Materials and methods

Microbial resources. A collection of twelve bacteria obtained in a previous work, from an agricultural soil and a pine forest soil, were evaluated as PGPB for their ability to produce in-vitro IAA (I), siderophores (S) and /or solubilizing phosphate (P) (^{Flores-Núñez et al. 2018}). All bacteria produce IAA, and based on their PGP activities, they were grouped as follows: bacteria that only produce IAA, called as BI group (n = 3); bacteria that also produce siderophores, BIS group (n = 2); bacteria that additionally solubilize phosphate, BIP group (n = 4); and bacteria that solubilize phosphate and produce siderophores, BIPS group (n = 3). Table 1 shows the bacteria used in this study.

Bacterial inoculum. Bacteria were inoculated onto nutrient agar plates and incubated for 24 h at 28 °C. Subsequently, a suspension of each bacterial species was prepared with sterile saline solution and adjusted to OD_600mm = 0.5, then 1 mL of this suspension was inoculated into flasks containing 49 mL of nutrient broth and was incubated at 20 ºC overnight. Subsequently, bacterial growth was evaluated by CFU and a concentration of 10⁷-10⁸ CFU mL^-1 was inoculated to each seedling.

The bacteria that had significant results in the different growth parameters of the sorghum seedlings in the hydroponics bioassay were selected to make mixtures among them and tested in the low-nutrient soil bioassay (as mentioned below). Multi-bacterial inoculum was prepared from pure cultures, following the previously described procedure. For this, 5 mL of each bacterium was taken and mixed.

Hydroponic nutrient solution (HNS). The synthetic hydroponic nutrient solution (HNS) was prepared from a macronutrient stock solution that contained: 1 mM NH₄H₂PO₄, 5.75 mM Ca(NO₃)₂, 1.3 mM CaCl₂, 2 mM MgSO₄•7H₂O, 6 mM KNO₃, 24 µM H₃BO₃, 12 nM Na₂MoO₄•H₂O, 0.24 µM CuCl₂•H₂O, 5µM ZnCl₂, 0.88 µM MnCl₂•4H₂O and 14.2 nM Fe-EDTA. Modifications to the HNS were made in the assays of assimilation of phosphorus, wherein the phosphorus concentration was changed as follows: a) without phosphorus, HNS-wP, NH₄H₂PO₄ was replaced by the addition of 0.5 mM NH₄NO₃; b) half of phosphorus as insoluble tricalcium phosphate Ca₃(PO₄)₂ and the other half as soluble NH₄H₂PO₄, HNS-mP, was composed of 0.5 mM NH₄H₂PO₄ (instead of 1 mM), 0.5 mM Ca₃(PO₄)₂ and 0.25 mM NH₄NO₃ (to compensate the nitrogen source); and c) total phosphorus as insoluble tricalcium phosphate Ca₃(PO₄)₂, HNS-tP, 1 mM Ca₃(PO₄)₂, and 0.5 mM NH₄NO₃.

Sorghum seed germination. The seeds of Sorghum bicolor (L.) Moench were disinfested with 3 % NaClO for 1 min and subsequently washed with sterile distilled water. The disinfected seeds were placed on a paper previously wet with sterile distilled water with a separation of 3 cm between each seed, the paper containing the seeds was rolled up and incubated at 28 ºC in darkness for 6 days.

Hydroponics bioassay. The hydroponics bioassay was done with tyndallized vermiculite as the support for the development of sorghum seedlings. Inoculation of IAA-producing bacteria was conducted by dispensing a bacterial suspension from 10⁷ to 10⁸ CFU mL^-1 over the neck area of the five-day-old sorghum homogenous seedlings, which were in containers with vermiculite. Subsequently, 15 mL of HNS was added to allow the inoculum to cover the root, facilitating colonization of the rhizosphere. The bioassay was kept during four weeks under greenhouse conditions with the exposure of 12 h of light and 12 h of darkness at room temperature (25-30 ºC), watering every other day with 15 mL of HNS diluted with distilled water in a ratio of 1:4. Each treatment included a total of 6 replicates and a control group was included without the inoculation of any bacteria.

