Soil characteristics and experimental setup

The soil was a Typic Natraqualf from San Vicente (Buenos Aires, Argentina). The plant community in the sampling site was dominated by L. tenuis, Distichlis spicata (L.) Greene, Paspalum vaginatum Swartz, Eleocharis viridans Kük. ex Oken, and Cynodon dactylon (L.) Pers. The soil is characterized by: pH 9.53, electrical conductivity 10.10 dS/m, Na⁺ 19.45 cmol_c/kg, available P 8.72 mg/kg (Bray I), total carbon 9.10 g/kg, and total N 1.02 g/kg ⁹. The soil was air-dried and sieved through a 2-mm mesh screen.

Seeds of L. tenuis cv. Esmeralda and S. arundinaceus cv. Arizona were surface-sterilized and pregerminated under sterile conditions. Fifty closed-bottom 1.6-L pots were filled with 950 g of air-dried soil. Three seedlings of each plant species were planted together in each pot and the soil surface was covered with 1 cm of sterilized sand to conserve moisture. During the experimental period, pots were placed in a greenhouse at a mean relative humidity of 65 ± 15%, mean temperature of 30 ± 5°C and 20 ± 3°C during days and nights, a photoperiod ranging from 10 to 12 h, and a photon-flux density at midday of 900-1,300 μmol m^-2s^-1. Pots were randomized and daily rotated to minimize potential gradient effects.

After 62 days of growth in soil at 80% of field capacity (46.10% w/w, 33 kPa), five pots were harvested (initial plants) and the remaining pots were subjected to a combination of two defoliation frequencies and two water stress conditions (water excess and deficit). Defoliation treatments comprised removing all plant material 4 cm above ground level at different time intervals over a 28-day period, starting with defoliation of all plants, except controls (non-defoliated plants under all irrigation conditions), at the beginning of this period. L. tenuis and S. arundinaceus were defoliated by clipping at a frequency of 2 and 4 times representing intervals of 2 and 1 weeks between defoliations, respectively. The 2-time frequency is here referred to as low defoliation frequency (LF) and the 4-time frequency as high defoliation frequency (HF). Water-stress treatments consisted in subjecting plants to water excess (WE), water deficit (WD), or 80% of field capacity (WW- well watered plants) for a 28-day period. The WE treatment consisted in maintaining a water level of 1 cm above the soil surface, whereas the WD treatment consisted in keeping soil moisture at 24.50 w/w (1,488 kPa), near the permanent-wilting-point level of soil hydration (1,500 kPa). Control pots contained non-defoliated plants maintained at the three different water levels (WW, WE, and WD). Five replicates were assigned to each combination of defoliation frequency and water treatment. The water status of pots was examined daily by weighing them and the amount of water lost was replaced. At the end of the 28 days of treatment, all plants were harvested (final plants).

Plant yield and analytical determinations in plant tissue and soil

After each harvest (initial and final plants), L. tenuis and S. arundinaceus biomasses were separated into shoots and roots. All shoot biomasses of each plant species removed during the defoliation and water treatments were dried and weighed to be later included in the shoot fraction of the corresponding plant at the end of the experiment, determining accumulated shoot dry weight. Shoots and roots of each plant species were oven-dried at 70°C for 48 h and weighed. A portion of fresh root was used to measure AM fungal colonization in both plant species and number of rhizobia nodules in L. tenuis roots. For each species, defoliation or stress-tolerance efficiency (STE) for each water level was calculated using the following equation:

STE (%) = (dry biomass of defoliated plant/ dry biomass of non-defoliated plant) x 100

Shoots and roots of each species were digested separately in sulfuric acid to determine P content by a modification of the Murphy Riley method and N content by the Kjeldahl method ¹⁶. Total absorbed N per nodule of rhizobia was calculated ¹⁰. Soil P availability was measured by Bray and Kurtz. The root fraction colonized by AM fungi (mycorrhizal colonization index) and the root fraction colonized by DSE fungi (DSE colonization index) were assessed following McGonigle et al. (1990). Septate, melanized or hyaline hyphae and microsclerotia were classified as DSE fungi ². Rhizobia nodules were counted in whole fresh root system under a binocular stereomicroscope (7.5x). Results were analyzed by ANOVA with Tukey’s mean separation test (P ≤0.05) using InfoStat statistical software ⁸.

Plant growth

Under each combination of treatments (water level+defoliation) studied in a saline-sodic soil, L. tenuis contributed near 80% of the total biomass produced by the co-culture (Figure 1a, page 104).

Different letters indicate significant differences (P< 0.05) for each plant species among treatments according to Tukey’s test. Letras diferentes indican diferencias significativas (P< 0,05) para cada especie entre tratamientos según test de Tukey.

