Enhancing Soybean Yield Through Inoculation of Multifunctional Microbial Consortia

Abstract

Inoculating multifunctional microbial consortia offers potential benefits for enhancing plant growth and grain yield formation. This study verified the feasibility of inoculating soybean (Glycine max) seeds with multifunctional microbial consortia to improve soybean productivity. Seeds were inoculated with 12 combinations of inoculants [Bradyrhizobium japonicum and B. diazoefficiens for biological N₂ fixation, Azospirillum brasilense for growth promotion via phytohormone release, Bacillus megaterium (=Priestia megaterium) and B. subtilis for enhancing P uptake, and Trichoderma harzianum as biopesticide] and grown in BOD chamber, greenhouse, and field experiments. In the chamber, inoculated seeds were subjected to germination tests. In the greenhouse, inoculated seeds were sown in pots with nonsterile soil, and plant growth was monitored until the flowering stage. In the field, plants were cultivated until physiological maturity. Soil and plant samples were collected at three growth stages: vegetative, reproductive, and maturation. Measurements included shoot, root, nodules, grain masses, and grain yield, alongside analyses of seed N, P, lipid, protein, and carbohydrate contents. The highest number of microbial inputs and the inclusion of T. harzianum in the microbial consortia impeded seed germination, hindered initial vegetative growth, and decreased plant densities in the plots. Likely due to the crop’s plasticity and the stimulation of microorganisms, the initial setbacks did not affect grain yield and soybean grain lipid content. Therefore, inoculating multifunctional microbial consortia holds promise as a sustainable approach in agriculture. Still, care should be taken concerning an excessive number of inoculants composing the consortia.

This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.

Summary

• Bradyrhizobium, Azospirillum, and Bacillus are compatible microorganisms to be inoculated in soybean.

• The soybean plasticity compensates for initial decay in growth due to excessive inoculants or Trichoderma.

• Inoculation of multifunctional microbial consortia might shift host photosynthate metabolism.

1. Introduction

Microbial inoculation of crops with elite strains is an interesting approach to advancing sustainable agriculture. By promoting plant growth, enhancing plant nutrition, and boosting grain yields, this technique promises to reduce reliance on synthetic fertilizers, curtailing production costs and mitigating environmental pollution [1]. Moreover, the augmentation of root growth and soil biological activity through microbial inoculation can elevate soil organic matter, a vital component for soil physical structure and nutrient exchange [2]. Microbial inoculation with elite strains also confers increased crop tolerance and resilience to environmental adversities such as drought, high temperatures, pest attacks, and diseases. Consequently, it could facilitate agricultural acclimation to the evolving challenges of climate change [3].

There are microbial inputs recommended for most commodity crops. In certain countries, such as those of the European Union [4], Brazil [1], and China [5], the production and commercialization of microbial inputs are government-regulated. Microbial inputs are designed to be employed exclusively for registered crops and designated scenarios in adherence with regulatory agencies. Microbial inputs may fit into several agronomic functions. For instance, promoting nodulation in legume crops that enable biological N₂ fixation [e.g., Bradyrhizobium spp. for soybean (Glycine max)], inducing plant growth through rhizospheric phytohormone release (e.g., Azospirillum brasilense), facilitating the solubilization of soil phosphates and improved phosphate uptake (e.g., Bacillus spp.), and bolstering plant resilience against diverse stressors through balanced phytohormones (e.g., B. aryabhattai) [1, 6]. Microbial inputs are also used as biopesticides, for instance, for controlling the growth of phytopathogenic microorganisms (e.g., Trichoderma sp.) and addressing pest infestations (B. thuringiensis). In contexts where the crop is limited by soil phosphate availability, Bacillus megaterium (=Priestia megaterium) and B. subtilis increase the availability of soil phosphate by producing organic acids, protons, and siderophores that alter the pH equilibrium in the rhizosphere and release inorganic phosphate adsorbed to iron and aluminum oxides [7]. Bacillus spp. also produces phosphatases that break down organic phosphates, releasing available forms (HPO₄²⁻ and H₂PO₄⁻) into the soil solution [8].

Traditional recommendations typically advise inoculating specific microbial strains for a particular crop, one at a time. However, experimental data treated through meta-analyses have shown that legumes co-inoculated with two types of microbial inputs, such as rhizobia combined with mycorrhizal fungi [9], Bradyrhizobium spp. with Azospirillum [10, 11], rhizobia with Trichoderma spp. [12], and rhizobia with bacilli [13] exhibit better symbiotic performance and higher grain yields than legumes inoculated only with rhizobia. Notably, the inoculation of efficient Bradyrhizobium strains facilitates biological N₂ fixation, which allows the replacement of synthetic nitrogenous fertilizers, increasing plant growth and yields while decreasing production costs and greenhouse gas emissions [14]. The co-inoculation of Bradyrhizobium with A. brasilense [15, 16] or Bacillus subtilis [17] in soybeans increases root and nodule masses, the leghemoglobin content in nodules, the total shoot N and P contents, and the grain yields in comparison to the single inoculation with Bradyrhizobium spp.

Moreover, it has been shown that the multiple inoculation of Bradyrhizobium, Azospirillum, and Bacillus with microbial metabolites increased root, shoot, nodule, and grain masses, under field conditions, in comparison to the single inoculation with Bradyrhizobium [18]. The approach of inoculating multifunctional microbial consortia was also successfully applied to nonlegume crops such as sugar cane (Saccharum officinarum) [19] and maize (Zea mays) [20], producing positive results for crop yields. Furthermore, a meta-analysis confirmed that adopting microbial consortia leads to more considerable plant growth and yield increments than single inoculant applications [21].

However, these consortia studies also emphasized the need for engineering multifunctional consortium arrangements to achieve even more compelling results. In fact, microorganisms may compete for similar resources (e.g., rhizospheric carbon compounds and mineral nutrients) and may interact in diverse ways, either stimulating or disrupting each other’s colonies [21, 22], which may indirectly affect plant physiology.

In this study, we evaluated the effects of inoculating multifunctional microbial consortia using inoculants recommended for soybeans. With a production of 154.6 million metric tons in 2023 [22], soybean is one of the three most important grain crops produced worldwide, with the largest production occurring in Brazil, followed by the United States of America, Argentina, and China [23]. This crop is highly valued due to its high lipid and protein contents [9], which are only possible because of its high photosynthetic capacity [9, 24, 25], supported by its symbiotic association with efficient diazotrophic Bradyrhizobium strains able to supply the N demanded by the crop [26, 27]. The crop also benefits from co-inoculation of Bradyrhizobium with arbuscular mycorrhizal fungi and associative plant growth-promoting rhizobacteria (e.g., [9, 13]; Zeffa et al., 2020; [12]), making it a model plant for studying the responses to microbial consortia.

We hypothesized that the inoculation of multifunctional microbial consortia will positively influence plant growth and grain yield formation processes for soybean. We aim to test the feasibility of inoculating multifunctional microbial inoculants to maximize crop productivity.

2. Materials and Methods

2.1. Microbial Treatments

The treatments were designed prioritizing the inoculant containing Bradyrhizobium japonicum and B. diazoefficiens, and from there, introducing co-inoculation of multifunctional microbial inoculants, resulting in a total of 12 treatments for the various conducted experiments. The microorganisms used and the dosages employed are presented in Table 1. All inoculants were subjected to a qualitative assessment for hydrogen cyanide (HCN) production following the method of Bakker and Schippers [29]. Among them, only Trichoderma sp. demonstrated the ability to produce HCN.

Table 1

Treatments of co-inoculations with plant growth-promoting microorganisms in soybean subjected to field, greenhouse, and BOD chamber experiments in Campo Largo and Curitiba, PR, Brazil.