Low-nutrient soil bioassay. This was performed with a mixture of sterile sand (121 °C, 15 PSI, for 15 min) with sieved soil (< 2 mm) from Cerro de la Estrella, Mexico City (19° 20' 31" N, 99° 05' 22" W, altitude 2,460 m); the sieved soil was sterilized by tyndallization by 3 cycles of flowing steam for 45 min, with 24 h of aeration at room temperature in each cycle. The sand and soil were mixed in a ratio of 2:5 volume/volume. The soil is considered as a low-nutrient soil due to the organic matter (4.46 %), nitrogen content (0.13 %) and soluble phosphorus (1.7 % mg Kg^-1) as determined previously (^{Gómez-Acata et al. 2019}). Two control groups without bacterial inoculation were included, one was irrigated only with water and the other watered with HNS.

In the case of the phosphate-solubilizing bacteria, the test was carried out in a similar way to the inoculation of the bacteria described above, with the following modifications: a) HNS without phosphorus (HNS-wP); b) HNS with the half of phosphorus as insoluble tricalcium phosphate and half of the remaining phosphorus as NH₄H₂PO₄ (HNS-mP) and c) HNS with total of phosphorus as insoluble tricalcium phosphate (HNS-tP).

The bioassay was maintained for a duration of four weeks under greenhouse conditions with the exposure of 12 h of light and 12 h of darkness at room temperature (25-30 ºC), irrigating every seven days with the respective modified HNS and every other day with distilled water. Each treatment included six replicates.

Growth parameters of the sorghum seedlings. For both bioassays (hydroponics and low-nutrient soil), the seedlings growth parameters were determined. The height of the shoot was determined from the neck to the tip of the apical tissue. The stem diameter was determined at the height of the neck of the plant with a vernier caliper. The shoot and the root of the plant were cut, and both were placed in an oven at a temperature of 70 °C, for three days. After this time, the dry weight of the shoot and the root was determined.

Statistical analysis. Firstly, the results obtained from the bioassays were determined if they comply with the normality and homoscedasticity tests, to apply a parametric test. In all experiments, the results behaved as parametric data (Table S1). The parametric test used for the statistical analysis was a one-way ANOVA by Dunnett's statistical test with a significance level of 0.05. All statistical tests were performed with the GraphPad PRISM^® program ver. 8.0.1.

Results

Evaluation of non-solubilizing phosphate (PO ₄ ^-3 ) bacteria on sorghum seedlings in hydroponics bioassays. It was observed that there is a significant difference in plant growth in the presence of bacteria with respect to the control. Regarding the stem diameter of sorghum, the most noticeable result was that obtained with Bacillus simplex BIS-4 (Table 1; BIS group; Figure 1A), which produced an 18 % increase compared to the control group (sorghum without bacterial inoculum; P < 0.05). Regarding the height of shoot, it was observed that in four of the five bacteria, there was an increase that ranged from 7 to 15 % with respect to the control group (P < 0.05), except for Bacillus drentensis BIS-5 (Figure 1B). Regarding the effect on the biomass of the shoot, all the bacterial strains increased from 28 to 55 % with respect to the control group (P < 0.05; Figure 1C). In addition, a significant increase in the dry weight of the root was also observed with the treatments of the strains Pseudomonas monhii BI-2, Bacillus subtilis BI-3 and B. simplex BIS-4, of 76, 59 and 42 % compared to control group (P < 0.05; Figure 1D), respectively.

Figure 1 Effect of inoculation of IAA-producing strains (groups BI and BIS) on sorghum seedlings under hydroponic conditions. A) diameter of the stem, B) height of the shoot, C) shoot dry weight, and D) root dry weight. Error bars represent the standard deviation. *Treatments that show a significant difference with respect to the control (HNS) using the Dunnett's test (P < 0.05) (n = 6). 

