Effects of co-inoculation of Bradyrhizobium japonicum SAY3-7 and Streptomyces griseoflavus P4 on plant growth, nodulation, nitrogen fixation, nutrient uptake, and yield of soybean in a field condition

ABSTRACT

Co-inoculation of nitrogen-fixing bacteria with plant growth-promoting bacteria has become more popular than single inoculation of rhizobia or plant-growth-promoting bacteria because of the synergy of these bacteria in increasing soybean yield and nitrogen fixation. This study was conducted to investigate the effects of Bradyrhizobium japonicum SAY3-7 and Streptomyces griseoflavus P4 co-inoculation on plant growth, nodulation, nitrogen fixation, nutrient uptake, and seed yield of the ‘Yezin-6’ soybean cultivar. Nitrogen fixation was measured using the acetylene reduction assay and ureide methods. Uptake of major nutrients [nitrogen (N), phosphorus (P), potassium (K), calcium (Ca), and magnesium (Mg)] was also measured. This study showed that single inoculation of SAY3-7 significantly increased shoot biomass; nodulation; Relative Ureide Index (RUI %), percent nitrogen derived from N fixation (% Ndfa); N, P, K, Ca, and Mg uptakes; during the later growth stages (R3.5 and R5.5), compared with control. These observations indicate that SAY3-7 is an effective N-fixing bacterium for the plant growth, nodulation, and nitrogen fixation with an ability to compete with native bradyrhizobia. Co-inoculation of SAY3-7 and P4 significantly improved nodule number; nodule dry weight; shoot and root biomass; N fixation; N, P, K, Ca, and Mg uptake; at various growth stages and seed yield in ‘Yezin-6’ soybean cultivar compared with the control, but not the single inoculation treatments. Significant differences in plant growth, nodulation, N fixation, nutrient uptake, and yield between co-inoculation and control, not between single inoculation and control, suggest that there is a synergetic effect due to co-inoculation of SAY3-7 and P4. Therefore, we conclude that Myanmar Bradyrhizobium strain SAY3-7 and P4 will be useful as effective inoculants in biofertilizer production in the future.

KEY WORDS:

• Co-inoculation
• nitrogen fixation
• nodulation
• plant growth
• yield

View correction statement:

Corrigendum

1. Introduction

Legumes are the second largest crops in Myanmar after rice according to the cultivated areas. Seventeen kinds of pulse, including soybean, are produced widely in Myanmar. Among them, black gram, green gram, soybean, butter bean, pigeon pea, and cowpea are mainly exported. Due to the relatively low cost of cultivation and increasing demand for domestic consumption and export, cultivation areas for legumes have increased substantially from 0.73 million hectares in 1988–1989 to 5.91 million hectares in 2014–2015 (MOAI 2015). Among these legumes, soybean is a major grain legume crop for domestic consumption. In addition to being a rich source of protein and oil, soybeans are very efficient in fixing nitrogen (N) among the leguminous crops (Unkivich and Pate 2000; Van Kessel and Hartley 2000).

The application of effective inoculants is an alternative strategy for agricultural sustainability and reduction of the environmental impact of chemical fertilizer application due to decreased use (Adesemoye et al. 2009; Hungria et al. 2013). Rhizobia are very important for crop production because biological N fixation, the process by which atmospheric elemental N₂ is converted to ammonia (NH₃) (Franche et al. 2009), provides the required N to leguminous plants and accounts for 65% of the N currently used in agriculture (Matiru and Dakora 2004). Moreover, legume–rhizobia symbiosis is an easy and inexpensive way to maintain soil fertility and improve crop production. Therefore, rhizobium inoculation has become a popular agronomic practice to provide adequate amounts of N to leguminous plants, instead of the use of nitrogenous fertilizer.

Successful rhizobium–legume symbiosis depends largely on whether the host and inoculum are compatible to form N fixation nodules. In the case of soybean symbiosis, this compatibility is related to the nodulation type of bradyrhizobia and nodulation regulatory genes (Rj genes) of the soybean cultivar. Nodulation types and Rj genes have been reported to play roles in the control of nodule formation (Ishizuka et al. 1991a, 1991b). Type A strains induce nodulation on all Rj genotype cultivars but prefer non-Rj soybean cultivars. Type B and C strains have restricted nodule formation on Rj ₂ Rj ₃ and Rj ₄ genotype cultivars, respectively, preferring the Rj ₄ and Rj ₂ Rj ₃ genotype cultivars, respectively, for nodulation. Therefore, type A, B, and C strains must be inoculated with their specific soybean cultivars to successfully nodulate and fix N.