Microbial input	Bj	Ab	Bs	Th
Function	Nodulation and biological N2 fixation	Phytohormone release	Soil P solubilization	Biopesticide
Microorganisms	Bradyrhizobium japonicum SEMIA 5079 = CPAC 15 + B. diazoefficiens SEMIA 5080 = CPAC 7	Azospirillum brasilense Ab-V5 + A. brasilense Ab-V6	Bacillus subtilis CNPMS B2084 + B. megaterium (=Priestia megaterium) CNPMS B119	Trichoderma harzaianum Simbi-T5
Concentration	7.2 × 109 viable cells mL−1	4 × 108 viable cells mL−1	4 × 109 viable cells mL−1	1 × 109 CFU (colony-forming units) mL−1
Recommended dose	1 mL·kg−1	2 mL·kg−1	1.4 mL·kg−1	2 mL·kg−1
Treatment	Inoculation: mL in 450 g of seeds
T1 = no inoculation	—	—	—	—
T2 = Bj	0.32	—	—	—
T3 = Ab	—	0.9	—	—
T4 = Bs	—	—	0.65	—
T5 = Th	—	—	—	0.9
T6 = Bj + Ab	0.32	0.9	—	—
T7 = Bj + Bs	0.32	—	0.65	—
T8 = Bj + Th	0.32	—	—	0.9
T9 = Bj + Ab + Bs	0.32	0.9	0.65	—
T10 = Bj + Ab + Th	0.32	0.9	—	0.9
T11 = Bj + Bs + Th	0.32	—	0.65	0.9
T12 = Bj + Ab + Bs + Th	0.32	0.9	0.65	0.9

Note: We used liquid commercial inoculants that are approved for use in Brazilian territory and are formally monitored by Brazilian authorities, as specified in Law No. 6894 [28]. Dosages were defined according to the manufacturer’s instructions. To the inoculation mixture, 0.5 mL of a maltodextrin-based sealant and thickener was added to provide better adhesion of the inoculants to the seeds. Bj: inoculant with Bradyrhizobium japonicum SEMIA 5079 = CPAC 15 + B. diazoefficiens SEMIA 5080 = CPAC 7, Ab = with Azospirillum brasilense Ab-V5 + A. brasilense Ab-V6, Bs = with Bacillus subtilis CNPMS B2084 + B. megaterium (=Priestia megaterium) CNPMS B119, and, Th = with Trichoderma harzianum Simbi-T5.

2.2. Germination Test

Samples of 100 seeds of the indeterminate-cycle soybean cultivar AS 3590IPRO were inoculated according to Table 1 and incubated in a Biochemical Oxygen Demand (BOD)-type chamber from March 3 to 9, 2022. This experiment was carried out using a completely randomized design, following the procedures for seed analysis as described by ISTA [30].

2.3. Greenhouse Experiment

The experiment was conducted from March 15, 2022, to June 06, 2022, in an open greenhouse at coordinates 25°46′23″S and 49°10′13″W, Curitiba, Paraná State (PR), Brazil. The soil used in the experiment was classified as Typic Dystrohumic Cambisol soil [31], sampled at the coordinates 25°23′30″S and 49°07′30″W, Pinhais, PR, Brazil. The indeterminate-cycle soybean cultivar AS 3590IPRO was used. The results of chemical analyses indicated the need for liming and fertilization. Subsequently, the soil was passed through a 4 mm sieve and amended with 5 g·dm⁻³ of high-reactivity lime with an effective calcium carbonate equivalence (ECCE) of 90% to achieve a base saturation of 60%. Following this, the soil was fertilized with 200 mg·P·dm⁻³ and 150 mg·K·dm⁻³, equivalent to 1.5 g·dm⁻³ of Triple Superphosphate (monocalcium phosphate) and 0.5 g·dm⁻³ of Potassium Chloride, respectively.

The experiment was performed under a completely randomized design with 12 replicates and two sampling dates. Each experimental unit consisted of a seven-liter pot in which 6 soybean seeds were sown. At the end of the germination period, on April 4, 2022, thinning was performed, retaining the most vigorous plant per pot or the only emerged plant in the pot. Plants were grown in a greenhouse with side openings and natural ventilation. The average daily temperature was 18.4°C for the plants collected at the V4 stage and 16.8°C at the R1 stage. Soil pots were routinely wetted to maintain 70% of the soil field humidity capacity.

The sampling of plants was conducted at the V4 stage (four true expanded leaves) on May 6, 2022, and at the R1 stage (beginning of flowering) on June 6, 2022. Soil samples were taken during the same stages to check soil fertility changes due to liming and fertilization (Table 1). At the V4 and R1 stages, the plants underwent analyses for root length, shoot height, dry root and shoot mass, number and dry mass of nodules, and N and P content in the shoot.

Emerged seedlings in the greenhouse experiment were counted to assess seed germination in pots at 7, 10, 15, 18, and 21 days after planting. The data collected regarding the number of germinated seeds were converted into percentages for statistical analysis.

2.4. Field Experiment

The experiment was conducted in the 2021/2022 summer cropping season, in the municipality of Campo Largo, Paraná State, at coordinates 25°18′49.0″S 49°33′49.7″W, in a Typic Dystrohumic Cambisol soil [31], at an altitude of 934 m. According to the classification by Alvares et al. [32]; the climate in the area is a temperate oceanic climate (Cfb type). Climatic conditions during the experiment are shown in Figure 1. The average daily temperature and cumulative precipitation were 20.8°C and 314 mm, respectively, between sowing and the beginning of flowering (R1 stage; [33]), but with a short drought during seedling emergency (Figure 1). Between the R1 stage and harvest, the temperature averaged 20.5°C, and the cumulative precipitation was 345 mm, summing 669 mm for the whole cycle.

[figure(s) omitted; refer to PDF]

The area has been cultivated with annual crops under no-till management for over 8 years, and in the past 2 years, has been cropped first with soybean (summer), followed by onion (Allium cepa) (winter), and then, by maize (summer). Subsequently, avena (Avena strigosa) was cultivated as a cover crop and desiccated with glyphosate before soybean sowing. The soybean and maize crops were respectively inoculated with commercial inoculants containing B. japonicum strain SEMIA 5079 and B. diazoefficiens strain SEMIA 5080, and A. brasilense strains Ab-V5 + Ab-V6. Fifteen composite soil samples were taken in an transect immediately after the harvest of maize (April 14, 2021) from the 0–20 cm depth layer, for soil chemical and physical characterization (Table 2). The average results from chemical, granulometric, and total content analyses indicated no need for fertilizer and liming application.

Table 2

Soil analyses for multifunctional microbial inoculation experiments in soybean grown in Curitiba (greenhouse) and Campo Largo (field) during the growing season of 2021/2022.

Parameter	Unity	Greenhouse			Field
		Native soil	V4	R1	Fallow	Harvest
pH CaCl2		4.33	5.20	5.08	5.70	5.50
pH SMP		4.88	5.62	5.61	6.80	6.60
SOM	g·dm−3	66	74	75	61	69
TOC	g·dm−3	38.4	43.2	43.7	36.0	40.0
Mehlich-1 P	mg·dm−3	6.5	36.0	26.7	112.0	112.3
Total P	mg·kg−1	246.9	489.9	454.9	973	975
Exchangeable Ca	cmolc·dm−3	3.7	13.6	12.3	7.6	7.9
Total Ca	g·kg−1	0.4	1.2	1.5	0.9	0.86
Exchangeable Mg	cmolc·dm−3	1.5	6.6	5.7	1.9	1.62
Total Mg	g·kg−1	1.1	1.6	1.5	0.40	0.38
Exchangeable K	cmolc·dm−3	0.10	0.50	0.40	0.40	0.63
Total K	g·kg−1	2.0	2.5	2.4	0.4	0.56
Na	cmolc·dm−3	0.02	0.06	0.08	0.02	0.01
Fe	g·kg−1	25.1	24.5	25.2	12.7	12.14
Mn	mg·kg−1	122.2	131.4	122.2	144	130.6
Cu	mg·kg−1	16.8	16.3	16.2	11.9	7.8
Exchangeable Al	cmolc·dm−3	2.5	0	0	0	0
H + Al	cmolc·dm−3	13.2	1.9	1.9	2.7	3.16
CEC effective	cmolc·dm−3	7.8	20.7	18.4	9.9	10.1
CEC pH 7	cmolc·dm−3	18.5	22.6	20.3	12.6	13.33
SB	cmolc·dm−3	5.3	20.7	18.4	9.9	10.1
V	%	28	91	90	78	76
Clay	g·kg−1	638	638	675	425	437
Silt	g·kg−1	150	163	175	109	120
Sand	g·kg−1	212	199	150	466	443
Classification	SiBCS	Cambisol			Humic dystrophic typical cambisol
Classification	NRCS	Inceptisol			Inceptisol

Note: pH CaCl₂, pH SMP, exchangeable elements P (Mehlich-1), Na, K, Ca, Mg, and Al, Al + H (potential acidity), CEC pH 7, m = aluminum saturation, and V = base saturation were determined according to Santos et al. [31]; total contents of Ca, Mg, P, Fe, Mn, and Cu were extracted by microwave and determined by inductively coupled plasma optical emission spectrometry; and textural fractions were determined according to Gee and Bauder [34]. Field means were obtained from three composite samples during fallow and five composite samples (one per block) at harvest; greenhouse means were obtained from one composite sample of the native soil and one composite sample of all pots at growth stages V4 and R1. Soil classification in SiBCS according to Brazilian Soil Classification System [31] and in NRCS according to Keys to Soil Taxonomy [35].

