Fertilization constitutes a tool to increase the forage supply and consequently, animal production. In this way, not only the nutrients extracted from the soil are restored with the biomass intake by cattle, but the nutritional value and the persistence of grasses are also improved (Neves et al. 2019).
Of all the mineral nutrients that are applied with fertilization, nitrogen is, quantitatively, the most important for forage species, due to its immediate effect on increasing their productivity and quality (Franco et al. 2015). In the specific case of Urochloa, hybrid cv. Mulato II, whose extension in tropical areas has increased due to its high potential of biomass production and nutritional value, the requirements of this element are usually high (Marques et al. 2017).
In spite the mentioned, the growing increase in fertilizer prices, and the need for environmentally friendly technologies, have increased interest in designing fertilization strategies that guarantee adequate nutrition of grasses and forage crops, reduce the use of external supplies and, at the same time, ensure the protection of natural resources (Gupta et al. 2015).
Among these strategies is include biofertilization, due to its potential to increase crop productivity, improve the biological properties of soils and the efficiency of nutrients use (Mahanty et al. 2016 and Finkel et al. 2016). The use of associative bacteria, of the Azospirillum genus, has been shown to increase the performance of forage grasses, due to their ability to fix atmospheric nitrogen, as well as other benefits related to the phytohormones production and the protection of plants against abiotic stresses (Cassan and Diaz-Zorita 2016 and Basu et al. 2017).
The arbuscular mycorrhizal fungi (AMF) are integral components of the grasses rhizosphere, whose plants remain interconnected by a hyphal network that increase the soil volume, explore the roots and facilitate the absorption of nutrients and water (Motta et al. 2017 and Jach-Smith and Jackson 2018), in addition to other important services, such as the restoration of photosynthetic tissues after defoliation (Ambrosino 2018). In fact, the management of biofertilization with these microorganisms has also positive results in increasing yields and improving the absorption of N and other nutrients in grasses (Koziol and Bever 2017 and Teutscherova et al. 2019).
These premises, in addition to the information about the beneficial effects of co-inoculation with A. brasilense and AMF in other agricultural crops (Villarreal et al. 2016 and Lopes et al. 2019), suggest that biofertilization with both microorganisms could contribute significantly important to the reduction of nitrogen fertilization, especially in forage species that require high amounts of this nutrient to achieve higher yields.
The objective of this study was to evaluate the effect of the joint application of the rhizobacteria A. brasilense and the AMF Rhizoglomus irregulare in the reduction of nitrogen fertilization in Urochloa hybrid cv. 36087 (Mulato II).
Materials and Methods
The experiment was carried out in the typical dairy 23, from Unidad Básica de Producción Cooperativa Juan Oramas, located in Guanabacoa municipality, Havana province. The facility is located at 23o08 ’north latitude and 82º11’ west longitude, on a soft brown carbonate soil (Hernández et al. 2015), which the main chemical characteristics are showed in table 1. During the time the experiment was conducted (May 2016-April 2017), the rainfall was 1250 mm. Of this, 75 % occurred during the rainy season (May-October 2016), and the rest between November 2016 and April 2017 (INSMET 2018).
pH H2O | OM (%) | P2O5 | Ca2+ | Mg2+ | Na+ | K+ | BEC |
---|---|---|---|---|---|---|---|
(mg 100 g-1) | (cmolc kg-1) | ||||||
7.7 | 3.85 | 122 | 50.8 | 5.0 | 0.32 | 0.99 | 59.09 |
OM: organic matter, BEC: base exchange capacity
Values in parentheses show confidence intervals (α = 0.05)
Six treatments were evaluated, resulting from the combination of three nitrogen doses (0, 70 and 100 kg N ha-1), alone and combined with the joint application of A. brasilense and R. irregulare, in a random block design with factorial arrangement and four replications. The plots constituted the experimental unit, and had a total area of 21 m2 and a calculation area of 14 m2.
The soil was prepared by plowing (plowing), harrowing, crossing (plowing) and harrowing, at approximately 25 d intervals between each one. The sowing of the grass was carried out in May 2016, in rows separated by 50 cm and in row seeding, with doses of 10 kg of total seed ha-1 (1 kg of pure germinable seed ha-1), at a depth of 1.5 cm.