Co-inoculation of N-fixing bacteria with plant-growth-promoting bacteria has become more popular than single inoculation of rhizobia or plant-growth-promoting bacteria because co-inoculation increases soybean yield and improves the sustainability of agriculture (Hungria et al. 2015). Vessey and Buss (2002) reported that co-inoculation of rhizobia with plant-growth-promoting bacteria improved nodulation and N fixation. Streptomyces griseoflavus P4, which is an endophytic bacterium, is known to enhance symbiotic nitrogen fixation of some bradyrhizobia under controlled greenhouse and open field conditions (Soe et al. 2012; Soe and Yamakawa 2013a; Soe and Yamakawa 2013b). Our previous research demonstrated that co-inoculation of Bradyrhizobium japonicum SAY3-7 with the endophytic bacterium Streptomyces griseoflavus P4 increases N fixation rates and plant growth compared with single inoculation of B. japonicum SAY3-7 under environmentally controlled conditions (Htwe and Yamakawa 2016). However, field investigation is needed to determine the performance of single and co-inoculation of SAY3-7 and/or P4 under open field conditions. Therefore, this study was conducted to investigate the effects of co-inoculation of SAY3-7 and P4 on plant growth, nodulation, N fixation, nutrient uptake, and seed yield of the ‘Yezin-6’ soybean cultivar.

2. Materials and methods

2.1. Experimental site

The experiment was conducted at Kyushu University Farm, Fukuoka Prefecture, Japan (33°37ʹN, 130°25ʹE) from July to November 2016.

2.2. Experimental design and treatments

In this study, a randomized complete block design was used with three replications. The four treatments were an uninoculated control, single inoculation of P4, single inoculation of SAY3-7, and co-inoculation of SAY3-7 and P4.

2.3. Soil analysis

Before conducting the field experiment, soil samples were collected from five locations in the field at 10 cm depth using a 5-cm-diameter soil sampling tube. The soil samples were spread and air dried at room temperature for 24 h. Then, they were crushed by hand and sieved through a 2-mm-mesh sieve. They were stored at 4°C before the soil analysis. Soil $\mathrm{p}{\mathrm{H}}_{{\mathrm{H}}_2\mathrm{O}}$ (1:2.5 soil:H₂O) was measured using a pH meter. The soils were digested using the salicylic acid–H₂SO₄–H₂O₂ digestion method (Ohyama et al. 1991); total N was determined using the indophenol method (Cataldo et al. 1974); and total phosphorus (P) was analyzed using the ascorbic acid method (Murphy and Riley 1962). Mineralisable N was assessed using the soil incubation method (Sahrawat 1983) followed by Cataldo et al. (1974). Available P was measured using Truog’s method (Truog 1930). Cation exchange capacity and exchangeable cations were determined using the ammonium acetate shaking extraction method (Muramoto et al. 1992). Exchangeable cations were measured by atomic absorption spectrophotometry (Z-5300; Hitachi, Tokyo, Japan). The physicochemical properties of the experimental soils are described in Table 1.

Table 1. Physicochemical properties of soil.

Physicochemical property (field)	Values
Soil pH (Soil:H2O; 1:2.5)	6.38
Total N (%)	0.15
Total P2O5 (%)	0.28
Available N (mg N/100 g soil)	0.07
Available P (mg P2O5/100 g soil)	29.75
CEC (cmolc kg^−1)	10.05
Exc. K (cmolc kg^−1)	0.12
Exc. Ca (cmolc kg^−1)	2.84
Exc. Mg (cmolc kg^−1)	0.28

The rhizobial populations of the collected soil samples were evaluated by the most probable number (MPN) method (Vincent 1970) using ‘Yezin-6’ (non-Rj) as the host plant.

