The study was performed in the “Hauts Bassin” region located in the western part of Burkina Faso. Two sites were selected to assess the impact of cultivation practices on the diversity of plant growth-promoting bacteria. The samples were collected at two sites of the Bama and Bobo Dioulasso districts (11˚21'8.766''N; −4˚23'16.6272''E and 11˚9'6.9948''N; −4˚15'16.236''E, respectively). The climate at both sites is South Sudanian, characterized by an alternating rainy season (May to October) and a dry season from November to April [14]. Maize productivity was particularly higher in these sites. Since soil fertilization management at these sites was mainly based on the use of organic fertilizers, this high productivity could be due to the effect of plant growth-promoting bacteria. Indeed, it has been shown that organic farming promotes the development of plant growth-promoting rhizobacteria [10].

During the sampling, the farming practices in the two study areas were recorded. The main parameters noticed were the cropping system, the fertilization mode, and the pesticides used.

Root samples were collected from the study sites following the method described by Agbodjato et al. [15]. At each site, the roots of ten (10) plants at maturity spaced ten (10) meters apart were harvested. A composite sample was made from the collected roots, and then three hundred grams (300 g) were taken and placed in sterile bags. All samples were stored at +4˚C.

Rhizobacteria were isolated using the method described by Agbodjato et al. [15] with modified media. A reconstituted Luria-Bertani (LB) medium supplemented with Agar (15 g/mL) and antifungal Nystatin (25 µg/mL) was used. The inoculated plates were incubated at 30˚C for 24 hours. Then, well-isolated colonies were purified after successive transfers on the same media. The purified colonies were stored at +4˚C in yeast-dextrose-CaCO₃ broth supplemented with 20% glycerol [16].

To evaluate the plant growth-promoting properties of the isolates, several tests were performed, including nitrogen fixation and the production of ammonia, hydrogen cyanide, Indole-3-Acetic Acid (IAA), and exopolysaccharides.

2.5.1. Nitrogen Fixation

The nitrogen fixation test was carried out using the method described by AlAli et al. [17], on the nitrogen-free medium (NFDM) broth. Each strain was inoculated in 5 ml of NFDM broth. And the optical density of the inoculated media was measured using a Spectronic 601 spectrophotometer at 600 nm. The measure of the OD was repeated after an incubation at 30˚C for 7 days. The nitrogen fixation was estimated using the method described by Giroux [18]. The positive media for nitrogen fixation were subcultured on the same culture medium supplemented with agar (15 g/l) and Nystatin (25 µg/ml), to confirm the results obtained.

2.5.2. Ammonia Production

For each strain, one colony was used to inoculate 5 ml of peptone broth. The mixture was incubated at 30˚C for 72 hours. The amount of ammonia produced was assessed by adding 1 ml of Nessler’s reagent to the culture. The light yellow color of the mixture was attributed to low ammonia production, while the dark yellow color was considered as to high ammonia production [19].

2.5.3. Hydrogen Cyanide Production

For this test, the nutrient agar media supplemented with glycine (4.4 g/l) was inoculated with each strain. The inoculated plates were then covered with Whatman paper soaked in 5 ml of a mixture of 2% sodium carbonate and 0.5% picric acid, and incubated at 30˚C for 24 hours. The positive production of hydrogen cyanide was evidenced by the color change of the Whatman paper from yellow to reddish-brown [20].

2.5.4. Production of Indole-3-Acetic Acid (IAA)

The ability of the strains to produce Indole-3-Acetic Acid (IAA) was assessed using the modified method described by Al-Kahtani et al. [21] using Tryptophan broth as media. Tryptophan broth (10 ml) was inoculated with each isolate and incubated at 30˚C for 24 hours. After 24 hours, the supernatant from each culture obtained after centrifugation at 6000 g for 30 minutes was collected. Then, 2 ml of Salkowski’s reagent was added to 1 ml of the supernatant. The appearance of a pink color indicated the production of IAA.

2.5.5. Assessment of the Exopolysaccharides Production

The exopolysaccharide production was evaluated using RVC-Sucrose broth. A quantity of 50 ml of broth was inoculated with each isolate and incubated at 30˚C for 24 hours. The exopolysaccharides were collected by centrifugation at 8000 g for 20 minutes according to the method described by Bajpai et al. [22].