In general, it is observed that all the strains of the BI group had a positive effect on most of the growth parameters of the sorghum seedlings, with a better performance of the Pseudomonas monhii BI-2 strain over the Pseudomonas sp. BI-1 strain. In the BIS group, only the B. simplex BIS-4 strain exerted a positive effect. Taking into consideration these results, all bacteria from groups BI and BIS were selected for the following low-nutrients soil bioassay.

Effect of phosphate (PO ₄ ^-3 )-solubilizing bacteria on sorghum under hydroponic bioassay. The groups of bacteria that produce IAA and solubilize phosphate, BIP and BIPS groups (Table 1), were tested for their ability to solubilize phosphate, for this purpose two bioassays of availability of phosphate were included, one corresponds to applying half of the phosphorus (mP) as tricalcium phosphate (insoluble phosphate) and the remaining as diacid ammonium phosphate (soluble phosphate) (Figure 2), and in the other bioassay to adding only insoluble tricalcium phosphate (tP; Figure 3). The following controls were included, without the inoculation of any bacteria with HNS, HNS without phosphorus (HNS-wP), water only, and HNS with 50 % tricalcium phosphate (HNS-mP). In addition, a treatment with the inoculation of a non-solubilizing phosphate bacterium (Pseudomonas sp. BI-1) with HNS-mP was included.

Figure 2 Effect of inoculation of phosphate solubilizing bacteria (BIP and BIPS groups) on sorghum seedlings under 50 % phosphate condition. Diameter of the stem in a bioassay with 50 % of tricalcium phosphate using vermiculite as support. Error bars represent the standard deviation. * Treatments showing significant difference with respect to the control of sorghum without bacterial inoculum and with HNS including 50 % tricalcium phosphate (HNS-mP). The Dunnett test (P < 0.05) (n = 6) was utilized. 

Figure 3 Effect of inoculation of phosphate solubilizing bacteria (BIP and BIPS groups) on sorghum seedlings under 100 % phosphate condition. A) shoot dry weight, and B) root dry weight in a bioassay with 100 % of tricalcium phosphate using vermiculite as support. Error bars represent the standard deviation. * Treatments showing significant difference with respect to the control of sorghum without bacterial inoculum and with HNS including 100 % tricalcium phosphate (HNS-tP). The Dunnett test (P < 0.05) (n = 6) was utilized. 

In the bioassay with 50 % insoluble tricalcium phosphate, the controls corresponding to the growth of sorghum in the presence of HNS and 50 % insoluble phosphate (HNS-mP) did not show a significant difference between them; however, in the sorghum controls growing without phosphate (water or HNS-wP), there was a significant difference in all the growth parameters compared to the control in the presence of HNS, indicating the importance of phosphate for the growth of sorghum.

Regarding to the inoculated strains in the bioassy with 50 % insoluble tricalcium phosphate, Pseudomonas mohnii BIPS-10 and Pseudomonas corrugata BIPS-11 improve the stem diameter of the plant with an increment of 22 and 43 %, respectively, as compared to the HNS-mP control, without bacterial addition (P < 0.05; Figure 2), the Pseudomonas sp. BI-1 strain did not improve this parameter because it does not solubilize phosphate (control). Regarding the other three parameters no significant differences were observed (Figure S1).

In the bioassay with 100 % of insoluble tricalcium phosphate, the controls behaved similarly to the previous bioassay with 50 % of insoluble tricalcium phosphate. In the treatments with the inoculated strains, significant increases were observed in the shoot dry weight (Figure 3A) and the root dry weight (Figure 3B) for P. mohnii BIPS-10 and P. corrugata BIPS-11 strains, with respect to the HNS-tP control without bacterial addition (P < 0.05); in contrast to the Pseudomonas sp. BI-1 strain (control) in which the percentage was lower. With the other two parameters no significant differences were obtained (Figure S2). From these results, P. mohnii BIPS-10 and P. corrugata BIPS-11 strains are the best options, since they have greater ability to solubilize phosphate in both conditions tested.