2.4. Bacterial inoculant preparation

B. japonicum SAY3-7 (type A) strain isolated from soybean at Aungban City, Myanmar (Htwe et al. 2015). The type A described in parentheses expresses the nodulation type of Bradyrhizobium strain and it was identified by Htwe et al. (2015). B. japonicum SAY3-7 was incubated in A1E broth culture medium (Kuykendall 1987) on a rotary shaker at 30°C for 7 days. S. griseoflavus P4 isolated from sweet pea root at Kurima, Tsu City, Japan (Thapanapongworakul 2003), was incubated in IMA-2 broth culture medium (Shimizu et al. 2000) on a rotary shaker at 30°C for 7 days. A peat-based inoculum was used in this experiment. Peat soil was collected from Heho, Myanmar. A total of 100 soybean seeds were mixed thoroughly with 10 g peat soil, 7 mL 20% liquid solution of gum arabic, and 0.1 mL 1 × 10⁸ cell mL⁻¹ SAY3-7 and/or 0.1 mL 1 × 10⁸ cell mL⁻¹ P4 to obtain the required inoculation density (1 × 10⁵ cells seed⁻¹) for both bacteria during preparation of the peat-based inoculum. Our previous experiment highlighted that the proper inoculation density for SAY3-7 and P4 was 1 × 10⁵ cells mL⁻¹ for co-inoculation.

2.5. Soybean variety

The soybean ‘Yezin-6’ (non-Rj) was used as the host plant. Rj gene described in parentheses indicates the nodulation regulatory gene of soybean and it was identified by Htwe et al. (2015). Yezin-6 is one of the most widely grown soybean cultivars in Myanmar. The term ‘non-Rj’ in parentheses expresses the nodulation regulatory gene. Although non-Rj is compatible with all types of bacterium (Vest and Caldwell 1972), it prefers type A strain bacteria for nodulation. Therefore, the non-Rj soybean cultivar is suitable for use as a host plant when inoculating SAY3-7 (type A strain).

2.6. Cultivation and crop management

The experimental field was prepared for conventional practice in Japan. Compound fertilizer (Kumiai Mame-kasei 300, Ryoto Fertilizer Co., Ltd., Ooita, Japan) which contains 3% N, 10% P₂O₅, and 10% K₂O was applied at a rate of 800 kg ha⁻¹ 1 week before the final land preparation. After the basal application of fertilizer, the experimental field was leveled and divided into individual plots. The experimental site was 15 m long and 20 m wide. Each plot was 3 m long and 6 m wide. The row and plant spacing were 60 and 40 cm, respectively. Four inoculated seeds were sowed in one hill and covered with soil just after seed sowing. To eliminate contamination, seeds were first sown in control plots, followed by the single inoculation of P4, single inoculation of SAY3-7, and co-inoculation of SAY3-7 and P4. After sowing, granular forms of herbicide and pesticide were applied to prevent competition with weeds and to prevent insect and pest infestation. Sixteen days after sowing (DAS), the first, inter-cultivation was done using an inter-cultivator to control weeds and reduce competition with the soybean plants. At 28 DAS, the second inter-cultivation was done for earthing-up. Then, plants were thinned as necessary to maintain two plants per hill. Pesticides were sprayed fortnightly from the flowering stage (R2) to the beginning of maturation (R7).

2.7. Plant sampling

Plant samples were collected from five growing stages: V6 (six unfolded trifoliate leaves), R2 (Full flowering stage), R3.5 (early pod-fill stage), R5.5 (early seed-fill stage), and R8 (maturity stage). The growth stages refer to Fehr et al. (1971).

2.8. Acetylene reduction assay

Two plants at the V6 and R2 stages from one hill in each plot were uprooted and carefully washed with water so as not to detach the nodules. The acetylene reduction assay (ARA) was performed according to Haider et al. (1991) to measure nitrogenase activity. The soybean plants were cut at the cotyledonary nodes. Then, the soybean root with intact nodules was placed in a 200-mL conical flask and sealed with a serum stopper. A 25-mL aliquot of acetylene (C₂H₂) gas was injected into the flask to replace the air with acetylene. The flasks containing roots with intact nodules were incubated at room temperature and 1.0-mL subsamples were analyzed at 5 and 65 min, respectively. The ARA value, in terms of ethylene (C₂H₄) production per plant, was measured using a flame ionization gas chromatograph (GC-14A, Shimadzu, Kyoto, Japan) equipped with a stainless steel column (3 mm diameter, 0.5 m length). The column was filled with Porapak R 60–80 mesh (Nacalai Tesque, Inc., Kyoto Japan). Colum, injection, and detection temperatures were 35, 45, and 170°C, respectively. N gas was used as the carrier gas at a flow rate of 45 mL min⁻¹. The number of nodules was counted after the assay. Shoots, roots, and nodules were collected separately and oven dried at 70°C for 72 h to record their dry weights.

