Evaluation of Urochloa decumbens cv . Basilisk in Response to Nitrogen Fertilization and Inoculation With Diazotrophic Bacterium

Nitrogen fertilization provides a great response in pasture productivity and quality but, after applied to the soil, this element undergoes several transformations, what increase its losses. To minimize this problem, a promising alternative currently suggested is diazotrophic bacteria use, which can contribute to a greater use of nitrogen by plants. This study aimed to evaluate the effect of nitrogen doses with and without inoculation of seeds with Azospirillum brasilense on the structural characteristics, chemical composition, and mass production of Urochloa decumbens cv. Basilisk. The experimental design was completely randomized, arranged in a 2 × 5 factorial scheme, with four replications. Treatments consisted of forage seed inoculation or not with Azospirillum and five nitrogen doses (0, 50, 100, 150, and 200 kg ha). The variables analyzed were plant height, number of tillers, shoot dry mass (SDM), root dry mass (RDM), SDM/RDM ratio, chlorophyll index, nutrient content in forage shoot, crude protein (CP), neutral detergent fiber (NDF), and nitrogen use efficiency. The inoculation of forage seeds with A. brasilense associated with nitrogen doses up to 100 kg ha contributed positively to dry mass, plant-shoot nutrient content and bromatological composition of U. decumbens cv. Basilisk. The inoculation of seeds of U. decumbens cv. Basilisk, with A. brasiliense, is a viable alternative for partial substitution of nitrogen fertilization.


Introduction
In Brazil, nearly 180 million hectares have been grown with pastures, mainly grasses of the genus Urochloa (Dias, 2011).The state of Mato Grosso do Sul owns an area of 16 million hectares under pasture, about 50% of which are degraded due to the low system productivity, which compromises the economic potential of livestock (Holsback, 2016).
An improved management of pastures during extended periods seems to be one of the main causes of soil fertility decline (Pereira et al., 2013).A poor input of nutrients and, hence, a decrease in soil organic matter stand out among the fertility decline factors.According to Alcântara et al. (2000), sustainable management of inputs and especially fertilization is the most viable alternative, aiming not only to recover areas already degraded but also to prevent the degradation of new areas.
Nitrogen (N) is the main nutrient responsible for maintaining productivity, being responsible for leaf and stem morphological characteristics, development of tillers, among others (Taiz & Zeiger, 2009).Therefore, N fertilization is among the most important technologies able to improve plant productive potential, besides increasing dry matter production and enhancing forage quality (Juarez Lagunes, Fox, Blake, & Pell, 1999;Martha Júnior & Corsi, 2000;Teutsch, Fike, & Tilson, 2005).For Benett et al. (2008), increasing N doses up to 200 kg ha -1 in U. brizantha cv.Marandu pasture improved its bromatological composition by rising the contents of nitrogen and reducing the levels of neutral detergent fiber (NDF).
Another factor to be considered is the use of N by Urochloa species through N 2 fixation, in which these plants can associate with diazotrophic bacteria (Reis, 2007).Endophytic diazotrophic microorganisms can play an important role in the recovery and sustainability of ecosystems by incorporating atmospheric nitrogen (N 2 ) to the soil, besides producing and releasing plant growth regulators such as auxins, gibberellins, and cytokines.These substances are responsible for increasing the root system and thus improving mineral nutrition and water use by plants (Bazzicalupo & Okon, 2000).
The objective of this study was to evaluate the effect of nitrogen doses, with and without inoculation of the seeds with Azospirillum brasilense, on the structural characteristics, chemical composition, and mass production of Urochloa decumbens cv.Basilisk.