Abbreviations: CEC = cation exchange capacity, SB = sum of bases, SOM = soil organic matter content, TOC = total organic carbon.

Before sowing, furrows were opened in rows 0.45 m apart using a sowing machine (Vence Tudo Panther SM 6000) pulled by a tractor (New Holland TL75E) on November 16, 2021. The indeterminate-cycle soybean cultivar AS 3590IPRO was sown from November 23 to 26, 2021, in previously delimited plots, along with the different treatments. Seeds were mixed with pipetted aliquots of microbial inputs as specified in Table 1 in polystyrene plastic bags puffed up with its own air to mimic balloons, stirred gently, and applied in-furrow with a jab sowing machine. The doses of each microbial input were applied, as shown in Table 1. Sowing started with the noninoculated treatment, followed by the treatment single inoculated with Bradyrhizobium, and then the others. The jab sowing was disinfested with 70% alcohol before each treatment change. Sowing considered a proportional number of seeds in the plots of 350,000 seeds per hectare.

The experiment was conducted in a randomized block design with five replications. The experimental units were plots measuring 5.5 m in length and 2.25 m in width (5 planting rows with 0.45 m between rows), spaced 1 m apart between actions. The useful area consisted of 5.5 m and 3 central rows, with the removal of the border from each of the rows parallel to the useful area.

During the R1 development stage on February 1, 2022, five plants per plot were randomly harvested to determine the length with a caliper and dry weight of the aboveground and root parts, the number and dry weight of nodules, and the total N and P contents of the aboveground parts.

Phytosanitary treatments were applied when necessary. However, significant disease problems were not identified during the experiment. A preventive application of protioconazole (triazolinthione) + trifloxystrobin (strobilurin) and acetamiprid (neonicotinoid) + bifenthrin (pyrethroid), along with orange oil as a sealer, was performed on January 5, 2022.

During the physiological maturity stage from April 18 to April 27, 2022, the available area of the plot was harvested to determine grain yield, 1000-grain mass, and protein, lipid, and carbohydrate contents in the grains. On the same occasion, soil samples were collected from the 0–10 cm layer between the rows of the plots to assess the overall fertility content of the area, with an average of the five blocks (Table 2). The total number of plants per available area in the plots was counted.

2.5. Laboratory Analyses

Harvested plant parts were dried under ventilation at 60°C per 72 h, shoots and roots were measured with a caliper, nodules were counted, and all parts were weighed. Dried shoots were ground into 2 mm fragments. Then, total N and P in aboveground mass were determined using Kjeldahl and colorimetry [7].

Grain protein and lipid contents were determined, respectively, by the methods of Kjeldahl [36] and Soxhlet extraction [37], while the carbohydrate contents were estimated by difference. The 1000-grain mass was obtained by counting and weighing grains.

Soil chemical analyses were performed according to Santos et al. [31] and textural characterization with Bouyoucos densimeter methodology [34].

2.6. Statistical Analyses

Statistical analyses were conducted according to the experimental designs. The data were subjected to the Shapiro–Wilk test to confirm the assumptions of analysis of variance (ANOVA) using the ExpDes.pt package [38] in the open-access R software [39]. When the data were not normally distributed, they were transformed using the square root. Means were compared using the Tukey test at a 5% significance level.

3. Results

3.1. Seed Germination and Emergency

In the BOD-chamber, germination was estimated on the sixth day. The seeds germinated at different times in the greenhouse and in the field. Therefore, the count of emerged plants in the field was performed at the end of the vegetative stage. In the greenhouse, emerging seedlings were counted at 3- or 5-day intervals, up to 21 days when the germination stage ended. Microbial consortia consistently affected soybean seed germination under the three different environmental conditions (field, greenhouse, and BOD). The inoculation of Bradyrhizobium only ensured the highest germination rate in all three evaluated conditions, suggesting a beneficial effect, especially under field conditions (Table 3). The treatments T4 = Bs, T5 = Th, T6 = Bj + Ab, T11 = Bj + Bs + Th, and T12 = Bj + Ab + Bs + Th reduced seed germination by half or less than half than T2 = Bj under BOD. A similar result was found in the greenhouse, where eight treatments had lower seed germination than T2 = Bj and noninoculated control. However, consortia inoculations involving Trichoderma (T10 = Bj + Ab + Th and T12 = Bj + Ab + Bs + Th) decreased the number of emerged plants compared to T2 = Bj, while those having Bacillus did not impair germination (Table 3).

Table 3

Percentage of soybean germination in B.O.D. and seedling emergence in greenhouse, and plant per plot under field condition in 12 multifunctional microbial inoculation experiments.

Experiment	B.O.D.% emergence	Greenhouse% of emerged seedlings					FieldPlants per plot (% of emerged seedlings)
Days	7	7	10	15	18	21	150
T1	78.0ab	44.4a	55.6a	66.5a	69.2a	70.7a	95.8ab	(32)
T2	86.0a	44.5a	54.2a	68.0a	70.8a	70.8a	134.4a	(38)
T3	68.0bc	32.0ab	45.8ab	59.7ab	62.5ab	65.3ab	93.6b	(27)
T4	40.5d	14.1bc	25.1bc	36.0bc	40.2bc	44.3bc	91.4b	(26)
T5	38.5d	9.8c	19.5c	26.5c	27.9c	29.3c	103.6ab	(30)
T6	48.0d	7.1c	12.7c	27.8c	27.7c	27.8c	93.2b	(27)
T7	78.0ab	7.0c	11.2c	18.0c	18.0c	19.4c	109.4ab	(31)
T8	74.5abc	5.7c	9.8c	18.1c	18.1c	22.3c	102.0ab	(29)
T9	66.5bc	9.8c	15.3c	27.8c	29.2c	32.0c	104.6ab	(30)
T10	65.5c	5.7c	11.3c	18.0c	18.0c	19.4c	81.8b	(23)
T11	45.5d	2.8c	9.8c	19.5c	20.8c	22.3c	98.2ab	(28)
T12	43.5d	7.0c	16.8c	23.7c	23.7c	23.8c	85.6b	(24)
CV (%)	7.8	90.8	71.3	57.3	53.14	50.6	18.7

Note: Treatments combine microbial inputs as specified in Table 1: T1 = no inoculation; T2 = Bj; T3 = Ab; T4 = Bs; T5 = Th; T6 = Bj + Ab; T7 = Bj + Bs; T8 = Bj + Th; T9 = Bj + Ab + Bs; T10 = Bj + Ab + Th; T11 = Bj + Bs + Th; T12 = Bj + Ab + Bs + Th. Bj: inoculant with Bradyrhizobium japonicum SEMIA 5079 = CPAC 15 + B. diazoefficiens SEMIA 5080 = CPAC 7; Ab = with Azospirillum brasilense Ab-V5 + A. brasilense Ab-V6; Bs = with Bacillus subtilis CNPMS B2084 + B. megaterium (=Priestia megaterium) CNPMS B119; and, Th = with Trichoderma harzianum Simbi-T5. The field experiment was conducted with five repetitions, the greenhouse experiment with six replicates, and the B.O.D. experiment with four repetitions. Field plots had an area of 7.4 m² each. Between brackets, the % of germinated seeds in relation to what would be expected considering that 100% of the seeds germinated. Means followed by the same letter do not differ statistically (Tukey’s test at 5% probability). Means followed by the same letter in the column do not differ statistically (Tukey’s test at 5% probability).

3.2. Greenhouse Experiment

The different consortia did not affect plant height, shoot dry, or root dry mass in the greenhouse. Root length at the reproductive stage (R1) was stimulated by co-inoculation with Bradyrhizobium and Azospirillum (T6 = Bj + Ab). The opposite was obtained with the inoculation of Bacillus and the inoculation of the consortium of Bradyrhizobium, Bacillus, and Trichoderma, resulting in a difference between T6 = Bj + Ab versus T4 = Bs and T11 = Bj + Bs + Th (Table 4).