Effect of PGPB on sorghum seedlings using a low-nutrients soil bioassay. Based on the results obtained in the hydroponic conditions, the strains selected from the BI and BIS groups (Table 1) and the P. monhii BIPS-10 and P. corrugata BIPS-11 strains provided the best results. The next challenge was to test these selected strains in a low-nutrients soil bioassay. The bacteria were tested alone and in a multi-strains inoculum M1 (P. monhii BI-2 + B. simplex BIS-4 + P. monhii BIPS-10) and M2 (P. monhii BI-2 + B. simplex BIS-4 + P. corrugata BIPS-11).

The treatments with Pseudomonas sp. BI-1 and P. mohnii BIPS-10 strains increased significantly (P < 0.05) the stem diameter by 32 % compared to the control (sorghum without bacterial inoculum and irrigated only with water, Figure 4A). In contrast, eight of the nine treatments significantly improved the shoot height (P < 0.05) from 23 to 48 % in relation to the control (Figure 4B), being the strain Pseudomonas sp. BI-1 the one that achieved the highest percentage (48 %). Finally, seven of the nine treatments increased the biomass of the shoot dry weight significantly (P < 0.05), from 61 to 140 % compared to the control, highlighting the results obtained with Pseudomonas sp. BI-1 (140 %) and P. mohnii BIPS-10 (123 %) strains, with respect to the control (Figure 4C), a similar result was observed in the dry weight of the root of Pseudomonas sp. BI-1 (79 %) and P. mohnii BIPS-10 (67 %), with respect to the control (Figure 4D). Although most strains presented increases in sorghum growth, Pseudomonas sp. BI-1 presented the highest percentages in the four parameters tested. For example, 32, 48, 140 and 79 % in stem diameter, height and dry weight of the shoot and dry weight of the root, respectively, followed by the P. mohnii BIPS-10 strain. The other bacterial strains also improved the growth of sorghum, but not in the four parameters analyzed. For this reason, they are of less interest.

Figure 4 Effect of inoculation of selected PGPB on sorghum seedlings using a low-nutrient soil bioassay. A) diameter of the stem, B) height of the shoot, C) shoot dry weight and D) root dry weight. *Treatments that show significant difference with respect to the control irrigated with water. The Dunnett test (P < 0.05) (n = 6) was utilized. 

Discussion

PGPB promotes plant growth and development through a variety of mechanisms. The exact mechanism by which PGPB stimulates plant growth is not clearly understood, although various mechanisms such as the production of phytohormones, the suppression of harmful organisms, the phosphate solubilization and the improvement in absorption of mineral nutrients are involved in promoting plant growth (^{Santos et al. 2019}). This work was focused on determining the effect of PGPB for promoting plant growth in sorghum.

First, we analyzed the bacteria that only produced IAA in the hydroponic bioassay using vermiculite as support. It was observed that most of the tested strains increased different growth parameters of sorghum. These parameters are established as indicators of the PGP, since an increase in the shoot height of the plant is highly relevant for the access to sunlight to carry out photosynthesis (^{Mahanty et al. 2017}), making it more competitive by minimizing its permanence in greenhouse conditions, and therefore the cost associated with it. Furthermore, the increase in the height of shoot of the sorghum has the effect of a higher productivity of the grain (^{Grover et al. 2013}). The same occurs with the increase in root biomass, which represents a benefit to the plant, since it increases the absorption surface for the uptake of nutrients and water, in addition to improving its stability since it has a better anchorage to the ground (^{Santos et al. 2019}). It was observed that the majority of the tested strains increased different growth indicators of sorghum, and the increase in sorghum root biomass generated by the studied strains could be attributable to the IAA production, since the IAA production has been reported to promote root proliferation (^{Khalid et al. 2004}, ^{Keswani et al. 2020}).

