Phosphate solubilizing rhizobacteria isolated from jujube ziziphus lotus plant stimulate wheat germination rate and seedlings growth

Sampling and bacterial isolation

The sampling site, located in the experimental farm of Mohammed VI Polytechnic University (UM6P), Benguerir-Morocco (32.219731E, −7.892268N), is characterized by a temperate continental monsoon climate. The annual rainfall is 290, six mm and mainly occurs from October to January. Eight samples of roots and rhizospheric soil of jujube were collected from a 5 to 25 cm depth, packed and labeled in sterile plastic bags, then transported immediately to the laboratory in cool boxes. Fractions from all soil samples were mixed and analyzed for some physicochemical properties at the Agricultural Innovation and Technology Transfer Center (AITTC) of UM6P. The results are shown in Table 1. Rhizospheric samples were serially diluted under aseptic conditions by suspending 1 g of rhizosphere soil in 9 mL of sterile deionized water. Next, 100 µL of serial dilutions were subsequently plated on Tryptic Soy Agar medium (TSA) (EMD Millipore, Berlin, Germany) and incubated at 28 ± 2 °C till the appearance of bacterial colonies. Screening of individual colonies was carried out by repeated streaking.

Screening of phosphate solubilizing bacteria

All bacterial isolates were qualitatively screened for inorganic P solubilization by inoculating a single colony of each strain in National Botanical Research Institute’s Phosphate growth medium (NBRIP) containing 10 g L⁻¹ glucose; 0.1 g L⁻¹ (NH4)₂ SO₄; 5 g L⁻¹ MgCl₂ 6H₂O; 0.2 g L⁻¹ KCl, 0.25 g L⁻¹ MgSO₄,7H₂O and finally 5 g L⁻¹ Ca₃(PO₄)₂ (TCP: insoluble tricalcium phosphate) as a sole source of phosphate (Nautiyal, 1999). The initial media pH was adjusted to 7.00 before use. Each bacterium was incubated on NBRIP plate at 30 °C for 7 days and only colonies surrounded by clear halos were selected for further studies as potential P solubilizer candidates. PSB were subsequently sub-cultured in TSB (Tryptic Soy Broth) (Professional lab, Casablanca, Morocco) liquid media and cryopreserved at −80 °C until use. 

Quantification of phosphate solubilization by bacteria

Inorganic P-solubilizing activity was quantified using TCP (in a modified NBRIP liquid medium). Briefly, bacterial suspension (0.1 mL of OD_600nm= 0,8) was inoculated in a 100 mL flask containing 50 mL of NBRIP broth in triplicate. Non-inoculated medium was used as blank, while Rhizobium tropici CIAT 899 served as a positive control. Bacterial cultures were incubated at 28 ± 2 °C during five days under shaking condition at 150 rpm. The cultures were then harvested by centrifugation at 13.000 rpm for 10 min and the soluble P, contained in the supernatant, was quantified by colorimetric method using SKALAR (SKALAR SAN++ SYSTEM). Dissolved P concentration was determined by subtracting the P concentration of the blank from the final concentration of soluble P in the inoculated broths. The final pH of each culture supernatant was also measured. The experiments were performed in triplicate and the results are means of the replicates. 

Bacterial antibiotic resistance and heavy metal tolerance

Antibiotic resistance profile of selected PSB was determined using TSA medium supplemented with selected antibiotics namely kanamycin (50 µg mL⁻¹), streptomycin (100 µg mL⁻¹), tetracycline (10 µg mL⁻¹), ampicillin (100 µg mL⁻¹), chloramphenicol (20 µg mL⁻¹) and spectinomycin (60 µg mL⁻¹).

Heavy metal tolerance of selected isolates was tested using the same method (TSA plates) with the addition of increasing concentrations (ranging from 0 to 1500 µg mL⁻¹) of three heavy metals; cadmium sulfate (CdSO₄), copper sulfate (CuSO₄.5H₂O) and nickel nitrate (N₂NiO₆). The plates were incubated at 30 °C for 24 h.