2.9. Xylem sap collection

Two plants at the R3.5 and R5.5 stages from one hill in each plot were cut just under the cotyledonary nodes and inserted into a silicon tube. The xylem sap was collected for 1 and 2 h after cutting at the R3.5 and R5.5 stages, respectively. The sap samples were stored at −30°C for long-term use. Amino acid (Moore and Stein 1954), nitrate (Cataldo et al. 1975), and ureide (Young and Conway 1942) were analyzed from the root-bled sap. The relative ureide index (RUI) of the root-bled sap was calculated using the following formula (Peoples et al. 1989):

$\mathrm{R}\mathrm{U}\mathrm{I}\left(\mathrm{\%}\right)=4\times \frac{\mathrm{u}\mathrm{r}\mathrm{e}\mathrm{i}\mathrm{d}\mathrm{e}}{\left(4\times \mathrm{u}\mathrm{r}\mathrm{e}\mathrm{i}\mathrm{d}\mathrm{e}+\mathrm{a}\mathrm{m}\mathrm{i}\mathrm{n}\mathrm{o}\mathrm{a}\mathrm{c}\mathrm{i}\mathrm{d}+\mathrm{n}\mathrm{i}\mathrm{t}\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{e}\right)}\times 100.$

The percentage of N derived from N fixation was calculated using the following formula (Herridge and Peoples 1990):

$y=21.3+0.67x,$

where y is RUI (%) and x is the percentage of N derived from N fixation (%Ndfa), respectively. After xylem sap collection, the roots were uprooted and washed. Then, the nodules were counted. The shoots, roots, and nodules were oven dried at 70°C for 72 h to record their dry weights.

2.10. Measurements of total N, P, K, Ca, and Mg uptake

The shoots were divided into leaves, stems and petioles, shells, unfilled seeds, and filled seeds. Then, each plant part was dried and separately ground into a powder using a mill (100–120 mesh, Tecator AB, Hoedanaes, Sweden). After digestion of the nutrients using the H₂SO₄–H₂O₂ digestion method (Ohyama et al. 1991), total N accumulation in the shoot was measured by the indophenol method (Cataldo et al. 1974); total P was analyzed using the ascorbic acid method (Murphy and Riley 1962); and total K, Ca, and Mg were measured using atomic absorption spectrophotometry (Z-5300, Hitachi, Tokyo, Japan).

2.11. Yield and yield components

Four hills from each plot at R8 were selected randomly and plants were cut at the cotyledon nodes to determine seed yield and yield component parameters, such as the number of pods per plant, number of seeds per pod, 100-seed weight.

2.12. Data analysis

Data were statistically analyzed using STATISTIX 8 (Analytical Software, Tallahassee, FL, USA) and the means were compared by Tukey’s HSD test at P < 0.05.

3. Results

3.1. Indigenous rhizobia

Before cultivation, the population of indigenous rhizobia in the soil from the experimental field was estimated by the MPN method using the Yezin-6 cultivar as the host plant. The population density of indigenous rhizobia was 0.85 × 10⁷ rhizobia in 1 g dried soil.

3.2. Effects of single and co-inoculation on biomass production at different growth stages

The results of shoot and root biomass production are shown in Table 2. No difference in shoot and root dry biomass was observed at the V6 and R2 stage. Significantly different shoot and root dry biomasses were detected at the R3.5, R5.5, and R8 stages. A synergistic effect was formed by combined inoculation of both bacteria as significant increase in shoot dry weight and root dry weight was observed between co-inoculation and control, but not between single inoculation and control at R5.5 and R8 stages.

Table 2. Effect of co-inoculation of B. japonicum SAY3-7 and S. griseoflavus on nodule number, nodule, shoot, and root dry weights of Yezin-6 soybean cultivar at different growth stages.