Regarding the effect of phosphate solubilizing bacteria on the growth of sorghum seedlings using an HNS with 50 % tricalcium phosphate, the results were unexpected. In the study reported by ^{Flores-Núñez et al. 2018}, it was exhibited that Burkholderia phytofirmans BIP-7, Phylobacterium sp. BIP-8, Chitinophaga japonensis BIP-9, Pseudomonas corrugata BIPS-11 and Burkholderia graminis BIPS-12 solubilized tricalcium phosphate in vitro. However, under the experimental conditions tested here, these same strains did not improve the growth of sorghum seedlings. This result was unexpected since there is evidence that phosphate-solubilizing bacteria increase the yields of different crops in both, at greenhouse level and in an experimental field (^{Chuang et al. 2007}, ^{Valverde et al. 2007}). The probable causes that led to this result could be that these strains inhibit the activity of solubilizing phosphate in the presence of the sorghum plant, another possibility is that the activity of solubilizing phosphate by these strains is low or the amount of inoculum used was not adequate to meet the demand of available phosphate for sorghum. More experiments are required to answer these questions.

The evaluation of the effect of PGPB on sorghum seedlings using a low-nutrient soil as support yielded significant increases in stem diameter, height and biomass of shoot, and biomass of root. These increases are attractive since it demonstrates that the plant-microorganism interaction allowed a better utilization of the resources provided by the soil, allowing the plant to grow more effectively.

It was observed that there was a better growth of sorghum when a single bacterial inoculum was used instead of a multi-strains inoculum. This could be explained by the antagonistic activity of some metabolite or by a simple competition for the space on the root surface among the members of the inoculum. However, it would be necessary to carry out the compatibility experiments that confirm the above. The biochemical activity of the bacteria being in consortium is complex and perhaps our bacterial mixtures were not adequate because of their similar PGP activities, such as P. mohnii BI-2, B. simplex BIS-4 and P. mohnii BIPS-10, with PGP activities that overlap, since they all produce IAA and two of them produce siderophores. It is possible that a multi-strains inoculum with different PGP activities produce a synergistic effect, such as Pseudomonas sp. TR15 (ACC deaminase producer) and Bacillus aerophilus TR15c (siderophore producer) which have a synergistic effect promoting growth of Helianthus annuus L. (^{Kumar et al. 2021}). In fact, our results are similar to those obtained in another study (^{Abdel-Salam et al. 2015}), in which inoculated sorghum seedlings with Azotobacter chroococcum, Azospirillum brasilense, Bacillus megaterium and Bacillus circulans individually and in different bacterial mixtures, obtaining increases of 63-81 % with single strains rather than bacterial mixtures. Similarly, the individual inoculation of Luteibacter sp. XiU1292, Pseudomonas vranovencis XiU1297 or Acinetobacter iwoffii XiU12138 in sorghum seedlings, improved the growth parameters evaluated 4 weeks after inoculation (^{Mull et al. 2017}). Luteibacter sp. XiU1292 produces IAA, siderophores and solubilizes phosphate, similarly to the P. mohnii BIPS-10 strain reported in this work. It has also been reported that the inoculation of Bacillus spp. in sorghum seedlings generated better growth, in terms of shoot length and root biomass, with an average increase of 7 to 9 % compared to non-inoculated plants (^{Grover et al. 2013}). The strains used in that study are IAA producers, solubilize phosphate and do not produce siderophores, which are the same PGP activities of our strains. In contrast, soybeans co-inoculated with Bradyrhizobium japonicum and A. brasilense produced a better result compared to the individual-strain inoculum (^{Hungria et al. 2013}). Furthermore, the same result was obtained with the co-inoculation of beans with Rhizobium tropici and A. brasilense. This suggests that the synergistic effects depend on the bacterial species involved.