Before sowing, 10 soil samples were taken with an earth drill, by the zigzag method, at a depth of 0-20 cm. It was determined the pH in H2O (potentiometry, soil-water ratio 1: 2.5), the contents of organic matter (Walkley and Black), assimilable P (extraction with H2SO4 0.5 mol L-1 and colorimetric determination), interchangeable bases (extraction with NH4Ac 1 mol L-1 pH 7, determination by complexometry for Ca and Mg, flame photometry for Na and K) and base exchange capacity (sum of exchangeable bases), according to the analytical techniques established in the soil and plant laboratory of the Instituto Nacional de Ciencias Agrícolas (INCA), described by Paneque et al. (2011).
The doses of nitrogen fertilizer were divided. A 50 % was applied 30 days after sowing and the rest after the first cut, both during the rainy season. For this, small rows, 10 cm deep, were opened with a hoe, 10 cm from the rows of plants, and they were covered after placing the fertilizer. In the dry season, nitrogen fertilizer was not applied, since the experiment was conducted under dry conditions. As carrier of the nitrogen fertilizer, urea was used. Phosphoric and potassium fertilizers were not applied because it was considered that the contents of both nutrients in the soil were sufficient for the grass.
For biofertilization with A. brasilense, the commercial product Nitrofix® was used, from the Instituto Cubano de Investigaciones de los Derivados de la Caña de Azúcar (ICIDCA), which contained strain 8I, with a concentration of 109 CFU mL-1. For the application of the mycorrhizal biofertilizer, the INCAM-11 strain of the AMF species Rhizoglomus irregulare (Sieverding et al. 2014), from the INCA collection, was used. The inoculum was multiplied on a clay substrate, sterilized in an autoclave at 120 °C for one hour for three days, with the use of Urochloa decumbens cv. Basilisk as a host plant. It contained 30 spores per gram of inoculant, as well as abundant fragments of roots and hyphal of the fungus.
The co-inoculation was carried out by the seed coating method. These were immersed in a fluid paste, made by mixing 1 kg of mycorrhizal inoculant, 6 mL of Nitrofix® and 600 mL of water. The seeds were dried in the shade and immediately sown. After the first cut, a mixture of Nitrofix® and water in a 1:10 ratio was prepared. Using a manual knapsack sprayer, it was applied to the soil, very close to the rows, at a rate of 20 L ha-1 of the commercial product.
Four cuts were made, the first at 120 days after sowing and the rest every 60 and 90 days in the rainy and dry season, respectively, at a height of 10 cm with respect to the soil surface. In each cut, the fresh matter of the aerial part of the grass, which occupied the calculation area of the plots, was weighed and 200 g samples were taken. They were taken to an air circulation oven at 70 ºC for 72 h, to determine the dry matter percentage, the dry matter yield, and the concentrations of N, P, K in the biomass (Paneque et al. 2011), as well as the crude protein contents (N x 6.25), organic matter digestibility (Kesting 1977) and NDF (van Soest et al. 1991).
At the time of the second and fourth cut, framed in the rainy and dry periods, respectively, in each plot three sub-samples of roots and soil of the rhizosphere were taken, at a depth of 0‒20 cm by using a metal cylinder 5 cm in diameter and 20 cm in height. The sampling points were distributed equidistant and 10 cm separated from the rows.
The subsamples were homogenized to form a composite sample per plot, and 1 g of rootlets was extracted for staining and clarification (Rodríguez et al. 2015). The frequency of mycorrhizal colonization was evaluated using the intercepts method (Giovanetti and Mosseae 1980), the visual density or colonization intensity (Trouvelot et al. 1986), and the number of spores in the rhizosphere, from sieving and wet decanting of these structures and their observation under a microscope (Herrera et al. 1995).
The statistical processing of the data was carried out using the analysis of variance and Duncan (1955) multiple range test at P <0.05. In the variables corresponding to the chemical characterization of the soil, as well as in those the results are shown in graphs, the confidence interval (α = 0.05) was used, in that order, to estimate the variability of means and as a criterion for their comparison (Payton et al. 2000). In all cases, the statistical program SPSS 25 (2017) was used.
Results and Discussion
Figure 1 shows the interaction between the application of biofertilizers and the doses of N in the fungal variables.
The highest values of the frequency and intensity of mycorrhizal colonization and number of spores in the rhizosphere were reached in the treatments with the application of A. brasilense and R. irregulare without N, or with the addition of 70 kg N ha-1. This effect was maintained until the dry season. With the application of biofertilizers plus the addition of 100 kg N ha-1, in the sampling carried out during the rainy season, there were values of these variables that, although they exceeded the non-inoculated treatments, which showed the level of radical occupation of the AMF residents, were significantly lower than those not biofertilized and accompanied by doses of 0 and 70 kg N ha-1. In this treatment, the effect of biofertilizers in the increase of mycorrhizal structures disappeared in the season of lower rainfalls.