Method
The experiment was carried out in a greenhouse located in the municipality of Dourados, MS (Brazil) (22º11′43.7″S and 54º56′08.5″W, and 452-m altitude).The local climate is classified according to the Köppen's classification (1948) as Cwa-humid mesothermal.
The experimental design was completely randomized, in a 2 × 5 factorial, with four replications.The treatments were composed of two factors: inoculation with Azospirillum brasilense (presence or absence), and nitrogen doses (0, 50, 100, 150 and 200 kg ha -1 ) in topdressing.Each plot consisted of a 5 dm -3 pot, totaling 40 experimental plots.
The soil samples contained in the pots were submitted to two sequential incubations for a period of 30 days each, under humidity conditions equivalent to 60% of the total pore volume (TPV) occupied by water (Freire et al., 1980), controlled by daily weighing.
The first soil incubation was performed after soil liming for soil acidity correction.The liming was made to increase soil saturation to 50%, using a dolomitic limestone at a dosage of 3.5 Mg ha -1 (86% of PRNT, 31% CaO, and 21% MgO).After 30 days, implantation fertilization was performed in each pot applying phosphorus, potassium, sulfur, and micronutrients.These nutrients were applied in the form of PA salts corresponding to the following nutrient contents (in mg dm -3 soil): K (150); P (150); S (62); B (0.81); Cu (1.3); Zn (5.0); Mn (3.6); Fe (1.6), and Mo (0.15), whose sources were K 2 SO 4 , (NH 4 ) 2 HPO 4 , NH 4 NO 3 , KH 2 PO 4 , H 3 PO 4 , The forage Urochloa decumbens cv.Basilisk was sown after incubations, with fifteen seeds sown directly in the pots.The seeds were inoculated with Azospirillum brasiliense in the dosage of 30 g of peat inoculant for each kg of seeds and left for 15 minutes until material adherence.The dose used was 30 g kg -1 of seeds recommended by the product in grasses.Thinning was done ten days after sowing, leaving three forage plants per pot.Pots were maintained with moisture at 60% TPV (Freire et al., 1980).The plants were conducted up to 60 days after sowing.
Topdressing fertilizations with nitrogen were divided in four times.The first application was made at the time of sowing, and the other treatments were done at 10, 20, and 30 days after sowing (DAS).
Plant height was measured with a graded ruler, from the soil to the curvature of the plant canopy.The number of tillers was evaluated by counting all the tillers of the plants per pot.The estimation of chlorophyll content was performed indirectly by the SPAD-502 (Soil Plant Analysis Development) reading, using a chlorophyll meter.The chlorophyll index per experimental plot was determined by the average of five readings per leaf.
After collecting the shoot, the root system was separated by washing the remaining soil in the pot after sampling, under running water and using sieves with a 2 mm mesh.After collection and washing, all the material was dried in an air circulation oven at 65 ºC for 72 hours until reaching constant dry mass (Malavolta, 2006).The material was then weighed for shoot dry mass (SDM) and root dry mass (RDM) determinations, as well as the SDM/RDM ratio.After weighing the samples were ground in a Willey type mill with 1 mm diameter sieves to determine the contents of N, P, K, Ca, and Mg, as described by Malavolta et al. (1997).Crude Protein (CP) was also determined according to procedures described by the Association of Official Analytical Chemists AOAC (1995), and Neutral Detergent Fiber (NDF) was determined as described by Van Soest (1994).The NDF was determined using TNT bags with a porosity of 100 gm -2 , using a fiber determiner (TE-149-Tecnal ® ).
The estimated N use efficiency was obtained by dividing the shoot dry mass per pot by the amount of nitrogen applied per pot, values expressed in g of dry mass per g of N added (Moll, Kamprath, & Jackson, 1982).
The data were submitted to analysis of variance and the means were compared by the F test (p < 0.05) to evaluate two variables using the ASSISTAT statistical software (Silva & Azevedo, 2016).Regression analysis was used for the N doses when the dose significance was verified.The following mathematical model for multiple linear regressions was used in this study.
Where, y is the response variable and xi (i = 1, 2... n) are the explanatory variables.β0 represents the value of y when the explanatory variables are null, the terms βi are called regression coefficients and the residue (ε) is the prediction error, i.e. the difference between the response variable actual and expected values, which is assumed to be normally distributed with mean zero and variance σ2 (Hair Jr., Anderson, Tatham, & Black, 2005).

Results and Discussion
An interaction between nitrogen doses and inoculation with A. brasilense (p < 0.01) was observed for structural characteristics and forage mass production (Figures 1a,1b,1c,1d,and 1e).The variables plant height, number of tillers, shoot dry mass, and root dry mass were described by an increasing linear model (Figures 1a,1b,1c,and 1d).The dose of 200 kg ha -1 provided the highest values for these variables.This is due to the higher availability of N in the soil and its consequent absorption by the plants.A similar result was obtained by Hanisch, Balbinot Junior and Vogt (2017), who observed a linear increase of nitrogen fertilization on the availability of U. brizantha forage in the first evaluation year.
Nitrogen is the main nutrient responsible for the morphological characteristics of leaves and stems size and development of tillers (Taiz & Zeiger, 2009) and of the root system (Monteiro, 2010), since this element is part of the chlorophyll molecule, acting on the formation of substances such as proteins, enzymes, and nucleic acids (Gross, Von Pinho, & Brito, 2006).
The highest average plant height (60.75 cm), number of tillers (36.79 tillers per pot -1 ), shoot dry mass (30.11 g pot -1 ), and root dry mass (32.58 g pot -1 ), were found in treatments inoculated with Azospirillum.These results respectively indicate increases of nearly 6%, 14%, 6%, and 11% if compared to the treatments without inoculation.For these treatments were seen the lowest averages of plant height (57.20 cm), number of tillers (32.25 tillers per pot -1 ), shoot dry mass (28.51 g pot -1 ), and root dry mass (29.29 g pot -1 ).Similar results were found by Guimarães et al. (2011) which saw an increase approximately 8% in the number of leaves and 7% in the number of tillers inoculated in relation to the absence of the bacterium in pastures of Urochloa.
The results found with inoculation can be explained by the improvement in plant growth, mainly of the root system.According to Hungria, Campo, Souza, and Pedrosa (2010), the excretion of plant hormones by bacteria, especially indoleacetic acid, can promote plant growth and increase the absorption of nutrients and water.However, the highest efficiency of the A. brasilense bacterium is observed when associated with lower nitrogen doses (Figures 1a,1b,1c,and 1d).This result is due to the inactivation of the Azospirillum nitrogenase complex under conditions of high ammonia (NH 3 ) concentrations and low oxygen and carbon concentrations (Kavadia, Vayenas, & Aggelis, 2008).  .If the same N dose (184.28 kg ha -1 ) is considered without inoculation the estimated leaf N content will be 27.94 g kg -1 , which generates an increase of approximately 15% in N content, when combined with Azospirillum.
When compared to non-inoculated strains, the inoculated ones showed good efficiency as for BNF.The better response in the presence of Azospirillum is due to the characteristics of this bacterium, because, in addition to the N 2 fixation, it improves the root system, increasing the efficiency of nitrogen fertilization.A quadratic adjustment was observed for the K content without inoculation (Figure 3b), with an estimated maximum point at the dose of 118.67 kg ha -1 of N, with a content of 40.44 g kg -1 .This increase in K content is due to the high correlation between N and K, where there is synergism between applied N and leaf K (Primavesi Figure without

Figure 3 .
Figure 3. Nitrogen (a), potassium (b), copper (c), iron (d), and manganese (e) leaf contents in Urochloa decumbens plants as a function of nitrogen doses with and without inoculation with Azospirillum brasilense