DW, dry weight; ND, no defoliation; LF, low defoliation frequency; HF, high defoliation frequency; WW, well watered; WE, water excess; WD, water deficit. Mean (N=5)± SE. DW, peso seco; ND, no defoliada; LF, baja frecuencia de defoliación; HF, alta frecuencia de defoliación; WW, bien regada; WE, exceso hídrico; WD, déficit hídrico. Media (N=5)±

Figure 1: Figura 1: Biomass (a, b) of Lotus tenuis and Schedonorus arundinaceus grown in co-culture under a combination of different defoliation frequencies and water stress. Biomasa (a, b) de Lotus tenuis y Schedonorus arundinaceus crecidas en co-cultivo ante la combinación de diferentes frecuencias de defoliación y estrés hídrico. 

Lotus tenuis defoliated plants (low and high frequency) were not able to compensate for the shoot biomass produced by non-defoliated plants under well watered conditions through tissue regrowth (defoliation frequency mean value of STE: 59.7%) (Figure 1a, page 104 and Table 1, page 104).

Compensation of removed shoot tissue, especially after frequent defoliation, requires large amounts of energy investment, which is derived from reallocating energy stored in the remnant shoot and root tissue ¹⁸, resulting in reduced shoot biomass and root growth (Figure 1b, page 104 and Table 1, page 104). Unlike L. tenuis, S. arundinaceus produced similar amounts of accumulated shoot biomass in defoliated and non-defoliated plants under well watered conditions (Figure 1a, page 104 and Table 1, page 104).

Since S. arundinaceus was able to compensate for biomass clipping by increasing the allocated biomass below the cutting height (data not shown) and maintaining root growth relative to that of non-defoliated plants, defoliation at the two frequencies actually resulted in enhanced plant growth (Figure 1a, b and Table 1). These results are consistent with the highest values of STE recorded for this species (defoliation frequency mean value of 126.5%), which showed high tolerance to frequent defoliation under well watered conditions.

Under the water excess treatment, L. tenuis and S. arundinaceus non-defoliated plants recorded the highest accumulated shoot biomass (5.16 g for L. tenuis and 1.18 g for S. arundinaceus) (Figure 1a and Table 1). In agreement with that previously described by several authors ^11,27, the adaptation of both species to water excess conditions may be due to an increased soil nutrient availability ¹¹ along with the consequent reduction in soil salinity and sodicity as a result of soil water excess ¹¹. In the present study, the soil of non-defoliated plants under water excess showed a significant increase in available P with regard to the well watered condition (9.72 mg/kg and 12.23 mg/kg for well watered and water excess, respectively). The combination of water excess plus defoliation treatments resulted in a decrease in accumulated shoot and root biomasses of both plant species. These results were associated with lowest efficiencies in tolerating stress (Figure 1 a, b, page 104). Mean defoliation frequency of STE was 46.5% for L. tenuis and 68.9% for S. arundinaceus (Table 1, page 104). Both low and high defoliation frequencies in a waterlogged soil can be considered as the most stressful scenario for a L. tenuis and S. arundinaceus co-culture. The low STE of both species may be associated with a marked loss of nutrients, which affected plant ability to recover the defoliated biomass when defoliation frequency increased, or with decreases in root growth and soil P availability (defoliation frequency mean value of 10.43 mg/kg).

Under water deficit, non-defoliated and defoliated plants of both species produced similar quantities of shoot biomass (Figure 1 a, b and Table 1, page 104). A reduction in soil moisture resulted in a decrease in soil P availability (mean value for the combination of defoliation and water treatment of 8.39 mg/kg), affecting plant growth and P absorption. Defoliated plants growing under water deficit, however, have an adaptive advantage because a lower proportion of aerial biomass leads to an improvement in plant water balance and an avoidance of excessive losses of water through transpiration compared to non-defoliated plants. In these species, defoliation may alleviate the deleterious effect of water deficit on shoot biomass accumulation, enabling defoliated plants to compensate for the quantity of shoot biomass produced by control plants under drought, whose response becomes manifest in a high STE percentage (Figure 1a, page 104). Mean defoliation frequency of STE was 75.2% for L. tenuis and 95.5% for S. arundinaceusTable 1, page 104). Loss of apical dominance after defoliation and basal meristem activation may stimulate such clipped stems to develop more biomass ¹⁵. Moreover, an increase in the photochemical efficiency and photosynthetic capability of remnant leaves below a cutting induces a compensatory growth response through an increase in photosynthates ²⁷.

Defoliation generally results in reduced root growth ¹⁵. Root biomass loss in L. tenuis in response to defoliation frequency, especially in combination with water stress, would be attributable to such a decrease in root growth, a diversion of assimilated substances, and a mobilization of P and N from roots to regrowing shoot tissues (Figure 1b and Table 1, page 104). In contrast, the strategy of defoliated S. arundinaceus plants was to maintain root growth after low and high defoliation frequencies independently of water status, similar to the respective non-defoliated ones (Figure 1b and Table 1, page 104).

P and N nutrition

Lotus tenuis defoliated plants under well watered conditions were able to compensate total absorbed P by non-defoliated plants (Figure 2a, page 106 and Table 1, page 104).