Growth stage	Treatment	NN (No. hill^−1)	NDW (g hill^−1)	RDW (g hill^−1)	SDW (g hill^−1)
V6	Control	36.7 ± 0.5b	0.10 ± 0.03b	0.97 ± 0.28a	4.15 ± 1.34a
	P4	55.3 ± 16.5ab	0.19 ± 0.03ab	1.07 ± 0.02a	4.57 ± 0.37a
	SAY3-7	55.7 ± 6.3ab	0.14 ± 0.03ab	0.93 ± 0.21a	3.79 ± 0.25a
	SAY3-7 + P4	86.0 ± 10.2a	0.22 ± 0.02a	1.17 ± 0.19a	5.16 ± 0.61a
R2	Control	33.3 ± 6.2c	0.18 ± 0.08a	2.00 ± 0.55a	9.30 ± 0.48a
	P4	49.0 ± 6.2bc	0.22 ± 0.16a	2.13 ± 0.36a	11.34 ± 1.89a
	SAY3-7	82.7 ± 18.9ab	0.31 ± 0.01a	2.37 ± 0.23a	12.91 ± 0.75a
	SAY3-7 + P4	97.3 ± 9.2a	0.29 ± 0.15a	2.54 ± 0.16a	13.09 ± 0.79a
R3.5	Control	337.7 ± 17.9b	1.32 ± 0.17a	6.20 ± 1.01b	44.39 ± 5.47c
	P4	480.7 ± 49.8ab	1.50 ± 0.60a	7. 21 ± 0.81ab	52.18 ± 4.95bc
	SAY3-7	498.0 ± 60.6a	2.16 ± 0.34a	9.96 ± 0.92a	76.51 ± 8.61ab
	SAY3-7 + P4	536.0 ± 49.2a	2.21 ± 0.63a	10.60 ± 0.91a	80.04 ± 9.73a
R5.5	Control	430.3 ± 56.2b	2.21 ± 0.69b	7.80 ± 0.47b	101.20 ± 8.90b
	P4	512.7 ± 35.0ab	2.68 ± 0.23b	8.72 ± 1.70ab	97.12 ± 19.46b
	SAY3-7	585.7 ± 34.7a	2.87 ± 0.34b	10.38 ± 0.23ab	117.52 ± 4.25ab
	SAY3-7 + P4	638.7 ± 15.2a	4.63 ± 0.18a	11.30 ± 1.33a	139.26 ± 9.12a
R8	Control	–	–	–	123.74 ± 5.56b
	P4	–	–	–	139.38 ± 11.93ab
	SAY3-7	–	–	–	157.39 ± 10.00ab
	SAY3-7 + P4	–	–	–	172.40 ± 6.79a

Mean values ± standard deviation (SD) in each column at each growth stage followed by the same letters are not significantly different at P < 0.05 (Tukey’s test). NN: nodule number; NDW: nodule dry weight; RDW: root dry weight; SDW: shoot dry weight.

3.3. Effects of single and co-inoculation on nodulation at different growth stages

The nodule count and nodule dry weight results are shown in Table 2. Nodulation, in terms of nodule number and nodule dry weight, differed significantly among treatments at the V6 stage. Co-inoculation of SAY3-7 and P4 resulted in significantly more nodules and a higher nodule dry weight compared with the control treatment, but not the single inoculation treatments. At R2 and R3.5 stages, the numbers of nodules, but not nodule dry weight, differed significantly among treatments. The number of nodules and nodule dry weight differed significantly among treatments at the R5.5 stage. Co-inoculation of SAY3-7 and P4 produced the greatest nodule dry weight compared with other treatments. Moreover, co-inoculation resulted in significantly higher nodules than control treatment. The abundance of nodulation was occurred due to co-inoculation compared with control, but not the single inoculation at each growth stages. These results indicate that a synergetic effect was formed by co-inoculation.

3.4. Effects of single and co-inoculation on nitrogen fixation at different growth stages

The results of N fixation in terms of C₂H₄ production are shown in Fig. 1A. N fixation at the V6 and R2 stages in terms of C₂H₄ production was measured. C₂H₄ production at the V6 stage did not differ significantly among treatments. N fixation at the R2 stage in terms of C₂H₄ production differed significantly among treatments. Co-inoculation significantly increased N fixation by approximately 47%, compared with the control.

Figure 1. Effect of co-inoculation of B. japonicum SAY3-7 and S. griseoflavus on (A) acetylene reduction activity (ARA), (B) Relative Ureide Index (%), (C) N derived from N fixation (%Ndfa) of Yezin-6 soybean cultivar at different growth stages. The histograms with the same letter at each growth stage are not significantly different at P < 0.05 (Tukey’s test).

The results of N fixation in terms of the RUI (%) are shown in Fig. 1B. The RUI (%) at the R3.5 stage differed significantly among treatments. Single inoculation of SAY3-7 and co-inoculation of SAY3-7 and P4 resulted in significantly higher RUIs (%) of 88.21% and 88.30%, respectively, compared with the control (74.43%). The RUI (%) at the V5.5 stage did not differ among treatments.