After the evaluation of the plant growth-promoting properties of the rhizobacteria, some isolates were selected to determine their real effect on the germination of maize seeds.

2.6.1. Plant Material

The FBC6 variety developed by the Institute of Environment and Agricultural Research (INERA), Farako-bâ in Burkina Faso, was used. This variety is drought-tolerant, resistant to Striga, and can produce up to 5.6 tons/ha of maize grain [23].

2.6.2. Inoculum Preparation

Each colony was inoculated in 50 ml of LB broth and incubated at 30˚C for 24 hours. Then, each mixture was centrifuged at 4000 g and 4˚C to collect the bacterial pellets [24]. Each pellet was diluted with sterile distilled water to obtain a bacterial concentration of 10⁸ CFU/mL on the spectrometer at 600 nm according to the McFarland standard.

2.6.3. Inoculation of Maize Seeds with Selected Rhizobacteria

Before the maize seeds were inoculated with the rhizobacterial strains, they were sterilized: the seeds were treated with sodium hypochlorite (2%) for 2 minutes, washed with distilled water, and then with ethanol (95%) for 30 seconds [24]. The sterilized seeds were incubated in the inoculum of each strain for 30 minutes, and placed on cotton wool in Petri dishes at a rate of 10 seeds per dish. Five replicates were performed for each strain and incubated at 25˚C for 7 days. Sterile distilled water was used to water the seeds every day at a rate of 5 ml per Petri dish.

2.6.4. Data Collection on Seed Emergence

Seed emergence was observed every 24 hours for 7 days. The lengths of the radicle and coleoptile were measured each day using a cotton thread and a graduated ruler. Dry root biomass was determined by weighing using a RADWAG PS 210 R2 electronic balance (Polish company RADWAG, Poland).

Germination Rate (GR) and Elongations (E) of radicle and coleoptile were determined using Equations (1) and (2) described by D. Come [25], respectively.

n = total number of radicles or coleoptiles, and L_i = individual length of radicle or coleoptile.

Root moisture content was determined using Equation (3):

H = Moisture content; M₀ = Initial mass; M = Mass after drying.

For the most promising strains, the nature of the wall was observed under a microscope after Gram staining. Sporulation was determined using the heat shock method. Cytochrome oxidase and catalase production were determined using oxidase (OX) discs and hydrogen peroxide (10%), respectively. The use of sugars by the strains was tested on a minimal medium supplemented with agar (15 g/l) and 1% (w/v) of melibiose, glycogen, L-(+)-arabinose, D-(-)-fructose, D-(+)-glucose, and sucrose, respectively [26]. Bromothymol blue (0.25 g/l) was added to the broths to evaluate acid production. Mannitol metabolism and strain motility were tested using Mannitol Motility Test Agar. Glucose and lactose fermentation were tested on Kliger-Iron Agar. Citrate utilization, urea utilization, and indole production were tested using Simmons Citrate Agar and Urea-Indole Broth, respectively. The temperature effect on the growth of strains was tested at 4˚C, 41˚C, 45˚C, and 65˚C on LB broth supplemented with agar (15 g/l). The growth of strains at pH 5.7 was determined on LB broth supplemented with agar (15 g/l). Finally, the growth of the strains at 7% NaCl was assessed.

The physiological and biochemical characteristics of the strains were used to inform the ABIS Online platform, which was used for their presumptive identification. The databases consulted were those for Pseudomonas and Bacillus databases, 8.23.25-052025 and 4.8.11-032025 versions, respectively. The optimal number of tests for the Bacillus database was 18, with a minimum requirement of 9. The optimal number of tests for the Pseudomonas database was 24, with a minimum requirement of 9.

EXCEL-2019 software was used for data recording. Principal Component Analysis (PCA) was used to analyze the plant growth-promoting properties data of the bacterial strains using R software version 4.3.1. The results were visualized in the form of correlation circles and group graphs to identify promising strains. The means of the data were compared using the ANOVA test, Tukey at a probability threshold of p = 0.05.