Pseudomonas sp. BI-1, which only produces IAA as the only PGP activity, was one of the strains that yielded the best growth promotion results for sorghum. In this context, it has been shown that the inoculation of sorghum with A. brasilense strain SM with IAA production as the only PGP activity, produced increase of shoot height, and shoot dry weight (10-weeks), shoot height, and shoot dry weight (16-weeks) (^{Kochar & Srivastava 2011}). This suggests that in the case of sorghum, the association with a bacterium that produces IAA is enough to improve its growth.

In view that P. mohnii BIPS-10 (IAA producer, phosphate solubilizer and siderophore producer) improved sorghum growth below the Pseudomonas sp. B1, that only produces IAA, this could suggest that a bacterium with multiple PGP activities does not guarantee that it is better compared to a bacterium with mono-PGP activity. In the same context, ^{Kumar et al. (2012)}, mentioned that the possession of multiple PGP activities in an isolate does not imply that its utilization as inoculum for the plant will yield better results. In fact, these authors showed that the strain Pseudomonas sp. P28 (IAA producer, siderophore, HCN, protease, phosphate solubilizer) did not increase the growth of sorghum seedlings with respect to the control, unlike the strain Pseudomonas sp. P38 (with a single PGP activity, P-solubilization), which increased the root volume and root dry weights of sorghum in 10 %, compared to the control. There are other works that describe the same behavior, but in other bacterial genera (^{Mareque et al. 2015}, ^{da Silva et al. 2018}).

Apparently, in sorghum, the bacterial IAA is essential to improve its growth, or it may be possible that the bacterial species promote the improvement of this parameter. Strains of Pseudomonas genus inoculated to plants of Phaseolus vulgaris (beans), Zea mays (corn), Solanum tuberosum (potato) and Lycopersicum esculentum (tomato), under greenhouse conditions has promoted plant growth (^{Santillana-Villanueva 2006}). In another study, Solanum lycopersicum L. (tomato), Abelmoschus esculentus (okra), and Amaranthus sp. (African spinach) were inoculated with the cultures of P. aeruginosa and B. subtilis, and observed increases in the different growth parameters of these plants (^{Adesemoye et al. 2008}). Nonetheless, the above shows that the genus Pseudomonas are rhizobacteria that improve the growth parameters of various plants, including sorghum.

Furthermore, it has been shown that the most surprising results from the inoculation of PGPB are presented when utilizing soils with nutritional deficiencies, poor irrigation and adverse weather conditions (^{Grover et al. 2013}, ^{Inostroza et al. 2017}). In our case, the soil that was utilized contained serious nutritional deficiencies in organic matter, nitrogen and soluble phosphorus, which were previously determined (^{Gómez-Acata et al. 2019}). Therefore, with the results obtained, we corroborate the importance of the inoculation with Pseudomonas sp. BI-1 and/or P. mohnii BIPS-10 to improve sorghum growth under adverse conditions. It is worth mentioning that the strains utilized in this work were obtained from the same soil area that was used in the bioassays. The approach of selecting native strains has been implemented in other works, such as ^{Koskey et al. (2017)} which used indigenous strains with the ability to fix nitrogen in low fertility soils in Kenya in bean crops, obtaining an increase in grain yield of 30 % compared to the commercial inoculum for beans, Biofix (unreported strains).

In summary, sorghum growth has been scarcely studied in soils with nutritional stress; thus, the results of this work show that strains of Pseudomonas genus, with the activity of IAA production, is sufficient to improve the sorghum growth in a low-nutrient soil. This fact is attractive because it shows an alternative for the improvement of the sorghum crop by the utilization of PGPB like Pseudomonas. In addition, it also offers the possibility that Pseudomonas sp. BI-1 and P. mohnii BIPS-10, both producing IAA and isolated from the crop soil, could be considered as potential biofertilizers for sorghum.

Supplementary material

Supplemental data for this article can be accessed here: https://doi.org/10.17129/botsci.2841.

Supplementary material, Figure S1, Figure S2, Table S1.