Table 4

Plant growth parameters of soybean submitted to multifunctional microbial inoculation in the greenhouse.

Treatment	Plant height (cm)		Root length (cm)		Dry shoot mass (g)		Dry root mass (g)		Dry nodule mass (mg)		Nodule number
Stage	V4	R1	V4	R1	V4	R1	V4	R1	V4	R1	V4	R1
T1	22.5	28.5	34.7	34.5ab	3.7	4.1	0.93	1.17	22.0ab	40.3ab	10.5ab	14.0
T2	21.8	26.4	31.7	35.0ab	3.3	5.3	0.82	1.29	24.0a	43.7ab	15.5a	14.7
T3	22.5	34.7	40.5	39.2ab	3.8	7.6	0.88	1.71	23.0ab	179.8a	14.5ab	20.3
T4	19.4	30.3	35.2	31.5b	3.4	5.4	0.82	1.13	11.0ab	71.5ab	7.7ab	13.8
T5	21.7	27.3	38.2	37.0ab	3.3	4.3	0.91	1.21	15.3ab	25.2b	12.8ab	11.2
T6	18.7	32.5	32.3	46.5a	2.7	6.9	0.52	1.73	6.3b	114.4ab	8.0ab	17.8
T7	18.7	26.2	31.0	41.0ab	3.2	5.4	0.73	1.15	10.7ab	43.2ab	10.7ab	7.7
T8	20.3	21.5	33.3	34.5ab	2.9	3.9	0.64	0.88	12.4ab	26.6b	9.3ab	8.2
T9	21.5	27.2	35.8	39.5ab	3.13	5.0	0.72	1.22	17.0ab	75.2ab	15.0a	11.0
T10	21.2	26.3	36.3	41.3ab	3.3	4.2	0.83	1.31	10.4ab	104.2ab	12.2ab	14.3
T11	20.2	22.5	29.8	30.8b	2.7	4.0	0.48	0.89	8.5ab	26.5b	7.8ab	6.8
T12	20	27.5	32.7	41.0ab	3.0	5.1	0.56	1.24	8.3ab	39.8b	6.7b	13.8
CV (%)	18.0	28.6	19.8	18.9	29.3	53.2	35.6	49.3	62.9	114.1	37.6	59.4

Note: Treatments combine microbial inputs as specified in Table 1: T1 = no inoculation; T2 = Bj; T3 = Ab; T4 = Bs; T5 = Th; T6 = Bj + Ab; T7 = Bj + Bs; T8 = Bj + Th; T9 = Bj + Ab + Bs; T10 = Bj + Ab + Th; T11 = Bj + Bs + Th; T12 = Bj + Ab + Bs + Th. Bj: inoculant with Bradyrhizobium japonicum SEMIA 5079 = CPAC 15 + B. diazoefficiens SEMIA 5080 = CPAC 7; Ab = with Azospirillum brasilense Ab-V5 + A. brasilense Ab-V6; Bs = with Bacillus subtilis CNPMS B2084 + B. megaterium (=Priestia megaterium) CNPMS B119; and, Th = with Trichoderma harzianum Simbi-T5. Means followed by different letters are statistically different by Tukey’s test at 5% probability. Means without letters are not significantly different. During ANOVA, the original values of dry root mass and dry nodule mass (V4) and dry root mass, dry nodule mass, and number of nodules (R1) were extracted from their square roots to ensure the required normal distributions, but then the resulting means were returned to their powers. The values represent the average of six repetitions. The V4 stage means that at least 50% of the plants have already formed the fourth trifoil. The R1 stage means that 50% of the plants are at the beginning of flowering. Means followed by different letters in the column are statistically different by Tukey’s test at 5% probability.

The inoculation with only Bradyrhizobium (T2 = Bj) resulted in the highest nodule mass (at both V4 and R1). The co-inoculations of Bradyrhizobium with Azospirillum (T6 = Bj + Ab; V4), Bradyrhizobium with Trichoderma (T8 = Bj + Th; R1) and the multiple inoculations involving Bradyrhizobium, Azospirillum, and Trichoderma (T11 = Bj + Ab + Th, R1), and the inoculation of the four microbial inputs (T12 = Bj + Ab + Bs + Th, V4) decreased nodule growth (Table 4).

The plant nutrition measured by the foliar concentration of N and P was affected by treatment under greenhouse conditions but not in the field conditions. Under greenhouse, the inoculations of consortia Bradyrhizobium, Azospirillum, and Trichoderma (T11 = Bj + Ab + Th) and Bradyrhizobium, Azospirillum, Bacillus, and Trichoderma (T12 = Bj + Ab + Bs + Th) resulted in plants with higher shoot N concentration at the V4 stage than inoculation of only Bradyrhizobium (T2 = Bj) and noninoculated control (Table 5). The concentrations of P in plants at the R1 stage had the highest P concentration (3.7 g·kg⁻¹) for control and the lowest in the treatment inoculated with only Azospirillum (T3 = Ab) (Table 5). In contrast, the treatment with Bradyrhizobium, Bacillus, and Trichoderma (T11 = Bj + Bs + Th) produced plants with higher P accumulation than treatments noninoculated (T1), inoculated with Bradyrhizobium (T2 = Bj), Azospirillum (T3 = Ab), and Bacillus (T4 = Bs) in the greenhouse (Table 5).

Table 5

Shoot nutrients of soybean following the application of multifunctional microbial input in greenhouse and field experiments.

Treatment	Greenhouse								Field
	Concentration (g·kg−1)				Content (mg plant−1)				Concentration (g kg−1)		Content (g plot−1)
	N		P		N		P		N	P	N	P
	V4	R1	V4	R1	V4	R1	V4	R1	R1	R1	R1	R1
T1	36.6b	32.0	3.6	3.7a	140	131	13	15	27.8	3.3	79	9
T2	35.5b	31.2	3.8	2.8bc	120	160	13	13	32.4	3.5	104	11
T3	37.3ab	29.6	3.9	2.4c	141	226	15	18	30.1	3.6	77	8
T4	38.8ab	31.6	3.5	2.7bc	135	163	11	13	32.3	3.5	82	8
T5	39.3ab	30.4	4.1	2.6bc	128	131	15	11	31.9	3.4	90	9
T6	37.3ab	30.1	3.7	2.6bc	103	201	8	18	31.6	3.5	95	10
T7	38.9ab	29.9	4.6	2.6bc	135	170	15	15	32.0	3.7	98	10
T8	40.2ab	30.7	4.1	3.3ab	118	116	13	13	29.3	3.5	91	10
T9	40.9ab	30.2	4.4	2.7bc	128	143	15	11	30.4	3.6	100	11
T10	37.9ab	30.2	3.9	2.6bc	128	120	13	11	31.0	3.4	63	7
T11	42.7a	29.8	5.1	2.4bc	115	116	15	8	31.2	3.5	83	9
T12	42.5a	30.4	4	2.7bc	131	153	15	13	30.6	3.5	73	8
CV (%)	7.6	10.7	21.0	16.2	31.7	51.0	50.1	62.4	10.0	9.0	31.3	26.7

Note: Treatments combine microbial inputs as specified in Table 1: T1 = no inoculation; T2 = Bj; T3 = Ab; T4 = Bs; T5 = Th; T6 = Bj + Ab; T7 = Bj + Bs; T8 = Bj + Th; T9 = Bj + Ab + Bs; T10 = Bj + Ab + Th; T11 = Bj + Bs + Th; T12 = Bj + Ab + Bs + Th. Bj: inoculant with Bradyrhizobium japonicum SEMIA 5079 = CPAC 15 + B. diazoefficiens SEMIA 5080 = CPAC 7; Ab = with Azospirillum brasilense Ab-V5 + A. brasilense Ab-V6; Bs = with Bacillus subtilis CNPMS B2084 + B. megaterium (=Priestia megaterium) CNPMS B119; and, Th = with Trichoderma harzianum Simbi-T5. Means followed by different letters are statistically different by Tukey’s test at 5% probability. Means without letters are not significantly different. The field means represent the average of five replication and the greenhouse represent values of six replications. Stage V4 means that at least 50% of the plants have already formed the fourth trifoil. The R1 stage means that the plants are at the beginning of flowering. Means followed by different letters in the column are statistically different by Tukey’s test at 5% probability.