Different farming practices were applied in the study sites (Table 1). The soil fertility management option in the Bama district was crop association combined with organic fertilization. Maize was grown in association with tomato, combined with chicken droppings and compost made from neem residues at the recommended rate of 6 t/ha. The soil fertility management option in the Bobo Dioulasso area was based on the use of compost made from agricultural residues combined with NPK (14-23-14) and urea (46% N) at doses below those recommended. In addition, on both sites, no synthetic chemicals were used in the crop protection.

Table 1. Cultivation practices at the study sites.

A total of 235 bacterial strains were isolated, of which 168 strains were from Bama and 67 from Bobo Dioulasso district. Figure 1 presents the Principal Component Analysis (PCA) of strain diversity based on their plant growth-promoting properties. All strains were at least positive for one of the five (05) tested plant growth-promoting properties. The Principal Component Analysis (PCA) showed two main axes. The horizontal axis contributed 29.6% and was positively correlated with ammonia production, exopolysaccharides, IAA, and atmospheric nitrogen fixation. The vertical axis contributed 21.8% and was positively correlated with hydrogen cyanide production, exopolysaccharides, IAA, and atmospheric nitrogen fixation.

Figure 1. PCA of strain diversity according to plant growth-promoting properties. Variables were HCN: ability to produce hydrogen cyanide; AZO: ability to fix atmospheric nitrogen; EPS: ability to produce exopolysaccharides; AMM: ability to produce ammonia; IAA: ability to produce Indole-3-Acetic Acid.

The cluster graph showed three main clusters. Table 2 presents the percentages of isolates in each group, as well as their distribution by site.

Table 2. Group distribution frequency (%) of the strains per site.

The first group comprised 81% of all strains, with 71% from Bama and 29% from the Bobo Dioulasso site (Table 2). This group is characterized by the production of Indole-3-Acetic Acid, Ammonia, and to fix atmospheric nitrogen, and the absence of exopolysaccharide and hydrogen cyanide production (Table 3).

Table 3. Frequency distribution (%) of combined plant growth-promoting properties expressed by group 1 strains.

AZO: N₂ fixation; AMM: NH₃ production; IAA: Indole-3-Acetic Acid production.

The majority of strains in group 1 are capable of producing several plant growth-promoting factors at the same time. Indeed, 10% to 49% of the strains exhibit at least two properties simultaneously, while only 1% of the strains produce IAA alone (Table 3). Table 4 presents frequencies (%) of plant growth-promoting property combinations expressed by group n˚ 2 strains. The second group was composed of 15% of all strains with an abundance of 94% at the Bama site (Table 2). The rhizobacteria in this group were mainly unable to produce HCN (Table 4). In this group, only Bama strains, representing 3% of the strains, can fix atmospheric nitrogen, produce IAA, and exopolysaccharides simultaneously (Table 4). In addition, 14% of the bacteria, which were isolated from Bama, could produce exopolysaccharides, ammonia, and Indole-3-Acetic Acid simultaneously (Table 4).

Table 4. Frequencies (%) of plant growth-promoting property combinations expressed by group 2 strains.

AZO: N₂ fixation; AMM: NH₃ production; IAA: Indole-3-Acetic Acid production; EPS: exopolysaccharide production.

In the second group, 83% of the bacteria were able to fix atmospheric nitrogen and produce IAA, exopolysaccharides, and ammonia concurrently (Table 4).

The third group consists of only 4% of strains, all isolated from the Bobo Dioulasso site (Table 2). The bacterial strains in this group were mainly able to produce HCN (Table 5). The ability to fix atmospheric nitrogen, produce hydrogen cyanide, and IAA was expressed by 56% of the strains in this group (Table 5).

Table 5. Frequencies (%) of plant growth-promoting property combinations expressed by group 3 strains.

AZO: N₂ fixation; AMM: NH₃ production; IAA: Indole-3-Acetic Acid production; EPS: Exopolysaccharide production; HCN: Hydrogen cyanide production.

The ability to fix atmospheric nitrogen and produce hydrogen cyanide, ammonia, and IAA was expressed by 22% of the strains (Table 5). In addition, 11% of the strains can produce ammonia, IAA, and hydrogen cyanide. Globally, 11% of the strains of this group were found to have all of the desired properties.