The N fixation results in terms of %Ndfa are shown in Fig. 1C. %Ndfa at the R3.5 stage differed significantly among the treatments. Single inoculation of SAY3-7 and co-inoculation of SAY3-7 and P4 resulted in significantly higher %Ndfa values of 99.87% and 99.99%, respectively, compared with those of the control (79.30%). %Ndfa did not differ among treatments at the R5.5 stage.

3.5. Effects of single and co-inoculation on N, P, K, Ca, and Mg uptake at different growth stages

The results of N uptake are shown in Fig. 2. The uptake results for other nutrients are shown in Table 3. Total N, P, K, Ca, and Mg uptakes were not significant difference from other treatments at V6 and R2 stages. However, significantly different total N, P, K, Ca, and Mg uptakes occurred at the R3.5 and R5.5 stages. Total N, P, and K uptakes at the R8 stage differed significantly among treatments. However, Ca and Mg uptakes did not differ among treatments. Moreover, K, Ca, and Mg uptakes decreased at the R8 stage compared with those at the R5.5 stage. N and P uptakes did not decline until maturation (R8 stage).

Table 3. Effect of co-inoculation of B. japonicum SAY3-7 and S. griseoflavus on P, K, Ca, and Mg uptakes of Yezin-6 soybean cultivar at different growth stages.

Growth stage	Treatment	P uptake (g hill^−1)	K uptake (g hill^−1)	Ca uptake (g hill^−1)	Mg uptake (g hill^−1)
V6	Control	0.02 ± 0.003a	0.08 ± 0.036a	0.04 ± 0.012a	0.02 ± 0.005a
	P4	0.02 ± 0.002a	0.09 ± 0.010a	0.05 ± 0.005a	0.02 ± 0.001a
	SAY3-7	0.01 ± 0.001a	0.07 ± 0.016a	0.04 ± 0.004a	0.01 ± 0.001a
	SAY3-7 + P4	0.02 ± 0.004a	0.09 ± 0.020a	0.05 ± 0.006a	0.02 ± 0.003a
R2	Control	0.03 ± 0.003a	0.16 ± 0.013a	0.10 ± 0.009a	0.03 ± 0.003a
	P4	0.04 ± 0.011a	0.24 ± 0.054a	0.11 ± 0.024a	0.04 ± 0.008a
	SAY3-7	0.05 ± 0.009a	0.27 ± 0.035a	0.13 ± 0.003a	0.04 ± 0.001a
	SAY3-7 + P4	0.05 ± 0.002a	0.26 ± 0.020a	0.13 ± 0.011a	0.04 ± 0.004a
R3.5	Control	0.14 ± 0.020c	0.82 ± 0.109b	0.39 ± 0.041b	0.14 ± 0.018b
	P4	0.18 ± 0.008bc	0.91 ± 0.074b	0.48 ± 0.012b	0.16 ± 0.018b
	SAY3-7	0.27 ± 0.025ab	1.43 ± 0.064a	0.69 ± 0.046a	0.25 ± 0.023a
	SAY3-7 + P4	0.30 ± 0.048a	1.60 ± 0.182a	0.76 ± 0.020a	0.31 ± 0.003a
R5.5	Control	0.39 ± 0.031b	1.35 ± 0.090b	0.74 ± 0.081b	0.24 ± 0.023b
	P4	0.38 ± 0.067b	1.26 ± 0.239b	0.73 ± 0.154b	0.23 ± 0.050b
	SAY3-7	0.45 ± 0.019ab	1.57 ± 0.051ab	0.88 ± 0.035ab	0.28 ± 0.024ab
	SAY3-7 + P4	0.54 ± 0.031a	1.98 ± 0.115a	1.07 ± 0.096a	0.36 ± 0.038a
R8	Control	0.55 ± 0.026b	0.92 ± 0.083b	0.42 ± 0.023a	0.27 ± 0.005a
	P4	0.63 ± 0.043ab	1.03 ± 0.044ab	0.48 ± 0.068a	0.30 ± 0.034a
	SAY3-7	0.74 ± 0.044a	1.27 ± 0.134a	0.56 ± 0.046a	0.34 ± 0.032a
	SAY3-7 + P4	0.73 ± 0.021a	1.09 ± 0.057ab	0.53 ± 0.032a	0.34 ± 0.018a

Mean values ± standard deviation (SD) in each column at each growth stage followed by the same letters are not significantly different at P < 0.05 (Tukey’s test).

Figure 2. Effect of co-inoculation of B. japonicum SAY3-7 and S. griseoflavus on nitrogen uptake of Yezin-6 soybean cultivar at different growth stages. The histograms with the same letter at each growth stage are not significantly different at P < 0.05 (Tukey’s test).