Strains genotyping using 16S rRNA gene sequencing

PSB identification was performed using 16S rRNA gene sequencing. The polymerase chain reactions (PCR) were carried out directly with fresh bacterial suspension, using a pair of universal primers pA (5′- AGAGTTTGATCCTGGCTCAG-3′) and 926R_Quince (degenerated one) (5′-CCG YCAATTYMTTTRAGTTT-3′), and MyTaq Mix, 2X (ThermoFisher, Casablanca, Morocco) containing Taq DNA polymerase, dNTP, MgCl₂ and buffer. Amplification of 16S rDNA sequences was made in 50 µl reaction mixture containing 25 µL of MyTaq mix, 1 µL of each primer (20 µM), 22 µL of DNase/RNase-free distilled water and 1 µL of overnight bacterial culture as DNA template. The reaction was performed in a VWR^® thermal cycler using the following PCR optimized conditions: initial denaturation at 95 °C for 5 min, 35 cycles of denaturation at 95 °C for 30 s, annealing at 52 °C for 30 s, elongation at 72 °C for 1 min, and final elongation at 72 °C during 10 min. The amplified 16S rDNA fragments (910-bp) were sequenced by Genome Quebec, Canada. The generated DNA sequences were aligned to available standard sequences of bacterial lineage in the National Center for Biotechnology Information GenBank database (http://www.ncbi.nlm.nih.gov/) and the High-Quality Ribosomal RNA databases SILVA (https://www.arb-silva.de) using BLAST algorithm to carry out a taxonomic assignment of each isolate. The phylogenetic tree of identified PSB was built using Ugene software.

Indole-3- acetic acid (IAA) measurement assay

Bacteria were analyzed for the quantitative determination of indole acetic acid (IAA) production. For this purpose, 100 µL of each PSB strain (OD_600nm = 0.8) was grown in 50 mL Tryptic Soy Broth (TSB) supplemented with 0.1% L-tryptophan as IAA precursor at 28 ± 2 ° C in a shaking incubator at 200 rpm. After 7 days, two mL of Salkowski reagent [0.5M FeCl₃: 70% perchloric acid/water (2:49:49 ratio)] (Glickmann & Dessaux, 1995), was pipetted into test tubes containing one mL of culture supernatant filtrates. The tubes containing the mixture were gently vortexed and left for 30 min in dark for the development of color at room temperature (26 ± 2 °C). The absorbance was determined at an OD_535nm. The quantity of IAA produced in each supernatant was estimated in (µg mL⁻¹) from a calibration curve using a standard IAA (Sigma Aldrich, Overijse, Belgium).

Siderophores production assay

Qualitative production of siderophores by selected strains was detected on the chrome-azurol S (CAS) medium as previously described (Schwyn & Neilands, 1987). Briefly, each bacterial culture was spot-inoculated separately on CAS agar plates. The plates were kept at 30 °C for 3 days. After the incubation period, the appearance of orange halo (blue to yellow/orange) around the colony was considered as a positive result for siderophores production.

Extracellular enzymes production assay

Bacteria were qualitatively analyzed for the production of protease and cellulase by the plate method (Kavitha, Nelson & Jesi, 2013). Protease activity (casein degradation) was tested by inoculation of selected strains into nutrient agar medium containing casein 5 g L⁻¹, yeast extract 2.5 g L⁻¹, glucose 1 g L⁻¹, and agar 15 g L⁻¹ and amended with 10% of skim milk. After 48 h incubation at 30 °C, a clear zone around colonies indicated positive proteolytic activity. For cellulase activity, a mineral–salt agar plate containing 0.4% (NH₄)₂SO₄, 0.6% NaCl, 0.1% K₂HPO₄, 0.01% MgSO₄, 0.01% CaCl₂ with 0.5% carboxymethyl cellulose, and 2% agar were surface-inoculated with each strain and incubated 48 h at 30 °C. Plates were stained with 0.1% Congo Red (Sigma Aldrich, Casablanca, Morocco) for 15 min. Following de-staining during 15 min, using 1 M NaCl, the development of the halo zone around the colonies reflects cellulase production.

Ammonia production assay

Bacteria strains were tested, qualitatively and quantitatively, for ammonia production in peptone water as previously described (Cappuccino & Sherman, 1992). Briefly, freshly grown cultures were inoculated into 10 mL peptone water and incubated for 48 h at 30 °C on a shaker (150 rpm). Post incubation period, 0.5 mL of Nessler’s reagent was added to each tube. Ammonia production is proportional to the brown color intensity. It was measured spectrophotometrically at OD_450nm using the VICTOR NivoTM Multimode Plate Reader (PerkinElmer, Casablanca, Morocco) and determined using a standard curve prepared with 0.1–1 µmol mL⁻¹ ammonium sulfate.