Ten (10) bacterial strains exhibiting the best plant growth-promoting abilities were selected and tested on the germination of maize grain. The selected strains are able to produce IAA, ammonia, and exopolysaccharides, and are able to fix nitrogen (Table 6). In addition, the strain 614B was able to produce hydrogen cyanide. All selected strains except strain 614B originate from groups 2 and 3 and from the Bama site.

Table 6. Growth-promoting characteristics of the selected strains.

-: no production; ++: average production; +++: good production; ++++: very good production. EPS = Exopolysaccharide; NH₃ = Ammonia; N₂ = atmospheric nitrogen; HCN = hydrogen cyanide.

The inoculation of the grains with all of the selected strains significantly increased all of the germination parameters (p < 0.001) except the germination rate (Table 7). The longest radicle length was obtained with the strain 3127A (Table 7). The inoculated grains showed the highest coleoptile elongation compared to the non-inoculated control (p < 0.001). In addition, the number of roots of the inoculated seedlings was significantly higher than that of the control (p < 0.001). Indeed, seedlings treated with strains 668A, 3127A, 415A, 412A, and 614B had 6 seminal roots, compared with control seedlings, which had 02 seminal roots (Table 7). Furthermore, the maize grain inoculation with the strains significantly improved (p < 0.001) the seedling fresh root biomass (Table 7). The strains 493A and 3127A evidenced the best values with 214.8 mg/plantlet and 268 mg/plantlet obtained, respectively (Table 7). Moreover, the maize grain inoculation with the strains significantly improved the seedlings’ root dry biomass (p < 0.001), with strain 3127A showing the best value (30.40 mg/plantlet). The strains 7134A, 3123A, and 412A significantly increased the root moisture content of the seedlings (p < 0.001), with strain 7134A presenting the highest value of 92.44% of humidity.

Table 7. Effects of selected rhizobacterial strains on maize germination parameters.

For a given column, values with different letters are significantly different at the 5% threshold (Tukey α = 0.05). T = Treatments; CT = Control; GR = Germination rate; RL = Radicle length; CL = Coleoptile length; FRB = Fresh root biomass; DRB = Dry root biomass; RM = Root moisture content; NSR: Number of seminal roots.

Table 8 presents the biochemical and physiological characteristics of strains 668A, 3127A, 412A, 415A, and 614B. These strains were chosen because of their ability to significantly increase the number of seminal roots in plants. Since increasing the number of seminal roots improved mineral uptake by maize plants, these strains could further improve maize growth under natural conditions. All of the tested strains produced cytochrome oxidase, catalase, and indole (Table 8). Except for strain 412A, all strains were urease-negative. None of the strains was able to grow at pH 4. However, all of them are able to grow at the tested NaCl concentrations (Table 8). The strains are mesophilic since no growth was obtained at 4˚C, 45˚C, 55˚C, and 65˚C. However, they did grow at 20˚C and 30˚C. Except for strain 614B, all strains were able to grow at 41˚C. Citrate was metabolized by all strains. Mannitol was metabolized by all strains except strain 614B. However, its metabolism was followed by acid production by strains 3127A, 668A, and 415A (Table 8). All strains were capable of metabolizing Melibiose, glycogen, L-(+)-Arabinose, D-(-)-Fructose, D-(+)-Glucose, and Sucrose (Table 8). However, their metabolism was accompanied by acidification of the medium for strains 3127A, 668A, and 415A. Lactose and glucose were fermented by strains 3127A, 668A, and 415A with gas production (Table 8). None of the strains produced hydrogen sulfide.

Table 8. Biochemical and physiological characteristics of selected strains.

+: positive test; −: negative test; + a: with acid production.

Table 9 shows the presumptive identification of the strains. Based on these physiological and biochemical characteristics and using the ABISonline platform, the strains 3127A, 668A, and 415A could be affiliated with the Paenibacillus genus (Table 9). The strain 614B could be affiliated with the Sphingomonas genus with a membership probability of 80.6% (Table 9). The strain 412A could be affiliated with the genus Burkholderia with a membership probability of 64.4% (Table 9).

Table 9. ABIS identification of the strain.