The Enhancement of Soybean Growth and Yield in a Field Trial through Introduction of Mixtures of Bradyrhizobium japonicum , Bacillus sp . and Pseudomonas chlororaphis

The effects of plant growth promoting (PGP) bacteria on soybean growth and yield were tested in field conditions using four treatments: (A) Bradyrhizobium japonicum strain 526, combined with cell-free supernatants (CFS) of Bacillus sp. strain Q10 and Pseudomonas chlororaphis strain Q16; (B) B. japonicum 526 + Bacillus sp. Q10; (C) B. japonicum 526 + P. chlororaphis Q16 and (D) commercial fertilizer containing B. japonicum, which served as a control. The average values of dry weight per nodule and shoot dry weight had the maximum values in the B treatment. In dry shoots collected at the flowering stage, nitrogen and carbon content was similar across all treatments, while that of sulphur decreased in treatment A. Relative to the control (D), all treatments showed positive effects on pods number and grain mass per plant, with the best results yielded by treatment A. Nitrogen and sulphur content in grain were significantly higher in treatment C, whereas maximum carbon content was measured in treatment B. In the control, it was obtained the yield of 4,000 kg/ha, which was in accordance with data reported by the seed producer for the same growing conditions (the maximum value). The yields of 4,229, as well as 4,286 and 4,400 kg ha were measured for variants C, B and A, respectively, which were statistically significant higher (5.73 to 10%) than the commercial fertilizer (D). The improvement in soybean growth and yield in the field trial achieved by addition of PGP strains Bacillus sp. Q10 and P. chlororaphis Q16 to B. japonicum 526 can result in more productive agricultural practices.


Introduction
Plant Growth Promoting Rhizobacteria (PGPR) increase plant growth under some conditions and have beneficial effects on plants.PGPR are a very small portion of rhizobacteria (2-5%) (Antoun, 2013).They affect plant metabolism (by fixing atmospheric nitrogen, solubilizing phosphorus and iron, and producing plant hormones), improve the plant tolerance to stress and prevent the deleterious effects of phytopathogenic microorganisms, thus increasing nutrient availability in the rhizosphere, as well as positively influencing root growth and morphology, and promoting other beneficial plant-microbe symbioses (Vessey 2003;Lugtenberg and Kamilova, 2009;Solomon et al., 2012).Direct plant growth-promoting rhizobacteria enhance plant growth in the absence of pathogens.PGPR include representatives of the following genera: Acinetobacter, Agrobacterium, Arthrobacter, Azoarcus, Azospirillum, Azotobacter, Bacillus, Burkholderia, Enterobacter, Klebsiella, Pseudomonas, Rhizobium, Serratia, and Thiobacillus.However, Pseudomonas and Bacillus species, as well as Streptomyces species, are the bacteria most often found in the rhizosphere of many leguminous and nonleguminous crops (Bouizgarne, 2013).In recent years, interest in the use of PGPR to promote plant growth has increased.Bacteria that belong to Pseudomonas and Bacillus genera can stimulate plant growth and provide protection against pathogens.Mechanisms involved in Bacillus eliciting plant growth promotion include auxin production, increased phosphorus uptake availability, biocontrol abilities and induction of systemic resistance (Idris et al., 2004;Bouizgarne, 2013).Lugtenberg and Kamilova (2009) and Figueiredo et al. (2010) stated that PGPR promote plant growth because they can reduce damage caused by pathogens and therefore act as biopesticides.
As can be seen in the Table 1, according to the soil texture, the soil used belonged to sandy clay loam soil.The results show that the soil is medium heavy in terms of its mechanical properties, with total technical and colloidal clay fraction of about 51.7% and 31%, respectively.In this table the main fertility parameters of the studied soil were presented.The soil was weakly acidic in reaction (pH in 1M KCl = 4.98), and has high humus (SOM) content and low content of available phosphorus and potassium.
Meteorological condition during soybean growing season 2014 in the area western Serbia is shown in Table 2.