Wheat seeds germination assay

Our selected strains were assessed for their effect on seed germination. Seeds of durum wheat (Variety vitron) were surface sterilized with 2% sodium hypochlorite solution for 1 min, rinsed thoroughly with sterile distilled water, soaked in 70% ethanol for 1 min and washed 5 times in single distilled water followed by air-drying. PSB cell pellets were obtained by centrifuging an overnight culture (OD_600nm =0,8) at 10. 000 rpm for 5 min, the supernatant was removed, and the pellets were resuspended in 5 mL of sterile distilled water, vortexed and used for seed treatment. Fifteen sterilized seeds were treated with 5 mL of bacterial suspension for 30 min, air-dried, and then placed on sterile Petri dishes containing 0.7% agar medium and incubated at 25 °C. Triplicates were maintained for each treatment. Seeds were surface sterilized with 2% sodium hypochlorite solution for 1 min, rinsed thoroughly with sterile distilled water. Next, seeds were incubated in a dark incubator for 48 h, then left at room temperature in a day/night cycle. The germination rate was recorded after 24 h and 48 h. Root length, shoot length, fresh weight, and dry weight were measured after 7 days. The germination rate and vigor index were calculated using the following formula (Islam et al., 2016): (1)${\displaystyle Germinationrate\left(\text{\%}\right)=\frac{Numberofseedsgerminated}{\text{Total number of seeds}}\times 100\left(1\right)}$ (2)${\displaystyle Vigorindex=\left(\text{\%}\right)Germination\times Totalplantlenght.}$

Statistical analysis

Results presented here are the mean of triplicates (n = 3) ± Standard deviation. Statistical analysis was performed using IBM SPSS statistics 20 for windows. The differences between treatments were statistically analyzed using analysis of variance (ANOVA) and subsequently by Tukey’s multiple range test at p < 0.05.

Bacteria screening identified nine best phosphate solubilizing strains

The screening of P solubilizing bacteria from different rhizospheric soil samples of jujube on NBRIP led to the isolation of forty-one bacterial isolates. This microbial population has different aspects, but all exhibited a common character of tricalcium phosphate (Ca₃(PO₄)₂) solubilization on solid medium. Indeed, bacterial isolates were able to form a clear zone (halo) around their colonies on the NBRIP medium, indicating positive solubilization of P from tricalcium phosphate (TCP). Nine isolates were selected as being the best performers on plates and named J10 to J13, J15, and J153 to J156 (J for Jujube). Next, we tested their ability to solubilize inorganic phosphorus (Ca₃(PO₄)₂) in NBRIP liquid medium. The amount of soluble P and growth media’s pH were measured 5 days post incubation. Eight strains were found to release P from TCP with concentrations ranging from 20.5 mg L⁻¹ to 264 mg L⁻¹ (Fig. 1). Remarkably, the highest solubilization was recorded for strain J153, while the lowest one (20.5 mg L⁻¹) was measured for strain J11. The amount of P solubilization by the referenced strain, Rhizobium tropici did not exceed 67.5 mg L⁻¹. As expected, P solubilization was accompanied by a significant drop in pH, of the culture media, from 7.0 to 4.0 (Fig. 1).

Strains J10-13, J15 and J153-156 belong to the genera of Pseudomonas, Bacillus, and Paenibacillus

Characterization of the nine PBS strains to the genus level was performed by 16S ribosomal DNA gene partial sequencing. Generated sequences of 900-bp length were aligned to available 16S rDNA sequences using GenBank and SILVA databases. As summarized in Table 2, three strains (J12, J13, and J15) show 98% identity to the 16S rRNA gene sequences of Pseudomonas moraviensis; three strains J10, J153, and J154 share 98–99% identity to Pseudomonas sp.; two strains J11 and J156 exhibit 98 and 99% identity to Bacillus megaterium and Bacillus cereus, respectively. Lastly strain J155 shares 98% identity to Paenibacillus xylanexedens (Fig. 2).

Except for Pseudomonas sp. J11 and B. cereus J156, remaining strains displayed resistance at least to one antibiotic

PGPB tend to harbor genes that confer resistance to antibiotics (Kang et al., 2017). To assess bacterial resistance to antibiotics, we checked our strains for growth on plates supplemented with a set of different antibiotics, frequently encountered among bacteria isolated from soils. As reported in Table 3, out of the nine tested strains, seven presented resistance at least to one antibiotic. Strains P. moraviensis J12 & J15, Pseudomonas sp. J153 & J154 resist to chloramphenicol and ampicillin, while Pseudomonas sp. J10 & J13 confer resistance to chloramphenicol, ampicillin, and spectinomycin. Strains Pseudomonas sp. J11 and Bacillus cereus J156 are sensitive to all tested antibiotics. Strain P. xylanexedens J155 is resistant to both kanamycin and spectinomycin. Lastly, none of the tested strains are resistant to neither streptomycin nor tetracycline (Table 3). In the next steps, to avoid any potential contamination, we took advantage to these resistances to grow bacteria on selective media.

