Yield of maize hybrids: Is there any association among nitrogen rate, Azospirillum inoculation and fungicide treatment?

Optimization of land use can be attained by incorporating technologies to crop production, such as the use of diazotrophic bacteria, fertilizers, and pesticides. Seed inoculation with Azospirillum is an alternative that favors the incorporation of green agriculture in regions of conventional farming, such as the Brazilian savannah (Cerrado). However, limited information is available about this bacterium’s contribution to agriculture when other technologies are also incorporated. This study evaluated the performance of maize hybrids inoculated, or not, with Azospirillum brasilense, with or without fungicide applications, and subjected to different nitrogen rates under Cerrado field conditions. Each factor analyzed contributes to the increased maize grain yield. The use of inoculants containing plant growth promoting bacteria is a good option to ensure high yield of maize. Still, nitrogen should not be replaced, neither totally nor partially, by seed inoculation with Azospirillum. Fungicide applications should be done, as required, during maize cycle. Moreover, specific maize breeding programs should consider the affinity between Azospirillum strains and maize hybrids, mainly for regions with nitrogen deficient soils, like Cerrado. Thus, by incorporating additional technologies, maize crop farmers can optimize land use and, consequently, reduce the expansion into new agricultural areas.


INTRODUCTION
Optimization of land use has been the focus of international discussions for a long time.Recently, in RIO + 20, once again this aspect was addressed, now 2008; Valipour, 2012Valipour, , 2013;;Sá et al., 2013).Thus, agriculture moves into new areas, turning it in an unsustainable business (Klink and Machado, 2005;Gallardo and Bond, 2011).Furthermore, the global area available for agricultural purposes is becoming increasingly scarce and many experts state that the Brazilian savannah (Cerrado), a biodiversity hotspot, is the last agricultural frontier of the world (CEPF, 2015).
One of the biggest granaries of the world, Cerrado is responsible for most of the Brazilian commodity production, especially soybean and maize (Trivedi et al., 2012;CONAB, 2015a).Current maize production with new hybrids has potential yield between 9 and 15 t ha -1 .However, in Brazil, the average production is approximately 5.1 t ha -1 (CONAB, 2015b), demonstrating that natural resources are poorly managed.Thus, questions about how to solve this problem and how to give maize a label of green agriculture product become important.The immediate answer is the use of technologies that maximize plant genetic potential.One of these technologies, which has been adopted for some time, is seed inoculation with Azospirillum (Bashan et al., 2004;Cavaglieri et al., 2009;Hungria et al., 2010;Hungria, 2011).
Pesticide spraying and fertilization are among the most common practices used in cropping systems.Still, the wide use of pesticides in modern agriculture may cause side-effects on non-target microbiota (Pereyra et al., 2009).In this perspective, seed inoculation with Azospirillum is controversial.Some authors state that interactions between pesticides and microbes are compatible, such as for tebuconazole and A. brasilense sp245 on wheat (Pereyra et al., 2009); while others assert that these interactions are incompatible, such as for carbofuran, chlormephos, terbufos and benfuracarb with A. lipoferum strain CRT1 on maize (Revellin et al., 2001).Similarly, another interesting and controversial issue is the use of fertilizers, especially nitrogen, together with Azospirillum inoculation.Although strains of Azospirillum can improve plant growth and development (Cassán et al., 2009;Hartmann and Bashan, 2009), some studies suggest that nutrient supplementation with mineral fertilizers is needed for greater grain yields (Díaz-Zorita and Fernández-Canigia, 2009), especially for maize (Mehnaz et al., 2010;Ferreira et al., 2013;Myresiotis et al., 2014), in which practices such as fertilization and pesticide application may impair efficacy of treatments with Azospirillum.Studies about the Morais et al. 1151 interaction among these practices in maize are restricted, mainly on field conditions.Thus, questions are raised by farmers and scientists involved in the maize chain.Therefore, the association of Azospirillum seed inoculation, with nitrogen fertilization, and with plant protection in maize production, their combination on yield, and the consequences of such combination in nitrogen use from the physiological point of view were evaluated.