3.6. Effects of single and co-inoculation on yield at maturity stage

The yield and yield component results are shown in Table 4. Yield component parameters, such as the number of pods per plant, number of seeds per pod, and 100-seed weight, did not differ significantly among treatments. However, co-inoculation of SAY3-7 with P4 resulted in significantly higher yield (3.11 t ha⁻¹) than the control (2.28 t ha⁻¹). In this study, significant difference in yield was observed between co-inoculation and control, not between co-inoculation and single inoculations. Significant difference in yield between co-inoculation and control, not between single inoculation and control, suggests that there is a synergetic effect in increasing soybean productivity due to co-inoculation of SAY3-7 and P4.

Table 4. Effect of co-inoculation of B. japonicum SAY3-7 and S. griseoflavus on yield and yield components of Yezin-6 soybean cultivar at maturity stage.

Treatment	Pods (No. hill^−1)	Seeds per pod (No. pod^−1)	100 seed weight (g)	Yield (t ha^−1)
Control	225.6 ± 5.4a	1.5 ± 0.1a	16.41 ± 0.39a	2.28 ± 0.21b
P4	256.3 ± 35.8a	1.5 ± 0.1a	16.79 ± 2.25a	2.69 ± 0.07ab
SAY3-7	273.5 ± 10.6a	1.5 ± 0.1a	17.85 ± 1.64a	2.96 ± 0.36ab
SAY3-7 + P4	277.7 ± 58.7a	1.6 ± 0.4a	18.33 ± 0.70a	3.11 ± 0.07a

Mean values ± standard deviation (SD) in each column followed by the same letters are not significantly different at P < 0.05 (Tukey’s test).

4. Discussion

The combination of P4 with SAY3-7 resulted in significant increases in shoot and root dry biomasses at R3.5, R5.5, and R8 (Table 2). These findings are similar to those of our previous studies, in which the combination of P4 and SAY3-7 had a synergistic effect on shoot and root dry weights (Htwe and Yamakawa 2015, 2016). This plant-growth-promoting effect of P4 might be due to secretion of growth-promoting hormones. The P4 used in this experiment can induce production of the plant growth hormone indole acetic acid (Soe 2013).

The number of nodules and nodule dry weight was significantly increased due to after co-inoculation compared with the other treatments at some growth stages (Table 2). Similar findings were documented by Soe et al. (2012), in which co-inoculation of P4 with USDA110 significantly increased the nodule dry weight of a Myanmar soybean cultivar. Aung et al. (2013) reported a significant increase in nodulation of a Myanmar soybean cultivar due to co-inoculation of Azospirillum sp. with B. japonicum USDA110, compared with the uninoculated control and single inoculations of each bacterium in rhizobia-established soils collected from Myanmar and Thailand. In this study, soybean plants in the un-inoculated control and the P4 treatment group formed root nodules via indigenous rhizobia because the plants inoculated with P4 and control did not form nodules in sterilized vermiculite media (Htwe and Yamakawa 2016). One of the major problems in inoculation technology with soybean is the establishment of an inoculated strain from populations of indigenous bradyrhizobia existing in the soils (Tang 1979). Therefore, the occupancy of inoculated strains in nodules is relatively low compared with indigenous rhizobia because of their competition to occupy in root nodules (Weaver and Frederick 1974; Kvien et al. 1981). However, in this study, nodulation in control treatment was significantly lower than co-inoculation. The abundance of nodulation in co-inoculation treatment shows that the inoculated strain had the competitive ability with indigenous rhizobia for nodulation.