Bacterial strains and growth conditions
Bacterial strains used in the present investigation were already tested for PGP activities in earlier studies.The indigenous Bacillus sp.strain Q10 and P. chlororaphis strain Q16 were isolated recently as a part of III46007 Project and their respective PGP activities are described elsewhere (Jošić et al., 2012;Poštić et al., 2013;Pivić et al., 2015;Jošić et al., 2015).Different Bradyrhizobium japonicum strains have been in use in Serbia for more than 60 years in several commercial fertilizers.One of them is B. japonicum strain 526 used in this investigation.
The effects of PGP bacteria were tested using four treatments: A -B. japonicum 526 with CFS of Bacillus sp.Q10 and P. chlororaphis Q16 (2:1:1) B -B. japonicum strain 526 and Bacillus sp.strain Q10 (1:1) C -B. japonicum strain 526 and P. chlororaphis strain Q16 (1:1) D -commercial fertilizer containing B. japonicum, which served as a control.100 mL of each treatment was used for all treatments and mixed with sterile solid carrier (peat coal) before being applied to the seed just before planting.
In the production of leguminous crops, it is necessary to apply the appropriate rhizobial culture for elemental atmospheric nitrogen fixation (Solomon et al., 2012).Bradyrhizobium is capable of forming root nodules on leguminous crops, as well as fixing atmospheric nitrogen, reducing it to ammonia.In the case of soybean (Glycine max L.), it is important to inoculate seeds with relevant strains of Bradyrhizobium japonicum.According to Figueiredo et al. (2010), co-inoculation studies with PGPR and rhizobia have shown increased plant nodulation and N fixation.
Although bacteria from genera Pseudomonas and Bacillus exert a positive influence on the growth of cultivated plants and Bradyrhizobium is important for nitrogen fixation, the aim of the present investigation was to assess the enhancement of soybean growth and yield in the field achieved by addition of Bacillus sp.Q10 and P. chlororaphis Q16 to B. japonicum 526.

Soil properties
The field experiment was conducted in 2014 in the field Vajska (lat.45º 24' N, long.19º 06' E) AP Vojvodina, Serbia, on soybean cv.Angela.Previously, the fields were used for maize cultivation (Zea mays L.).The present study was performed on soil type Humogley (WRB, 2014).
Soil samples were obtained from the surface soil layer (0−0.3 m) in a disturbed condition.Composite soil samples were carried to the laboratory, dried, and passed through a 2 mm sieve (SRPS ISO 11464: 2004).Particle size distribution was determined by sieving and sedimentation (ISO 11277: 2009).pH in water and 1M KCl was analyzed potentiometrically using glass electrode (SRPS ISO 10390: 2007).Carbonate content was determined using Schiebler's calcimeter (SRPS ISO 10693: 2005), while total N was determined by employing elemental CNS analyzer Vario EL III (Nelson and Sommers, 1996).The humus content (SOM) was determined on the basis of total N content.Available P2O5 and K2O were analyzed by AL method according to Egner-Riehm (Riehm, 1958), using 0.1 M lactate (pH = 3.7) as an extract.After the extraction, K was determined by flame emission photometry, and P by spectrophotometer after color development with ammonium-molybdate and SnCl2.Field trial and laboratory testing of plant materials This experiment was conducted on Vajska locality.Each variant was performed in four replicates and every plot (length 97 m; width 9 m) consisted 18 rows.The total area under variant was 0.35 ha. 25 plants per replicate were selected for vegetative characters.One application of herbicides (Oxon 75WG + Pulsar 40 + Harmony 75 WG) and inter-row cultivation was performed without irrigation throughout the growing season.
The plant materials were collected in two vegetative soybean periods -flowering and maturity.At the flowering stage, plant height, trifoliate leaf number, root weight, root length, nodule number and nodule dry weight were measured, as along with nitrogen, carbon and sulphur content in dry shoot.At the maturity stage, plant height, trifoliate leaf number, pod number per plant, and grain mass per plant, along with nitrogen, carbon and sulphur content in grain, were measured.Most importantly, the yield was measured and compared to the value of 4,000 kg ha -1 pertaining to the control treatment, which was in accordance with the maximum value for the same growing conditions reported by the seed producer (Raiffeisen Agro, 2014).The contents of total forms of carbon, nitrogen and sulphur in soybean plant and grain were determined by elemental analysis using CNS analyzer Vario EL III (Benton Jones, 2001).
For statistical analysis, Duncan's Multiple Range Tests were performed to determine the level of significance, which was accepted at p < 0.05 (Statistica ver.12 StatSoft, Inc., Tulsa, Oklahoma, USA).