Site description
The study was done at 18°59'02'' S and 47°27'39'' W during the crop season of 2009/2010, under Cerrado field conditions.The region's climate is classified as humid subtropical (Cwa, according to Köppen's climate classification), with average temperature of 22.8°C and precipitation around 1539 mm per year.Weather was ideal for maize crop during the experiment conduction (Figure 1).Previously to the experiment, soil samples were taken arbitrarily from spatially distributed points, from the 0-20 cm layer, and chemically and physically analyzed.Chemical parameters evaluated were soil pH (in H2O), exchangeable P (in Mehlich -1 ), exchangeable K, Ca, Mg and Al.All parameters were analyzed according to the Committee of Soil Fertility of Minas Gerais State (CFSEMG) (1999).The main chemical and physical characteristics of the soil at the establishment of the experiment are shown in Table 1.The soil of the experimental area is classified as an Oxisol.

Experimental model design
A randomized block design was set up, with six replications, in a 4 × 2 × 2 × 6 factorial structure.Four maize hybrids inoculated, or not, with Azospirillum brasilense, with or without fungicide applications, and subjected to different nitrogen rates (50,100,150,200,250 and 300 kg N ha -1 ) were evaluated.Each plot consisted of six 5.2 m long rows, 0.6 m apart, covering an area of 18.7 m 2 per plot and an experimental area of 5,391.4m 2 .The four central rows were used for evaluations, discarding 1 m from each row end.

Seed inoculation
Maize seeds were inoculated with strains of the bacterium A. brasilense (Ab-V5 and Ab-V6) in a minimum concentration of 210 8 viable cells ml -1 .Mixture was carefully done, in plastic bags, to ensure a uniform distribution of the liquid inoculant on the seeds, at a dose equivalent to 100 ml ha -1 .Therefore, theoretical estimate of bacterium cells per seed was 285,714.Maize hybrids used in the study (coded 1 to 4) are genetically modified materials of high yield potential and belong to four different maize breeding companies.The hybrids were selected because they are recommended for Cerrado conditions.

Experiment conduction
Sowing was done immediately after seed inoculation with Azospirillum, in a no-tillage system and an approximate stand of 70,000 plants ha -1 .Basic fertilization was applied at sowing consisting of 625 kg ha -1 of the NPK formula 08-20-20 + 0.5% Zn.When maize plants were at V6 stage (Ritchie et al., 1992), 78 kg K2O ha -1 were applied broadcast, as well as different rates of urea according to the treatment.Herbicides were used for weed control and insects were controlled with biological and chemical insecticides, according to technical recommendations for the crop, described by the pesticides' manufacturers.Fungicide applications were done at V8, VT and R3 stages with a triazole + strobilurinbased product (in treatments with foliar protection).The dose of the commercial product and mineral oil used were 300 ml + 600 ml ha -1 , respectively, at the spray volume 150 L ha -1 .

Harvest
Ears from each experimental plot were mechanically harvested and processed when maize grains with 23% of moisture.Weight and humidity were determined on onboard scale and grain-moisture tester, in the harvester.Data were extrapolated to a one-hectare area and corrected to 13% moisture content, rendering productivity values in kg ha -1 .

Statistical analysis
All assumptions required for the analysis of variance (ANOVA) were confirmed.The error normality was evaluated by Kolmogorov-Smirnov and the variance of homogeneity by Levene, both at 0.01 significance level.Subsequently, the data set was submitted to the ANOVA (Table 2).When significant differences were detected (P ≤ 0.05), averages of inoculation effect and of foliar protection were compared by the Tukey test and averages of nitrogen rates by polynomial regression.All analyses were done at 0.05 significance level.

Effects of nitrogen fertilizer, Azospirillum inoculation, and foliar protection on maize grain yield
Each factor analyzed contributed to the increase of maize grain yield.Besides isolated effects, productivity was affected by the inoculation with Azospirillum combined with the fungicide application, and by the latter factor with the hybrids studied (Table 2).However, the second interaction will not be addressed in this paper since the focus is not the recommendation of maize hybrids, neither the study of their performance in the field, considering that new hybrids are constantly developed and released on the market.Thus, the effects of technologies on maize crop production are emphasized, regardless of the hybrid used by farmers.Nitrogen fertilization promoted greater maize yield.Crop production peaked up to 9.41 t ha -1 at 256 kg N ha  (Figure 2).In contrast, inoculation of maize hybrids with A. brasilense resulted in yield increases varying from 4 to 6% (approximately 400 to 600 kg ha -1 ) (Figure 3).Fungicide applications also contributed to increased crop productivity, regardless of the hybrid tested.Foliar protection promoted an increment of 30% in maize production (Figure 4).Besides the isolated effects of inoculation and foliar protection on hybrids' yield, a noteworthy increase in maize productivity was obtained when both practices were associated (Table 3).This increase was 22% above the control (with neither fungicide spraying nor Azospirillum inoculation).

