. Such traits strongly influence the acquisition of nutrients by plants (Hawkins et al., 2014). Root traits, such as volume, surface area and length, affect interception of N by plants (Ricroch et al., 2014). Higher root biomass and SRL generally improve the absorption of N by plants (Hawkins et al., 2014; Hong et al., 2018) by shortening the distances between nutrients and the root surface and expanding the root interception area. To sustain and improve N absorption, plants must adopt either an intensive strategy (morphological adaptations) or an extensive strategy (increasing root biomass) (Hong et al., 2018). In the presence of AMF, both root morphological traits (e.g. SRL) and root system size-related traits (e.g. R:S) showed stronger relationships with the NO₃⁻ uptake rate in C. thunbergii, but these traits make only a limited contribution to the acquisition of NH₄⁺. A larger root biomass and therefore R:S improves the absorption of N by plants owing to the larger contact area within the soil (Hawkins et al., 2014). At the same time, studies have also shown that lower root diameters optimize NO₃⁻ acquisition by enhancing SRL values (Weemstra et al., 2016). Long, thin roots (high SRL) are less energetically demanding to produce (Withington et al., 2006), making exploitation of soil resources more efficient and helping in resource foraging (Bowsher et al., 2016). However, because AMF association can lower the N:P ratio and given that the NO₃⁻ uptake rate is closely related to the N:P ratio, we suggest that AMF association might principally limit uptake of NO₃⁻. In the absence of AMF, however, these functional traits do not contribute positively to NO₃⁻ acquisition. The trade-off between relying on better root morphology built by more carbon vs. the use of a more energy-intensive physiological process to acquire N is an effective nutrient acquisition strategy for plants in variable environments (Duan et al., 2014).

In natural ecosystems, N is generally supplied to roots via diffusion (Lambers et al., 2008). Uptake of NH₄⁺, especially, is controlled predominantly by the soil water supply, and therefore N uptake rates are higher when there is an adequate water supply to promote diffusion of NH₄⁺ to the root surface. As a result, in partial support of our previous work, we observed that symbiotic AMF have an advantage for soil N acquisition in wet conditions (relatively high SWC; Wang et al., 2022). This potential benefit might also be reflected by the positive relationship between root length and uptake of NH₄⁺ in the case of non-inoculation with AMF. Plants can therefore compensate for reductions in their AMF associations with greater root length, which increases the volume of soil exploited per unit of root investment (Ostonen et al., 2007). Longer roots have a greater capacity for N uptake (Albano et al., 2021), which can ensure a high, steady NO₃⁻ absorption rate. In contrast, plants with greater mycorrhizal colonization often have shorter, thicker roots and therefore shorter SRL (Comas et al., 2014; Ma et al., 2018).

SUPPLEMENTARY DATA

Supplementary data are available at Annals of Botany online and consist of the following.

Table S1: multivariate ANOVA results for the effects of AMF inoculation, SWC and their interaction on plant N:P. Table S2: bivariate correlations among plant functional traits, SWC and plant N uptake rate included in the study in AMF⁺ treatment. Table S3: bivariate correlations among plant functional traits, SWC and plant N uptake rate included in the study in AMF⁻ treatment. Table S4: eigenvalue, cumulative percentage variance and the correlation coefficients between factors and the major axes of PCA ordination. Fig. S1: root AMF colonization of C. thunbergii. Fig. S2: soil inorganic N concentrations, either with or without AMF, at 10, 20 and 35 % SWC at harvest. Fig. S3: plant biomass partitioning of C. thunbergia, either colonized with AMF or without, at 10, 20 and 35 % SWC.

FUNDING

This work was supported by the National Natural Science Foundation of China (32271639), the Anhui Provincial Natural Science Foundation (2108085MC88) and the University Natural Science Research Project of Anhui Province (KJ2021A0087).

CONFLICT OF INTEREST

The authors declare that they have no conflict of interest.