3.3. Field Experiment

The influence on plant height was observed, but only the T10 = Bj + Ab + Th showed shorter plants than the control (1) and 7 other treatments (Table 6). The 1000-grain mass was also affected by treatment where T11 = Bj + Bs + Th had higher mass than the control and two other treatments, i.e., Bradyrhizobium and Bacillus (T7 = Bj + Bs) and Bradyrhizobium and Trichoderma (T8 = Bj + Th). Despite the difference in thousand grain weight and plant stand (Table 3), the yield was not impacted by treatment (Table 7). Likewise yield, there was no effect of treatments over above and below-ground biomass and root length (Table 6).

Table 6

Plant growth parameters of soybean submitted to multifunctional microbial inoculations under field-growing conditions in Campo Largo, PR, Brazil.

Treatment	Plant height	Root length	Dry shoot mass	Dry root mass	Nodule number	Dry nodule mass
	cm		g plant−1		No. plant−1	mg plant−1
T1	61.9a	14.3	31.8	5.0	16.9ab	108
T2	60.7a	12.3	23.7	3.9	21.2ab	158
T3	57.2ab	14.8	26.4	4.4	22.1ab	148
T4	59.7a	12.8	27.8	4.2	14.5ab	108
T5	60.8a	13.7	28.5	4.5	23.2a	166
T6	62.4a	14.7	31.9	4.7	15.7ab	116
T7	61.4a	14.1	27.4	4.2	20.4ab	142
T8	59.8a	13.9	30.9	4.9	20.3ab	126
T9	61.6a	15.8	31.6	4.8	16.8ab	118
T10	47.8b	12.2	24.4	4.0	13.6ab	84
T11	53.5ab	11.5	27.0	3.8	10.3b	74
T12	54ab	13.3	27.6	4.1	18.7ab	122
CV (%)	9.1	19.3	19.3	19.0	36.9	42.5

Note: Treatments combine microbial inputs as specified in Table 1: T1 = no inoculation; T2 = Bj; T3 = Ab; T4 = Bs; T5 = Th; T6 = Bj + Ab; T7 = Bj + Bs; T8 = Bj + Th; T9 = Bj + Ab + Bs; T10 = Bj + Ab + Th; T11 = Bj + Bs + Th; T12 = Bj + Ab + Bs + Th. Bj: inoculant with Bradyrhizobium japonicum SEMIA 5079 = CPAC 15 + B. diazoefficiens SEMIA 5080 = CPAC 7; Ab = with Azospirillum brasilense Ab-V5 + A. brasilense Ab-V6; Bs = with Bacillus subtilis CNPMS B2084 + B. megaterium (=Priestia megaterium) CNPMS B119; and, Th = with Trichoderma harzianum Simbi-T5. Means followed by different letters are statistically different by Tukey’s test at 5% probability. Means without letters are not significantly different. During ANOVA, the original values of root length, dry shoot mass, and number of nodules was extracted from their square roots to ensure the required normal distributions, but then the resulting means were returned to their powers. The values represent the average of five replications, each repetition being composed of the average of five plants. The plants were harvested at the R1 stage (beginning of flowering). Means followed by different letters in the column are statistically different by Tukey’s test at 5% probability.

Table 7

Grain yield, 1000-grain mass, and grain composition of soybean submitted to multifunctional microbial inoculation under field conditions in Campo Largo, PR, Brazil.

Treatment	Grain yield	1000 grains	Grain composition
			Lipid	Protein	Carbohydrate
	kg·ha−1	g	%
T1	3920	161.7b	12.6a	27.0	40.7
T2	3475	167.6ab	11.1ab	28.6	40.2
T3	3745	163.6ab	10.6ab	29.2	40.9
T4	3691	166.9ab	11.5ab	30.1	39.1
T5	3273	165.1ab	11.6ab	29.5	39.4
T6	3085	163.6ab	10.7ab	28.3	42.0
T7	3812	161.8b	10.1ab	29.1	41.0
T8	4310	159.9b	8.9ab	30.4	41.7
T9	3785	165.4ab	8.8ab	29.4	42.2
T10	2748	167.5ab	8.8ab	30.5	41.2
T11	4243	175.1a	8.3b	30.3	42.5
T12	4108	164.9ab	9.2ab	29.4	42.4
CV (%)	26.5	3.7	17.2	6.5	7.4

Note: Treatments combine microbial inputs as specified in Table 1: T1 = no inoculation; T2 = Bj; T3 = Ab; T4 = Bs; T5 = Th; T6 = Bj + Ab; T7 = Bj + Bs; T8 = Bj + Th; T9 = Bj + Ab + Bs; T10 = Bj + Ab + Th; T11 = Bj + Bs + Th; T12 = Bj + Ab + Bs + Th. Bj: inoculant with Bradyrhizobium japonicum SEMIA 5079 = CPAC 15 + B. diazoefficiens SEMIA 5080 = CPAC 7; Ab = with Azospirillum brasilense Ab-V5 + A. brasilense Ab-V6; Bs = with Bacillus subtilis CNPMS B2084 + B. megaterium (=Priestia megaterium) CNPMS B119; and, Th = with Trichoderma harzianum Simbi-T5. Means followed by different letters are statistically different by Tukey’s test at 5% probability. Means without letters are not significantly different. During ANOVA, the original values of proteins and carbohydrates were extracted from their square roots to guarantee the required normal distributions, but then the resulting averages were returned to their powers. The values represent the average of five replications. Grain yield was measured in the plots and extrapolated to hectare. Means followed by different letters in the column are statistically different by Tukey’s test at 5% probability.

The most significant number of nodules in the field were found in plants of T5 = Th, whereas the lowest number was found in T11 = Bj + Bs + Th; other treatments did not affect nodule numbers (Table 6). The treatments did not affect nodule mass (Table 6). Similarly, in the field, the treatments did not affect the concentrations and contents of total N and P in shoots (Table 6).

Regarding grain composition, the concentrations of protein and carbohydrates remained unaffected by the various consortium inoculations (Table 7). The highest lipid concentration was observed in the grains from the noninoculated plots, where the grains were smaller. Conversely, the lowest lipid concentrations were found in the T11 = Bj + Bs + Th plots, where the grains were larger (Table 7).

4. Discussion

The co-inoculation of Bradyrhizobium with other plant growth-promoting microorganisms (arbuscular mycorrhizal fungi, Bacillus, Azospirillum, and Trichoderma) has been increasingly used in agriculture [9, 13]; Zeffa et al., 2020; [11, 12]. Co-inoculation relies on the assumption that multifunctional microbial consortia can potentially enhance crops through complementary mechanisms of growth promotion [20, 21, 40, 41]. In this study, soybean plants were subjected to inoculation by individual strains or consortia of four microbial inoculants containing Bradyrhizobium, Azospirillum, Bacillus, and Trichoderma, with the hypothesis that the complementarity effects generated by multifunctional microbial inputs lead to higher productivity that the inoculation of single strains, putatively compatible between each other. Overall, our results indicated that inoculating multifunctional microbial consortia may be viable for maximizing crop yields. Still, certain aspects associated with consortia construction need to be carefully considered.

One significant aspect is that inoculation of specific multifunctional microbial consortia resulted in a low number of germinated seeds and negatively impacted plant emergence in both the greenhouse and the field, leading to lower plant density (Table 3). Seed germination is influenced by several factors, including seed quality and environmental conditions [42–44]. Additionally, it can be affected by inputs applied, including inoculants, during sowing [45]. In this study, the reduced germination rates in plants may have been attributed to the larger microbial cell density in the seeds. Specifically, despite applying the recommended dose for each inoculant, we combined multiple inoculants per seed, reflecting common farming practices of mixing microbial inputs. The increased microbial density may have disrupted the rhizosphere environment by introducing excessive biochemical compounds produced during the establishment of symbiotic or commensal associations [46]. Furthermore, it is plausible that the inoculated microorganisms engaged in antibiosis or competition during their growth in the rhizosphere [41, 46], which, in turn, affected seed germination and seedling development. As a result, plots treated with a higher number of inoculants exhibited a lower plant density per unit area. Typically, low plant density in the field is associated with reduced agricultural productivity across various crops [47]. However, in this study, crop grain yields were not adversely affected, likely because the inoculants stimulated photosynthesis and biological N₂ fixation on a per-plant basis, thereby compensating for the yield loss [43].