Effects of Azospirillum inoculation and foliar protection on maize nitrogen use
Regardless of the rate of nitrogen applied, maize grain yield increased due to the inoculation with A. brasilense (Figure 5), indicating better nitrogen fertilizer use by the plants.Fungicide spraying also optimized nitrogen use by the hybrids (Figure 6).Analyzing each nitrogen rate (50, 100, 150, 200, 250 and 300 kg ha -1 ), fungicide use led to increases of 14 to 17% in maize productivity, Table 3. Grain yield of maize hybrids (kg ha -1 ) in response to fungicide applications and A. brasilense inoculation*.corresponding to 1.2 to 1.5 t ha -1

DISCUSSION
The quadratic response of maize yield to increasing rates of nitrogen was also reported by Silva et al. (2005).This result could be explained by ammonia volatilization due to urea application to the soil.Urea hydrolysis raises the pH around the fertilizer granules and converts all of its N content into NH 4 + , which reacts with OH -resulting in H 2 O and volatile NH 3 + , which is phytotoxic.Thus, high nitrogen rates applied via urea can impair plant development, and, consequently, decrease production.
The inoculation of maize hybrids with A. brasilense led to greater yields.This increase varied among the hybrids tested due to their genetic constitution, which is consistent with several studies demonstrating affinity between Azospirillum strains and maize genotypes, altering their responses to inoculation (Salamone and Döbereiner, 1996;Salamone et al., 1996).This also emphasizes that research on selection of bacteria that are able to associate effectively to maize genotypes are essential in order to ensure investment return.
A wide range of responses of cereals to inoculation with Azospirillum is reported.Studies show yield increases varying from 5 to 30% (Okon and Labandera-González, 1994) and from 662 to 823 kg ha -1 in relation to non-inoculated controls (Hungria et al., 2010).In this study, grain yield increases varied from 4 to 6% (which represents 400 to 600 kg ha -1 ), meaning that even highly productive maize genotypes, obtained from conventional and biotechnological breeding, can have yield increased by seed inoculation with Azospirillum.
Therefore, inoculation enabled yield increases of maize crops growing under Cerrado conditions, which resulted from the affinity between Azospirillum and hybrids recommended for the region.Thus, inoculation allowed optimization of land use and even small and medium farmers (in low investment production systems) can obtain greater yields with this technology.
Increased production of maize can be attributed to the phytostimulatory effects of inoculation with Azospirillum, due not only to biological nitrogen fixation in the rhizosphere, but also to plant's greater efficiency in water and nutrient uptake due to greater growth of root system provided by the production of plant growth promoting substances by the bacteria (Döbereiner, 1992;Reis et al., 2000;Cassán et al., 2008).Better nitrogen fertilizer use was observed when maize hybrids were inoculated with Azospirillum.From this result it can be inferred that Azospirillum inoculation enhances nitrogen use, although it does not replace it.Thus, even if part of the maize nitrogen demand is supplied by association with diazotrophic bacteria, reduction of nitrogen fertilizer rates is not recommended.This result, however, contrasts with the one reported by Hungria (2011), who found substantial reduction of nitrogen fertilization in maize plants inoculated with Azospirillum.Applying a nitrogen rate equivalent to half of that recommended for maize in Brazil (100 kg N ha -1 ), the researcher obtained grain yield of 7.8 t ha -1 (Hungria, 2011).However, it must be stated that such production was achieved only with the strain Ab-V5 (+ 54 kg N ha -1 ) in a single crop season.Besides the previously mentioned aspects, foliar protection affected maize yield as well.The hybrids obtained greater yield potential, reaching up to 10.2 t ha -1 , after fungicide applications.This is certainly related to treated plants health.In treatments without fungicide application, hybrid photosynthetic activity may have been compromised, resulting in lower production.This ratifies the idea that investment in plant nutrition is jeopardized if correct phytosanitary management is not adopted.This statement is consistent with studies about effects of fungicide use to increase plant yield (Köhle et al., 2003).
Besides the already known foliar protection, it has been postulated that strobilurin-based fungicides can interfere in the physiology of some crops, such as dry beans (Rava, 2002) and soybean (Fagan et al., 2010), promoting a better fertilizer use by the plants, significantly increasing yields.Therefore, regardless of the N rate applied, this nutrient uptake by maize plants was optimized due to foliar protection (control of diseases) and to physiological effects also provided by the fungicide.These physiological effects comprise an increase in the enzyme nitrate reductase activity (Kaiser and Brendle-Behnisch, 1995), a decrease in ethylene synthesis (Grossmann and Retzlaff, 1997) and a greater plant tolerance to abiotic stresses (Grossmann et al., 1999).This result confirms that of Ruske et al. (2003) while studying the effects of a strobilurin-based fungicide on N uptake, partitioning, remobilization, and grain N accumulation in winter wheat cultivars.Thereby, it is possible to state that foliar protection as well as physiological effects due to the fungicide application positively influenced nitrogen use by maize, increasing yield of the hybrids tested.It is important to emphasize that this better nitrogen use can reduce production costs of maize, avoid degradation of natural resources and increase crop productivity.Foliar spraying with fungicide was not antagonistic to Azospirillum inoculation.Therefore, both technologies can be recommended for greater maize yields.This is important since A. brasilense is not restricted to organic crops and, therefore, will be exposed to a wide variety of pesticides commonly used in intensive agriculture.
Generally, agrochemicals have side-effects on nontarget micro-organisms (Bashan et al., 2007).However, these authors recognize the lack of studies addressing this important issue and that most of them were performed under in vitro conditions.Research on the effects of agricultural pesticides on Azospirillum species are available, focusing on herbicides (Jena et al., 1990;Salmeron et al., 1991;Omar et al., 1992;Rivarola et al., 1992;Forlani et al., 1995) and insecticides (Langenbach et al., 1991;Buff et al., 1992;Sánchez et al., 1994).
As to the effect of the fungicide applied in this study, one could wonder whether its absorption and translocation in the plant could affect bacteria development in maize rhizosphere.However, strobilurins and triazoles have low systemic activity, often bound to the outer layers of plant cuticle, showing limited transport on the boundary leaf layer resulting in long-lasting residual effects on plant pathogens (Köhle et al., 1994).Therefore, Azospirillum cells do not come into direct contact even with fully systemic fungicides as all of these compounds are translocated acropetally into leaves and the shoot tip (Diedhiou et al., 2004).The low or inexistent basipetal transport may explain why foliar fungicide applications have no direct effect on micro-organisms in the root zone (Sicbaldi et al., 1997;Chamberlain et al., 1998).In addition, indirect effects like greater photosynthesis activity of strobilurin-treated plants (Beck et al., 2002) may also be involved, and should promote Azospirillum development due to improved carbohydrate supply to the roots.Those effects could explain the positive interaction observed between Azospirillum inoculation and foliar protection on maize yield.