Figure 2: Figura 2: Plant P (a) and N (b) content in Lotus tenuis and Schedonorus arundinaceus grown in co-culture under a combination of different defoliation frequencies and water stress. Figure c shows total N per nodule of L. tenuis. Contenido de P y N en las plantas de Lotus tenuis y Schedonorus arundinaceus crecidas en co-cultivo ante la combinación de diferentes frecuencias de defoliación y estrés hídrico en un suelo salino-sódico. La figura c muestra el N total por nódulo de L. tenuis.  

This occurred in association with an internal recycling of nutrients from necrotic roots and/or an additional amount of absorbed soil P satisfying shoot regrowth demands. Under water stress plus defoliation, defoliated plants were not able to compensate total absorbed P by non-defoliated plants (Figure 2a, page 106 and Table 1, page 104). In this case, the plant strategy focuses on preserving the root system through increasing P allocation to roots while limiting P transport to shoots. This strategy was reflected by a decrease in shoot regrowth and plant P content under water stress and defoliation.

Defoliation induces a reduction in the N uptake and N² fixation capacity of forage legumes ¹⁴ until carbon supply to roots is restored. In the present study, plant N content in L. tenuis under defoliation decreased regardless of water status (Figure 2b, page 106 and Table 1, page 104). Total absorbed N per nodule and number of nodules in L. tenuis roots was similar under defoliation in all irrigation conditions (Figure 2c, 3c, page 106 and Table 1, page 104). These results indicate that the N₂-fixing efficiency of roots of defoliated plants was similar to that of non-defoliated plants. In this situation, defoliated plants mobilized N from roots to shoots to sustain the growth of the latter.

Schedonorus arundinaceus exhibited a similar pattern of plant P and N content in all water treatments (Figure 2 a, b and Table 1, page 104). In this species, root growth maintenance after defoliation allowed plants to maintain or increase soil P and N uptake in response to increased shoot demands. Iqbal et al. (2012) reported that P and N reallocation from roots to shoots would appear to be an essential component of S. arundinaceus tolerance to stressful conditions.

The results of plant growth of both species at high defoliation frequency and water excess are consistent with the proposed hypothesis, but responses to water deficit are inconsistent. In fact, both plant species showed higher STE after the combination of defoliation and water deficit than after that of defoliation and water excess.

Fungal colonization and root nodulation

Defoliation may influence plant function through changes in AM fungal colonization ³. Variations in the mycorrhizal response to the loss of above-ground biomass have been suggested to be due to the amount of tissue removed and root growth rate ¹³. Despite the decrease here observed in root biomass of L. tenuis defoliated plants (Figure 1b, page 104), defoliation frequency increased the AM colonization index in plant roots under well watered and water deficit conditions (Figure 3a, page 107 and Table 1, page 104).

Figure 3: Figura 3: Mycorrhizal colonization (MC) index (a) in both plant species, DSE colonization index in Schedonorus arundinaceus (b) and number of nodules in Lotus tenuis roots (c). FW, fresh weight. Índice de colonización micorrícica (MC) (a) en las raíces de ambas especies, índice de colonización de DSE en las raíces de Schedonorus arundinaceus (b) y número de nódulos en las raíces de Lotus tenuis (c). FW, peso fresco. 

This result contrasts the proposed hypothesis that defoliation decreases AM colonization. A positive effect on AM colonization may be a consequence of increased nutrient demand by defoliated plants ²⁵. In contrast, the AM colonization in L. tenuis roots was affected by the defoliation plus water excess treatment (Figure 3a, page 107), where the combination of stresses seemed to be the most stressful scenario for L. tenuis growth (Figure 1b, page 104). The decrease in AM colonization in defoliated plants was in line with a reduction in root biomass. AM colonization in S. arundinaceus roots under water stress and defoliation was minimal (Figure 3a, page 107 and Table 1, page 104), and similar quantities of root biomass were observed in the three water treatments evaluated (Figure 1b, page 104). Barto and Rilling (2010) found similar results and proposed that root biomass resilience to defoliation may explain the lack of effect on AM colonization of grasses.

In the present study, S. arundinaceus roots were co-colonized by AM fungi and DSE (Figure 3b, page 107 and Table 1, page 104). In agreement with the proposed hypothesis, DSE colonization was affected by defoliation frequency, water stress, and the combination of both stresses. Root DSE colonization in defoliated plants grown on well watered and water-deficit soils increased compared to non-defoliated plants, whereas that under water excess, decreased. Since DSE colonization reached lower values than AM colonization, it may be that AM colonization limits DSE colonization due to competition for host carbohydrates. In a previous field study, S. arundinaceus roots were much more extensively colonized by AM fungi when grown with L. tenuis in co-culture than when grown in monoculture ²¹.

The presence of L. tenuis in mixed communities is crucial to improve not only grassland quality for beef production ⁵, but also to maintain the AM fungal community, especially under water-stress conditions ^9,11. The high percent of AM colonization in L. tenuis roots may be another advantage over the less extensively colonized S. arundinaceus roots, accessing soil moisture and nutrients for plant regrowth, mainly under water deficit, given that S. arundinaceus is somewhat resistant to drought ¹.