During the early stages of crop development (V4), the inoculation of specific microbial consortia had a negative impact on plant growth. For instance, the inoculation of Bradyrhizobium, Azospirillum, and Trichoderma (T11 = Bj + Ab + Th) resulted in shorter roots and shoots compared to the co-inoculation of Bradyrhizobium and Azospirillum (T6 = Bj + Ab) (Tables 4 and 5). Remarkably, these results support the effectiveness of co-inoculation of Bradyrhizobium and Azospirillum in maximizing plant growth and yield [11]. However, these findings contradict our initial hypothesis that including a more diverse consortium of microorganisms would lead to higher plant growth. The consortium of microbial inputs having Trichoderma decreased soybean shoot biomass, consistent with findings reported elsewhere [48]. Therefore, we propose that the responses are not solely explained by the number of microbial inputs but also by the identity of the microorganisms within the consortia. In our experiment, this could potentially be attributed to the antibiosis effect of the Trichoderma strain.

Under greenhouse conditions, the inoculation of Trichoderma alone or in combination with Bacillus (T5 = Th, T8 = Bj + Th, T11 = Bj + Bs + Th, and T12 = Bj + Ab + Bs + Th) led to a significant reduction in nodule mass during the R1 stage, in comparison to the treatment single inoculated with Azospirillum (T3 = Ab) (Table 5). The reasons behind this inhibitory effect are unclear. It could be attributed to properties such as the production of HCN, a highly toxic volatile compound [49, 50], which is produced by Trichoderma strain (data not shown), or to the presence of another antibiotic molecule.

Nonetheless, any adverse effects attributed to Trichoderma may have exhibited transient characteristics throughout crop development. Plants subjected to the inoculation of multifunctional microbial consortia (T11 = Bj + Bs + Th and T12 = Bj + Ab + Bs + Th, both including Trichoderma) had higher shoot N concentration compared to those noninoculated or single inoculated with Bradyrhizobium (Table 6), even though their nodule masses were comparatively lower (Table 5). The variations in nodulation patterns imply that Trichoderma could regulate the allocation of plant photosynthates, thereby influencing nodulation negatively. However, its impact on various activities, such as phytohormone production [51], could potentially have stimulated bacteroid activity later in crop development. This stimulation, in turn, may have facilitated increased biological N₂ fixation, irrespective of the initial plant size.

Bacillus strains were incorporated into the experiment due to their role in solubilizing phosphate in the rhizosphere soil, enhancing plant P nutrition [7, 52]. Interestingly, the analysis of P content in plants grown in the greenhouse and inoculated solely with Bacillus did not reveal significant differences compared to their noninoculated counterparts (Table 5). However, at the R1 stage in the greenhouse, plants inoculated with microbial consortia containing Bradyrhizobium, Bacillus, and Trichoderma exhibited higher P contents than those inoculated solely with Bacillus (T11 = Bj + Bs + Th; Table 6) despite having shorter root length (Table 5). This suggests that plants inoculated with microbial consortia displayed greater efficiency in absorbing soil P.

In essence, for both N and P, greenhouse-grown plants inoculated with multifunctional microbial consortia demonstrated a superior nutritional status than plants inoculated with fewer microbial strains. The absence of differences in nutritional status observed in the field could potentially be attributed to its favorable soil fertility levels (Table 2), wherein nutrient absorption is not a limiting factor.

Inoculations of multifunctional microbial consortia may have altered the physiological functioning of soybean plants. First, the inoculation of consortia negatively affected the vegetative growth of the crop in the greenhouse (reduced root length and mass, and number of nodules; Table 4) and in the field (reduced height and number of nodules; Table 5). Later, these presumed negative results disappeared, as the soybean cultivar (indeterminate cycle) exhibited high phenotypic plasticity to compensate for initial adversities and recovered its productive capacity to satisfactory agronomic levels. Single inoculation with Bradyrhizobium resulted in the highest plant growth in both experiments (Tables 4 and 5), and despite showing differences in the number of plants per plot (Table 3), no statistical differences were verified in grain yield (Table 7). Conversely, the inoculation with multifunctional microbial consortia containing Bradyrhizobium, Azospirillum, Bacillus, and Trichoderma (T12 = Bj + Ab + Bs + Th) produced smaller plants in the vegetative stage but later resulted in higher grain yield. Similarly, Moretti et al. [18] demonstrated that inoculation of microbial consortia (i.e., Bradyrhizobium spp., Rhizobium spp., B. subtilis, and A. brasilense) increased grain yield by 485 kg·ha⁻¹ in comparison to the single inoculation with Bradyrhizobium. Given this, we hypothesize that soybean inoculation with multifunctional microbial consortia consisting of Bradyrhizobium and another plant growth-promoting microorganism can promote yield gain despite the reduced initial growth due to the crop’s phenotypic plasticity [25, 43, 44].

The grain composition analyses showed that multiple inoculations altered the redistribution of photosynthates during the reproductive period of the crop. In the case of T11 = Bj + Bs + Th (responsible for the highest grain mass), it produced grains with lower lipid values compared to the noninoculated T1 (Table 7), indicating a dilution effect on the composition. Moreover, it is possible that the high number of microorganisms decreased grain lipid concentration due to the degradation of plant photosynthates by rhizosphere respiration [53]. However, Kaschuk et al. [9, 25] demonstrated that photosynthetic rates can be compensated by the activity of rhizobia and mycorrhizal fungi in the rhizosphere, resulting in no differences in grain lipid concentration due to Bradyrhizobium inoculation. Furthermore, Marra et al. [54] showed that soybean inoculation exclusively with Trichoderma increased lipid concentration in the grains. In any case, the results highlight that inoculations of multifunctional microbial consortia strongly depend on the combination of the strains used when designing the consortia.

In summary, despite the potentially favorable perspective provided by the results of this study regarding the recommendation of multifunctional microbial consortia inoculations containing different plant growth-promoting microorganisms in soybean cultivation, several aspects remain to be comprehensively understood. There is a significant research interest in exploring various microbial consortia to ensure more sustainable agriculture with minimal environmental impact through microorganism utilization [21, 40]. Another beneficial aspect of using consortia lies in the potentially higher stability and survival of consortia rather than single strains when introduced in the soils. Results highlight the need to carefully design microbial consortia according to the intended aim.

5. Conclusion

Prudence is advised in the inoculation of multifunctional microbial consortia in soybeans. Excessive microbial inputs or including T. harzianum in the microbial consortia may adversely impact seed germination and hinder vegetative growth. Nonetheless, the increase in crop productivity despite varying plant densities suggests that inoculating multifunctional microbial consortium is a promising approach to sustainable agriculture.

Disclosure

An early version of this article has been made available as Rossetim et al. [55] at https://doi.org/10.21203/rs.3.rs-3304353/v1.

Author Contributions

Murilo Francisco Travençoli Rossetim: conceptualization; data curation; formal analysis; investigation; methodology; roles/writing–original draft; and writing–review and editing.

Antonio Carlos Vargas Motta: conceptualization; formal analysis; supervision; visualization; and writing–review and editing.

Yanka Rocha Kondo: formal analysis; methodology; visualization; and writing–review and editing.

Barbara Elis Santos Ruthes: formal analysis; methodology; visualization; and writing–review and editing.

Mariangela Hungria: funding acquisition; visualization; and writing–review and editing.

Joana Falcão Salles: investigation; visualization; and writing–review and editing.

Glaciela Kaschuk: conceptualization; funding acquisition; investigation; project administration; supervision; validation; visualization; roles/writing–original draft; and writing–review and editing.

Funding

Murilo Francisco Travençoli Rossetim acknowledges the Coordination for the Improvement of Higher Education Personnel (CAPES) for the scholarship granted throughout the study period. The research was conducted with financial support from the Federal University of Paraná-Graduate Support Program (PROAP/UFPR), and the INCT-Plant-Growth Promoting Microorganisms for Agricultural Sustainability and Environmental Responsibility (CNPq 465133/2014-4, Fundação Araucária-STI043/2019).