Soybean plant growth and yield enhancement
The co-inoculated soybean plants showed very similar development to plants inoculated with a commercial fertilizer (D) (Table 3).Trifoliate leaf numbers, root length and nodule numbers per root were not statistically significantly different at the flowering stage.The differences between nodule dry weight values at the flowering stage were statistically significant for all treatments.The shoot dry weight of plants co-inoculated with B. japonicum 526 and Bacillus sp.Q10 (B) was statistically significantly higher than that measured for plants inoculated with commercial fertilizer, as well as those in treatment A and C.
Significant plant development was observed during vegetation and maximum plant height during maturity stage was observed in treatment A, while the trifoliate leaf number and pod number per plant were higher in treatment A and B relative to C and D (Table 4).All mixtures of B. japonicum 526 and Bacillus sp.Q10 or P. chlororaphis Q16 resulted in a statistically significant increase in grain mass per plant; however, the maximum value was obtained for B. japonicum 526 with extracellular metabolites of both PGP strains.
Maximum yield of 4,400 kg ha -1 in the field trial was achieved for treatment A, whereas all co-inoculation variants were more efficient than the commercial fertilizer, with the yield increasing from 5.73 to 10%.Statistically content of total forms of N, C and S in soybean plants at the flowering stage and in grain is shown in Table 5.
Plant growth-promoting bacteria stimulate plant growth and some strains have the ability to fix biological nitrogen.In this work, the effects of Bacillus sp., P. chlororaphis and B. japonicum on soybean growth and yield enhancement in field conditions were tested.According to the obtained results, the average values of dry weight per nodule and shoot dry weight at the flowering stage were significantly higher in plants treated with the mixture of B. japonicum 526 and Bacillus sp.Q10 (B).The root weight, root length and nodule number were not statistically significantly different from those measured for the control, while maximum plant height was obtained in treatment A. According to Vacheron et al. (2013), PGPR modify root functioning, improve plant nutrition, influence vegetative growth and physiology of the whole plant, and play an important role in the plant hormonal network.
In the present study, the nitrogen and carbon content in dry shoots were similar in all treatments, while sulphur content was lower in treatment A (Table 5).Carbon content of 44.02-44.78% in the analyzed soybean dry shoots was similar to 43.6-45.2%reported by Al-Kaisi et al. (2005), but was higher than 41% measured in earlier work of Collins et al. (1997).
Nitrogen content was 4.85-5.08%and was significantly higher than or equal to the values reported by Sholihah et al. (2012).Araujo and Hungria (1999) demonstrated that B. subtilis introduction increases the contribution of the biological nitrogen fixation process in co-inoculated soybean seeds with crude or formulated metabolites.Sulphur content (1.17-1.52%)was significantly higher in our experiment compared to the levels obtained for the application of sulphur for fertilization, which ranged from 0.181 for soybean without S fertilization to 0.237% obtained when 60 kg ha -1 of sulphur was added (Dhage et al., 2014).
During the maturity stage, statistically significant increase in plant height and grain mass per plant was observed in treatment A, as well as in the trifoliate leaf number and pod number per plant in treatment A and B. All applied PGP strains improved soybean growth and yield relative to that obtained when commercial fertilizer was used.The higher yields of 4,229, 4,286 and 4,400 kg ha -1 were measured for variants C, B and A, respectively.These results correspond to those obtained by other authors, who concluded that the application of PGP increased yield (Gholami et al., 2009;Morel et al., 2012).Morel et al. (2012) posited that coinoculation results in a marked increase in legume yield when compared with single inoculation in