N fixation has been measured using various methods, such as N-difference and ureide methods, as well as ARA and ¹⁵N isotope methods. In this study, ARA and ureide method were used for assessment of nitrogen fixation. The ethylene production will be governed by the proportion of active nodules that remained intact on the plants in uprooting process (Danso 1995). At the early growth stages, most of the nodules were remained intact by carefully uprooting whereas nodules were easily detached from the plant at the later growth stages because a large number of nodules were formed on lateral roots. Danso (1995) stated that it is very difficult to recover 100% of intact nodules even in normal friable soil, especially where a large number of nodules are formed on lateral roots. In contract, root bled saps were difficult to collect at early growth stages because of lower flow rate, but not at later growth stages. Therefore, ARA and ureide method are suitable for early growth stages (V6 and R2) and the later growth stages (R3.5 and R5.5), respectively. For ARA measurement, uprooted plants with intact nodules were used to detect the acetylene reduction to ethylene. In this study, co-inoculation of Bradyrhizobium with P4 has been reported to have a synergistic effect on symbiotic N fixation by increasing nitrogenase activity (Soe and Yamakawa 2013a). These findings are in agreement with our results, in which co-inoculation of B. japonicum SAY3-7 with the plant-growth-promoting bacterium S. griseoflavus P4 significantly increased nitrogenase activity at the R2 stage (Fig. 1A). Our previous study showed that low-density co-inoculation of B. japonicum SAY3-7 and P4 improved nitrogenase activity by 15–75% in various soybean cultivars compared with single inoculation of SAY3-7 under environmentally controlled conditions (Htwe and Yamakawa 2016). Yamakawa et al. (2000) revealed that ureide is the main nitrogenous compound transported in xylem sap of the soybean. Streeter (1979) also stated that ureide-N, allantoin, and allantoic acid are the major forms of N transported from nodulated roots to shoots of soybean. Therefore, the RUI has become one of the most widely used methods to measure N fixation, particularly in soybean. In this study, RUI (%) was determined at the R3.5 and R5.5 stages. Our results show that the RUI (%) increased significantly in soybean plants inoculated with SAY3-7 and SAY3-7 + P4 at the R3.5 stage, although no significant increase occurred at the R5.5 stage (Fig. 1B). Consequently, N derived from N fixation was significantly increased at the R3.5 stage, but not at the R5.5 stage, compared with the control (Fig. 1C). These results support the findings of Soe et al. (2012) and Soe and Yamakawa (2013b), in which single and co-inoculation of bradyrhizobia with P4 promoted higher RUI (%) and %Ndf at the R3.5 stage.

Symbiotic N fixation of soybean can provide 40–70% of the total N requirement (Klubeck et al. 1988). In this study, single and co-inoculation of SAY3-7 improved total N uptakes in the shoot at R3.5, R5.5, and R8 stages (Fig. 2). Similarly, Soe and Yamakawa (2013b) reported that total N accumulation was significantly affected by single and co-inoculation at the V6 and R8 stages. Moreover, Soe et al. (2010) found that co-inoculation with Streptomyces spp. P4 and indigenous Myamar Bradyrhizobium was more effective for N uptake than single inoculation of Bradyrhizobium. In the present study, co-inoculation of SAY3-7 with P4 significantly improved P, K, Ca, and Mg uptakes at R3.5 and R5.5 stages (Table 3). This is the first report that P, K, Ca, and Mg uptakes were affected by co-inoculation of Myanmar Bradyrhizobium strain and P4.

Our group has reported significant increases in soybean yield after inoculation of bradyrhizobia (Soe et al. 2010; Yamakawa and Fukushima 2014) and co-inoculation of bradyrhizobia with P4 (Soe et al. 2012; Soe and Yamakawa 2013a, 2013b). Significant increase in yield was observed due to inoculation (Table 4). This results support the findings of our group (Soe et al. 2012; Soe and Yamakawa 2013a; Soe and Yamakawa 2013b) in which there is synergetic effect in increasing soybean productivity due to co-inoculation of P4 with Myanmar Bradyrhizobium.

In our previous study, we observed the effects of co-inoculation of the same combination of SAY3-7 and P4 bacteria on plant growth, nodulation, and nitrogen fixation in Yezin-6 in the controlled pot experiment where these bacteria were inoculated on the sterilized seeds and vermiculite (Htwe and Yamakawa 2016). Compared to the previous studies, the new findings in this study were the effects of co-inoculation SAY3-7 and P4 bacteria on plant growth, nodulation, nitrogen fixation, nutrient uptake, and yield in a field condition where indigenous bacteria were already established in the soil.

5. Conclusion

The results show that co-inoculation of SAY3-7 and P4 significantly improved plant growth; nodulation; nitrogen fixation; N, P, K, Ca, and Mg uptake; and yield of Yezin-6 compared with those in the control group. Therefore, we conclude that the combined use of SAY3-7 and P4 will be helpful for soybean production by enhancing plant growth, nodulation, N fixation, and major nutrient uptake.

Acknowledgment

This work was supported by Ministry of Education, Culture, Sports, Sciences and Technology (MEXT) of Japan.

Additional information

Funding

This work was supported by the Ministry of Education, Culture, Sports, Science, and Technology Grant Number: [130287].