Conclusions
Each factor analyzed contributes to an increase in maize average yield.Greater interest in the use of inoculants containing plant growth promoting bacteria has been observed and will probably increase in the coming years, due to fertilizer cost, awareness about pollution, and emphasis on sustainable agriculture.However, to ensure high yield, nitrogen rates should not be replaced, neither totally nor partially, by seed inoculation with Azospirillum.Fungicide applications should be done during maize cycle.In addition, specific maize breeding programs should consider the affinity between Azospirillum strains and maize hybrids, mainly for regions with N deficient soils, like Cerrado.Thus, by incorporating additional technologies, maize crop farmers can optimize land use and, consequently, reduce the expansion into new agricultural areas.

Figure 2 .
Figure 2. Grain yield of maize hybrids in response to nitrogen fertilization.

Figure 3 .
Figure 3. Grain yield of maize hybrids in response to A. brasilense inoculation*.*averagesfollowed by different letters, for each hybrid, are statistically different by the Tukey test (P ≤ 0.05).

Figure 4 .
Figure 4. Grain yield of maize hybrids in response to fungicide applications * .* averages followed by different letters, for each hybrid, are statistically different by the Tukey test (P ≤ 0.05).
by different letters, in lowercase in the columns and uppercase in the lines, are statistically different by the Tukey test (P ≤ 0.05).

Figure 5 .
Figure 5. Grain yield of maize hybrids in response to A. brasilense inoculation and nitrogen fertilization*.*differentletters, for each N dose, are statistically different by the Tukey test (P ≤ 0.05).

Figure 6 .
Figure 6.Grain yield of maize hybrids in response to fungicide applications and nitrogen fertilization*.*different letters, for each N dose, are statistically different by the Tukey test (P ≤ 0.05).

Table 2 .
Analysis of variance.
* and ns : significant and not significant by the F test (Snedecor statistics) at 0.05 significance level; DF: Degree of freedom; CV: Coefficient of variation.