Acknowledgments

The authors thank the farmers Marcio Domingues Mikos and Mauricio Domingues Mikos for providing the land and offering labor support during the field experiment and the laboratory technicians from the Federal University of Paraná, Heila Silva de Araújo, Carla Gomes Albuquerque, Fabiana Gavelaki, Josianne Meyer, Maria Aparecida Carvalho Santos, Roseli do Rocio Beggiora, and Jair José de Lima, for supporting the analyses. ChatGPT was used to ensure the accuracy of English grammar. Following the use of this tool, the authors reviewed and edited the content as needed, assuming full responsibility for the publication’s accuracy and quality.

References

[1] C. A. Bomfim, L. G. F. Coelho, H. M. M. do Vale, "Brief History of Biofertilizers in Brazil: From Conventional Approaches to New Biotechnological Solutions," Brazilian Journal of Microbiology, vol. 52 no. 4, pp. 2215-2232, DOI: 10.1007/s42770-021-00618-9, 2021.

[2] E. T. Alori, M. O. Dare, O. O. Babalola, "Microbial Inoculants for Soil Quality and Plant Health," Sustainable Agriculture Reviews. Sustainable Agriculture Reviews, vol. 22, pp. 281-307, DOI: 10.1007/978-3-319-48006-0_9, 2017.

[3] B. G. Kelbessa, M. Dubey, V. Catara, F. Ghadamgahi, R. Ortiz, R. R. Vetukuri, "Potential of Plant Growth-Promoting Rhizobacteria to Improve Crop Productivity and Adaptation to a Changing Climate," CABI Reviews, vol. 11,DOI: 10.1079/cabireviews.2023.0001, 2023.

[4] Ec Regulation EU, "2019/1009 of the European Parliament and of the Council of 5 June 2019 Laying Down Rules on the Making Available on the Market of EU Fertilising Products and Amending Regulations (EC) No 1069/2009 and (EC) No 1107/2009 and Repealing Regulation (EC) No 2003/2003," 2019. https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=celex%3A32019R1009

[5] J. Meng, X. Zhang, X. Han, B. Fan, "Application and Development of Biocontrol Agents in China," Pathogens, vol. 11 no. 10,DOI: 10.3390/pathogens11101120, 2022.

[6] M. S. Santos, M. A. Nogueira, M. Hungria, "Microbial Inoculants: Reviewing the Past, Discussing the Present and Previewing an Outstanding Future for the Use of Beneficial Bacteria in Agriculture," AMB Express, vol. 9 no. 1, pp. 205-222, DOI: 10.1186/s13568-019-0932-0, 2019.

[7] S. M. de Sousa, C. A. de Oliveira, D. L. Andrade, "Tropical Bacillus Strains Inoculation Enhances Maize Root Surface Area, Dry Weight, Nutrient Uptake and Grain Yield," Journal of Plant Growth Regulation, vol. 40 no. 2, pp. 867-877, DOI: 10.1007/s00344-020-10146-9, 2021.

[8] A. Saeid, E. Prochownik, J. Dobrowolska-Iwanek, "Phosphorus Solubilization by Bacillus Species," Molecules, vol. 23 no. 11,DOI: 10.3390/molecules23112897, 2018.

[9] G. Kaschuk, P. A. Leffelaar, K. E. Giller, O. Alberton, M. Hungria, T. W. Kuyper, "Responses of Legumes to Rhizobia and Arbuscular Mycorrhizal Fungi: A Meta-Analysis of Potential Photosynthate Limitation of Symbioses," Soil Biology and Biochemistry, vol. 42 no. 1, pp. 125-127, DOI: 10.1016/j.soilbio.2009.10.017, 2010.

[10] D. M. Zeffa, L. H. Fantin, A. Koltun, "Effects of Plant Growth-Promoting Rhizobacteria on Co-Inoculation With Bradyrhizobium in Soybean Crop: A Meta-Analysis of Studies From 1987 to 2018," PeerJ 8:e7905,DOI: 10.7717/peerj.7905, .

[11] J. Z. Barbosa, M. Hungria, J. V. d. S. Sena, G. Poggere, A. R. dos Reis, R. S. Corrêa, "Meta-Analysis Reveals Benefits of Co-Inoculation of Soybean with Azospirillum Brasilense and Bradyrhizobium Spp. In Brazil," Applied Soil Ecology, vol. 163,DOI: 10.1016/j.apsoil.2021.103913, 2021.

[12] J. Z. Barbosa, M. Hungria, S. A. Prior, M. C. Moura, G. Poggere, A. C. V. Motta, "Improving Yield and Health of Legume Crops via Co-Inoculation with Rhizobia and Trichoderma : A Global Meta-Analysis," Applied Soil Ecology, vol. 176,DOI: 10.1016/j.apsoil.2022.104493, 2022.

[13] G. Kaschuk, A. C. Auler, C. E. Vieira, F. D. Dakora, S. K. Jaiswal, S. P. da Cruz, "Coinoculation Impact on Plant Growth Promotion: A Review and Meta-Analysis on Coinoculation of Rhizobia and Plant Growth-Promoting Bacilli in Grain Legumes," Brazilian Journal of Microbiology, vol. 53 no. 4, pp. 2027-2037, DOI: 10.1007/s42770-022-00800-7, 2022.

[14] M. Hungria, I. C. Mendes, "Nitrogen Fixation with Soybean: The Perfect Symbiosis?," Biological Nitrogen Fixation, pp. 1009-1023, DOI: 10.1002/9781119053095, 2015.

[15] M. Hungria, M. A. Nogueira, R. S. Araujo, "Co-Inoculation of Soybeans and Common Beans With Rhizobia and Azospirilla: Strategies to Improve Sustainability," Biology and Fertility of Soils, vol. 49 no. 7, pp. 791-801, DOI: 10.1007/s00374-012-0771-5, 2013.

[16] C. A. Machaculeha, M. F. Guimarães, O. R. Brito, M. Nogueira, R. S. Araujo, M. Hungria, "Co-inoculation of Soybean With Bradyrhizobium and Azospirillum Promotes Early Nodulation," American Journal of Plant Sciences, vol. 6, pp. 1641-1649, DOI: 10.4236/ajps.2015.610164, 2015.

[17] F. F. Araújo, A. A. Henning, M. Hungria, "Phytohormones and Antibiotics Produced by Bacillus Subtilis and Their Effects on Seed Pathogenic Fungi and on Soybean Root Development," World Journal of Microbiology and Biotechnology, vol. 21 no. 8-9, pp. 1639-1645, DOI: 10.1007/s11274-005-3621-x, 2005.

[18] L. G. Moretti, C. A. Crusciol, E. E. Kuramae, "Effects of Growth‐Promoting Bacteria on Soybean Root Activity, Plant Development, and Yield," Agronomy Journal, vol. 112 no. 1, pp. 418-428, DOI: 10.1002/agj2.20010, 2020.

[19] A. L. M. Oliveira, M. Stoffel, M. Schmid, V. M. Reis, J. I. Baldani, A. Hartmann, "Colonization of Sugarcane Plantlets by Mixed Inoculations With Diazotrophic Bacteria," European Journal of Soil Biology, vol. 45, pp. 106-113, 2009.

[20] D. Molina-Romero, A. Baez, V. Quintero-Hernández, "Compatible Bacterial Mixture, Tolerant to Desiccation, Improves Maize Plant Growth," PLoS One, vol. 12 no. 11,DOI: 10.1371/journal.pone.0187913, 2017.

[21] X. Liu, S. Mei, "Inoculated Microbial Consortia Perform Better Than Single Strains in Living Soil: A Meta-Analysis," Applied Soil Ecology, vol. 190,DOI: 10.1016/j.apsoil.2023.105011, 2023.

[22] K. Vishwakarma, N. Kumar, C. Shandilya, S. Mohapatra, S. Bhayana, A. Varma, "Revisiting Plant–Microbe Interactions and Microbial Consortia Application for Enhancing Sustainable Agriculture: A Review," Frontiers in Microbiology, vol. 11,DOI: 10.3389/fmicb.2020.560406, 2020.

[23] Fao-Food and Agriculture Organization of the United Nations, "FAOStat," 2023. https://www.fao.org/faostat/en/#data/QCL

[24] D. N. Rodríguez-Navarro, I. Margaret Oliver, M. Albareda Contreras, J. E. Ruiz-Sainz, "Soybean Interactions with Soil Microbes, Agronomical and Molecular Aspects," Agronomy for Sustainable Development, vol. 31 no. 1, pp. 173-190, DOI: 10.1051/agro/2010023, 2011.