The lower figures of the fungal variables, as well as the lesser permanence of the inoculation effect, observed in the treatment where the biofertilizers and the dose of 100 kg N ha-1 were applied, seem to indicate an ineffective mycorrhizal function, caused by the addition of an amount of nitrogen fertilizer, which could exceed the needs of the biofertilized grass.
It is known that the availability of nutrients in the soil controls the growth of mycorrhizal structures, so that these are reduced when the grasses have been sufficiently fertilized, since the supply of these resources to the host plant through AMF loses importance (Baoming and Bever 2016).
The values of the fungal variables were lower in the dry season than in the rainy season, probably due to the fact that nitrogen fertilizer was not applied during that time, and to the seasonal performance of the biomass production of grasses. During the rains a rapid growth occurs, by virtue of the higher levels of rainfalls, temperature and luminosity, which increases the absorption of nutrients for the biomass formation (Bueno et al. 2019) and, consequently, leads to the formation of larger amounts of mycorrhizal structures to guarantee the plants access to soil resources (Adil et al. 2017).
Table 2 shows the effect of treatments on the macronutrients concentrations (N, P and K) in the aerial biomass of the grass. In the rainy season, there was an increasing response of N concentrations, which reached values of 17.1 g kg-1 of dry matter, with the addition of 100 kg N ha-1. However, with the co-inoculation of biofertilizers, even without the application of nitrogen fertilizer, similar values were obtained. This seems to show that the joint application of both can guarantee adequate nitrogen nutrition of the grass, at least during that time.
Treatments | Rainy season | Dry season | |||||
---|---|---|---|---|---|---|---|
N (kg ha-1) | Biofertilization | N (g kg-1) | P (g kg-1) | K (g kg-1) | N (g kg-1) | P (g kg-1) | K (g kg-1) |
0 | NB | 11.6c | 2.1 | 17.5 | 12.5b | 2.2 | 18.5 |
70 | NB | 14.1b | 2.2 | 16.9 | 13.3b | 2.4 | 17.9 |
100 | NB | 17.1a | 2.0 | 16.7 | 12.9b | 2.3 | 17.8 |
0 |
|
16.7a | 2.2 | 17.2 | 15.5a | 2.2 | 18.4 |
70 |
|
15.8a | 2.1 | 17.0 | 16.9a | 2.3 | 17.1 |
100 |
|
16.3a | 2.2 | 16.7 | 12.3b | 2.4 | 18.3 |
0.2** | 0.1 | 0.3 | 0.3** | 0.1 | 0.4 |
NB: not biofertilized. A. brasilense + R. irregulare: biofertilized with Azospirillum brasilense and R. irregulare. Averages with not common letters in the same column significantly differ at P <0.05, according to Duncan's test.
In the less rainy period, although the effect of nitrogen fertilization on the concentrations of N in the biomass disappeared, due to this nutrient was not applied, the effect of biofertilizers remained, alone or accompanied by the addition of 70 kg N ha-1 in the rainy season.
The treatments did not influence on the concentrations of P and K in the biomass, which join to the values that both elements showed, indicates that the grass was well supplied, based on its high content in the soil (table 1). In fact, this criterion was taken into account in order not to apply phosphoric and potassium fertilizers to the experiment.
The performance of the grass yield was similar to that of the N concentrations in the biomass of the aerial part (table 3). That is, in the rainy season, there was a growing increase in biomass production per unit area, reaching the highest values with the application of 100 kg N ha-1. However, with biofertilizers, the addition of nitrogen fertilizer was unnecessary to reach similar yields to those achieved with the highest dose of N.
Treatments | Rainy season | Dry season | |||
---|---|---|---|---|---|
First cut | Second cut | Third cut | Fourth cut | ||
N (kg ha-1) | Biofertilization | ||||
0 | NB | 5.10c | 4.87c | 2.6b | 2.32b |
70 | NB | 6.29b | 6.03b | 2.72b | 2.28b |
100 | NB | 7.39a | 7.13a | 2.84b | 2.31b |
0 |
|
7.37a | 6.87a | 4.07a | 3.68a |
70 |
|
7.43a | 6.93a | 4.23a | 3.77a |
100 |
|
7.51a | 7.01a | 2.71b | 2.29b |
0.19** | 0.17** | 0.15** | 0.13** |
NB: not biofertilization
A. brasilense + R. irregulare: biofertilized with Azospirillum brasilense and R. irregulare.
DM: dry matter
Averages with not common letters in the same column significantly differ at P <0.05, according to Duncan's test.