gnotobiotic laboratory, hydroponics, greenhouse and field conditions.According to the same authors, inoculation and co-inoculation experiments must be performed in field conditions in order to allow for a realistic assessment in practical farming settings.In the field, application of Bradyrhizobium co-inoculated with PGPR, including Pseudomonas spp.and Azospirillum spp., significantly improved yield and yield components of soybean when compared with the sole application of some of applied strains (Son et al., 2006;Seyed Sharifi, 2016).Masciarelli et al. (2014) co-inoculated B. amyloliquefaciens from soybean seeds with B. japonicum, a natural symbiont of soybean and reported enhanced plant growth parameters and significantly improved nodulation.
In the present study, the carbon content in grain ranged from 52.75 to 53.15%, and the maximum value obtained for B treatment was significantly higher than in the remaining treatments.Carbon content in grain was below 54% reported by Connor et al. (2011).Bacillus sp.Q10 included in treatment B was already described as wheat growth promoting strain by Pivić et al. (2015).
Maximum nitrogen and sulphur content in grain was measured in this study in treatment C, whereas P. chlororaphis Q16 promoted not only yield, but soybean seed quality as well.Based on the analysis of 289 grain samples, Salvagiotti et al. (2008) determined the minimum, maximum, median, as well as 25 and 75 percentile value of N. Values obtained in our experiment (6.53-7.24%)exceeded 75%.The S content in grain (0.28-0.33%) measured in our study was at the level reported by Dhage et al. (2014) for grain under fertilization with 20 kg ha -1 S. All treatments were effective in improving sulphur content in soybean grain, suggesting presence of high levels of sulphur-rich amino acids methionine and cysteine.Grain yield and quality increase in treatment C containing P. chlororaphis Q16 may be due to multiple PGP traits, including phosphosolubilization ability, production of indole acetic acid, siderophores, HCN, acyl homoserine lactones, enzymes (protease, chitinase, urease, phosphatase, gelatinase, lipase, pectinase, cellulase, amylase), and phenazine antibiotics, as shown in our earlier investigations (Jošić et al., 2012;Poštić et al., 2013;Jošić et al., 2015).Some PGP traits are responsible for phytostimulation, since production of antibiotics and lytic enzymes is involved in suppression of phytopathogens.P. chlororaphis Q16 have been used for yield improvement of potato, cardoon and wheat, as well as in biological control of phytopathogenic fungi affecting cardoon and wheat (Jošić et al., 2012;Poštić et al., 2013;Pivić et al., 2015).Cattelan et al. (1999) demonstrated that isolates positive for production of 1aminocyclopropane-1-carboxylate deaminase, siderophore and β-1,3-glucanase, and for P solubilization enhanced at least one aspect of early soybean growth.Hernandez et al. (2004) proposed the use of P. chlororaphis PCL1391 strain, which is capable of releasing soluble iron from insoluble ferric oxides, suggesting that phenazines might contribute to iron mobilization in soils.Field trials of pseudomonad strains as seed inoculants can result in growth promotion and have great influence on legume yields (Saharan and Nehra, 2011).

Conclusions
The economic yield and grain quality increases due to coinoculation observed in this field trial were influenced by all treatments using Bacillus sp.Q10 or P. chlororaphis Q16.The maximum yield of 4400 kg ha -1 was recorded in treatment A, comprising of B. japonicum 526 supplemented with CFS of both Bacillus sp.Q10 and P. chlororaphis Q16, representing 10 % increase relative to commercial fertilizer (D).Additional use of PGP Bacillus sp.Q10 and P. chlororaphis Q16 as a B. japonicum co-inoculant can be recommended for further trial to evaluate their suitability for wide soybean production.

Table 1 .
Physical properties and chemical characteristics of the studied soil prior to conducting the experiment (mean ± standard deviation) 275

Table 2 .
Meteorological condition during soybean growing season 2014 in the area western Serbia