[25] G. Kaschuk, X. Yin, M. Hungria, P. A. Leffelaar, K. E. Giller, T. W. Kuyper, "Photosynthetic Adaptation of Soybean Due to Varying Effectiveness of N 2 Fixation by Two Distinct Bradyrhizobium japonicum Strains," Environmental and Experimental Botany, vol. 76,DOI: 10.1016/j.envexpbot.2011.10.002, 2012.

[26] M. Hungria, J. C. Franchini, R. J. Campo, "Nitrogen Nutrition of Soybean in Brazil: Contributions of Biological N 2 Fixation and N Fertilizer to Grain Yield," Canadian Journal of Plant Science, vol. 86 no. 4, pp. 927-939, DOI: 10.4141/p05-098, 2006.

[27] G. Kaschuk, M. A. Nogueira, M. J. De Luca, M. Hungria, "Response of Determinate and Indeterminate Soybean Cultivars to Basal and Topdressing N Fertilization Compared to Sole Inoculation With Bradyrhizobium," Field Crops Research, vol. 195, pp. 21-27, DOI: 10.1016/j.fcr.2016.05.010, 2016.

[28] Brasil Ministério da Agricultura, P. E. Abastecimento, "Law No. 6,894. Diário Oficial da União-Section 1-17/12/1980, 25289," 1980. https://www.planalto.gov.br/ccivil_03/leis/1980-1988/l6894.htm

[29] A. W. Bakker, B. Schippers, "Microbial Cyanide Production in the Rhizosphere in Relation to Potato Yield Reduction and Pseudomonas Spp. Mediated Plant Growth-Stimulation," Soil Biology and Biochemistry, vol. 19 no. 4, pp. 451-457, DOI: 10.1016/0038-0717(87)90037-X, 1987.

[30] ISTA-International Seed Testing Association, International Rules for Seed Testing, 2008.

[31] H. G. Santos, P. K. T. Jacomine, L. H. C. Anjos, Sistema Brasileiro de Classificação de Solos, 2018.

[32] C. A. Alvares, J. L. Stape, P. C. Sentelhas, J. L. de Moraes Gonçalves, G. Sparovek, "Köppen’s Climate Classification Map for Brazil," Meteorologische Zeitschrift, vol. 22 no. 6, pp. 711-728, DOI: 10.1127/0941-2948/2013/0507, 2013.

[33] W. R. Fehr, C. E. Caviness, Stages of Soybean Development. Special Report 80, Iowa Agricultural Experiment Station, Iowa Cooperative External Series, 1977.

[34] G. W. Gee, J. W. Bauder, "Particle-Size Analysis," Methods of Soil Analysis, Part 1. Physical and Mineralogical Methods, Agronomy Monograph No. 9, pp. 383-411, 1986.

[35] Soil Survey Staff, Keys to Soil Taxonomy, 2022.

[36] AOAC-Association of Official Analytical Chemists, Official Methods of Analysis of the Association of Official Analytical Chemists (Method 991.20), 1995.

[37] ISO-International Standard Organization, ISO 1443: Meat and Meat Products-Determination of Total Fat Content, 1973.

[38] E. B. Ferreira, P. P. Cavalcanti, D. A. Nogueira, "ExpDes. Pt: Pacote Experimental Designs (Portuguese)," R package version, vol. 1, 2018.

[39] R. C. Team, R: A Language and Environment for Statistical Computing, 2013.

[40] P. Pandey, S. Bisht, A. Sood, A. Aeron, G. D. Sharma, D. K. Maheshwari, "Consortium of Plant-Growth-Promoting Bacteria: Future Perspective in Agriculture," Bacteria in Agrobiology Plant Probiotics,DOI: 10.1007/978-3-642-27515-9_10, 2012.

[41] Y. Besset-Manzoni, L. Rieusset, P. Joly, G. Comte, C. Prigent-Combaret, "Exploiting Rhizosphere Microbial Cooperation for Developing Sustainable Agriculture Strategies," Environmental Science and Pollution Research, vol. 25 no. 30, pp. 29953-29970, DOI: 10.1007/s11356-017-1152-2, 2018.

[42] D. M. Tekrony, D. B. Egli, A. D. Phillips, "Effect of Field Weathering on the Viability and Vigor of Soybean Seed 1," Agronomy Journal, vol. 72 no. 5, pp. 749-753, DOI: 10.2134/agronj1980.00021962007200050014x, 1980.

[43] M. J. d. Luca, M. Hungría, "Plant Densities and Modulation of Symbiotic Nitrogen Fixation in Soybean," Scientia Agricola, vol. 71 no. 3, pp. 181-187, DOI: 10.1590/S0103-90162014000300002, 2014.

[44] M. Javier de Luca, M. A. Nogueira, M. Hungria, "Feasibility of Lowering Soybean Planting Density Without Compromising Nitrogen Fixation and Yield," Agronomy Journal, vol. 106 no. 6, pp. 2118-2124, DOI: 10.2134/agronj14.0234, 2014.

[45] C. H. Queiroz-Rego, F. B. Cardoso, A. C. da Silva Cândido, P. E. Teodoro, C. Z. Alves, "Co‐Inoculation With Bradyrhizobium and Azospirillum Increases Yield and Quality of Soybean Seeds," Agronomy Journal, vol. 110 no. 6, pp. 2302-2309, DOI: 10.2134/agronj2018.04.0278, 2018.

[46] K. R. Gadhave, J. E. Hourston, A. C. Gange, "Developing Soil Microbial Inoculants for Pest Management: Can One Have Too Much of a Good Thing?," Journal of Chemical Ecology, vol. 42 no. 4, pp. 348-356, DOI: 10.1007/s10886-016-0689-8, 2016.

[47] D. B. Egli, "Plant Density and Soybean Yield," Crop Science, vol. 28 no. 6, pp. 977-981, DOI: 10.2135/cropsci1988.0011183X002800060023x, 1988.

[48] M. A. Silva, A. S. Nascente, M. C. C. D. Filippi, A. C. Lanna, G. B. D. Silva, J. F. A. E. Silva, "Individual and Combined Growth-Promoting Microorganisms Affect Biomass Production, Gas Exchange and Nutrient Content in Soybean Plants," Revista Caatinga, vol. 33 no. 3, pp. 619-632, DOI: 10.1590/1983-21252020v33n305rc, 2020.

[49] R. Rawat, L. Tewari, "Effect of Abiotic Stress on Phosphate Solubilization by Biocontrol Fungus Trichoderma Sp," Current Microbiology, vol. 62 no. 5, pp. 1521-1526, DOI: 10.1007/s00284-011-9888-2, 2011.

[50] R. Abdenaceur, B. T. Farida, D. Mourad, H. Rima, O. Zahia, S. H. Fatma, "Effective Biofertilizer Trichoderma Spp. Isolates With Enzymatic Activity and Metabolites Enhancing Plant Growth," International Microbiology, vol. 25 no. 4, pp. 817-829, DOI: 10.1007/s10123-022-00263-8, 2022.

[51] E. T. Iturralde, M. C. Stocco, A. Faura, "Coinoculation of Soybean Plants With Bradyrhizobium japonicum and Trichoderma harzianum : Coexistence of Both Microbes and Relief of Nitrate Inhibition of Nodulation," Biotechnology reports, vol. 26,DOI: 10.1016/j.btre.2020.e00461, 2020.

[52] A. K. Saxena, M. Kumar, H. Chakdar, N. Anuroopa, D. J. Bagyaraj, "Bacillus Species in Soil as a Natural Resource for Plant Health and Nutrition," Journal of Applied Microbiology, vol. 128 no. 6, pp. 1583-1594, DOI: 10.1111/jam.14506, 2020.

[53] H. Marschner, Mineral Nutrition of Higher Plants, 1995.

[54] R. Marra, N. Lombardi, G. d’Errico, "Application of Trichoderma Strains and Metabolites Enhances Soybean Productivity and Nutrient Content," Journal of Agricultural and Food Chemistry, vol. 67 no. 7, pp. 1814-1822, DOI: 10.1021/acs.jafc.8b06503, 2019.

[55] M. F. T. Rossetim, A. C. V. Motta, Y. R. Kondo, "Does Inoculation of Multifunctional Microbial Consortia Contribute to Compelling Increases in Soybean Yields?," Research Square, vol. 7,DOI: 10.21203/rs.3.rs-3304353/v1, 2023.