When analyzing the influence of the treatments on some variables of the nutritional value of grass (table 4), the applications of N without biofertilizers produced an increasing increase of crude protein content and organic matter digestibility, as well as a significant decrease of the neutral detergent fiber content. However, with the co-inoculation with A. brasilense and R. irregulare, alone or accompanied by nitrogen fertilizer, similar results were obtained to those achieved with the higher dose of N, in the absence of biofertilizers. This effect was more marked during the rainy season, although in the dry season the effect of biofertilizers remained on the crude protein contents of the grass.
Treatments | Rainy season | Dry season | |||||
---|---|---|---|---|---|---|---|
CP (%) | NDF (%) | OMD (%) | CP (%) | NDF (%) | OMD (%) | ||
N (kg ha-1 cut-1) | Biofertilization | ||||||
0 | NB | 7.23c | 70.3a | 62.5b | 7.8b | 71.7 | 61.7 |
70 | NB | 8.81b | 69.9a | 61.9b | 8.31b | 70.4 | 60.6 |
100 | NB | 9.95a | 68.1b | 64.7a | 8.06b | 69.9 | 60.8 |
0 |
|
10.07a | 67.9b | 65.2a | 9. 94a | 71.5 | 61.3 |
75 |
|
9.86a | 68.3b | 65.0a | 10.13a | 70.2 | 60.0 |
100 |
|
10.19a | 67.7b | 64.7a | 7.69b | 69.5 | 61.7 |
0.17** | 0.28** | 0.26** | 0.16** | 0.23 | 0.31 |
NB: noT biofertilized. A. brasilense + R. irregulare: biofertilized with Azospirillum brasilense and R. irregulare.
CP: crude protein; NDF: neutral detergent fiber; OMD: organic matter digestibility
Averages with not common letters in the same column significantly differ at P <0.05, according to Duncan's test.
The CP, OMD and NDF values, obtained with the application of the highest dose of N, coincide with those reported by Leal et al. (2017) and Teixeira et al. (2018) in Urochloa hybrids, fertilized with a similar amount of N. These authors attribute this result to the fact that the addition of this element stimulates the plant growth and increases the use of available carbohydrates for the formation of cells and protoplasm, instead of causing the thickening of the cell wall, which results in the decrease of the fiber content of the grass.
However the above, the most outstanding thing was that co-inoculation with both microorganisms, without additional applications of N, made possible these variables to reach values similar to those obtained with the addition of 100 kg N ha-1. In this way, the reduction of the N dose with the inclusion of biofertilization did not imply a reduction in the nutritional value of the biomass.
When analyzing the results, it was possible to verify the significant contribution of biofertilization to the nitrogen nutrition of the grass. This was evident in the levels of mycorrhizal colonization, which were the result of the effectiveness of the AMF strain R. irregulare, and the possible contribution of A. brasilense to the improvement of the levels of root occupation of the fungus. It has been shown that fungal structures can be used by bacteria as intermediaries to reach the epidermis of root tissue, and that their production of phytohormones stimulates roots growth and, in fact, mycorrhizal colonization (Villarreal et al. 2016). To this is added the contribution of A. brasilense to nitrogen nutrition, from the fixation of significant amounts of N.
In species from Urochloa genus, it has been found that the increase in biomass production, obtained with the inoculation with A. brasilense, can be equivalent to an additional application of 40 kg N ha-1 (Hungria et al. 2016), without rule out other benefits no less important. This is the case of the production of phytohormones and other photosynthates, which could stimulate plants growth (Hanisch et al. 2017 and Oliveira et al. 2018).
The synergistic interaction between both microorganisms could have been favored by the characteristics of the soil where the experiment was carried out. This has a loose soil in the A horizon, with a granular structure, not less than 18 cm thick (Hernández et al. 2015) and organic matter content higher than 3.5 %, which favored the roots growth and the absorption of N from the soil.
The combination of these factors resulted in an effective biofertilization, which guaranteed adequate nitrogen nutrition and the obtaining of biomass yields, with nutritional value indicators similar to those achieved with the application of 100 kg N ha-1 through mineral fertilization, at least during the rainy season. However, it is suggested to carry out longer-term studies to determine the permanence of the effect of biofertilization on the soil-grass system, and quantify its influence on the reduction of nitrogen fertilization over time.