Pseudomonas sp. J153 and B. cereus J156 withstand high concentrations of copper sulfate/cadmium sulfate and copper sulfate/nickel nitrate, respectively

Heavy metals such as lead (Pb), cadmium (Cd), chromium (Cr), and mercury (Hg) are exceptionally toxic and dangerous environmental pollutants (Tangahu et al., 2011). We investigated the capacity of our strains to grow under various concentrations of nickel nitrate (N₂NiO₆), copper sulfate pentahydrate (CuSO₄.5H₂O) and cadmium sulfate (CdSO₄). The nine strains exhibited various tolerance characteristics (Table 4). In the copper assay, Pseudomonas sp. J153 grows up to 1,500 µg mL⁻¹, Bacillus cereus J156 to 1,000 µg mL⁻¹, while strains Pseudomonas sp. J10, P. moraviensis J12, P. moraviensis J13, and P. moraviensis J15 supported a maximum of 500 µg mL⁻¹. The lowest tolerated concentration, 300 and 200 µg mL⁻¹ were seen in B. megaterium J11 and Pseudomonas sp. J154 strains, respectively. When tested for cadmium sulfate, only Pseudomonas sp. J153 grows up to 1500 µg mL⁻¹, whereas 300 µg mL⁻¹ was the maximal concentration tolerated by strains Pseudomonas sp. J10, P. moraviensis J12, P. moraviensis J13, P. moraviensis J15, Pseudomonas sp. J154 and B. cereus J156. Lastly, low tolerance at 100 µg mL⁻¹ and 10 µg mL⁻¹ were detected in strains P. xylanexedens J155 and B. megaterium J11, respectively (Table 4). In the nickel nitrate assay, Pseudomonas sp. J10 and Bacillus cereus J156 strains, grow up to 1,500 µg mL⁻¹ and strain P. xylanexedens J155 tolerated the lowest concentration of 300 µg mL⁻¹. However, the remaining strains tolerate growth up to 500 µg mL⁻¹. Taking together, our data highlighted the remarkable capacity of strains Pseudomonas sp. J153 and B. cereus J156 to withstand abnormal high concentrations of both copper sulfate/cadmium sulfate and copper sulfate/nickel nitrate, respectively.

B. megaterium J11 is the best indole-3-acetic acid producer

The production of indole acetic acid (IAA) is a major property shared by numerous rhizospheric bacteria that stimulate plant growth (Mohite, 2013). Seven days post incubation, all tested strains produced IAA although at various levels, ranging from 57.1 to 204.28 µg mL⁻¹ (Fig. 3). The highest concentration was produced by B. megaterium J11, whereas the lowest one was measured for P. xylanexedens J155 (Fig. 3).

P. moraviensis J13 is the best siderophores producer

Siderophores are best known for binding iron (Fe) and for mobilizing soil-immobilized Fe, although they may also contribute to improve the availability of P to plants (Sharma et al., 2013), by solubilizing minerals and chelating heavy metals, which in turn increases nutrient uptake and plant growth (Gontia-Mishra et al., 2016). The ability, in vitro, of selected PSB, to produce siderophores was qualitatively estimated using the CAS-agar plate assay. All tested strains were able to produce siderophores, although at various levels as deduced by the size of the halo zone and the intensity of the color change of the CAS-Agar (Table 3). P. moraviensis J13 was the most efficient siderophores producer, the six strains (Pseudomonas sp. J10, J153 & J154, P. moraviensis J12 & J15, and P. xylanexedens J155 produced intermediate level, whereas the lowest production was seen in both B. megaterium J11 and B. cereus J156 strains.

Pseudomonas sp. J153 is the best ammonia producer

Ammonia is a chemical compound having indirect plant health benefits, primarily by acting as metabolic inhibitor against phytopathogens (Kumar et al., 2012). All tested strains were able to produce ammonia with various concentrations. The highest value of 0.33 µmol mL⁻¹ was detected in Pseudomonas sp. J153, while the lowest one, 0.1 µmol mL⁻¹, was measured in P. xylanexedens J155 (Fig. 4).

Proteases are not produced by Pseudomonas sp. J153 and J154 and cellulase activity is restricted to Pseudomonas sp. J10, J155 and B. cereus J156

Bacterial extracellular enzymes such as proteases and cellulases play a dual important role in the biological control of phytopathogens and in soil fertilization (Mitchell & Alexander, 1963). The nine strains were tested for their ability to produce proteases and cellulases. Results of both proteases and cellulase assay are shown in Table 3. As for proteases production, except for Pseudomonas sp. J153 & J154, the remaining seven strains developed halo zone around the colonies. As a control, no halo zone was seen using E. coli strain DH5 α, used here as a negative control. Cellulase activity was solely detected in three strains: Pseudomonas sp. J10, P. xylanexedens J155, and B. cereus J156, each of which formed a yellow/whitish zone around their colonies and were considered as cellulase positive. No cellulase activity was observed in the remaining six other strains.

Inoculation with P. moraviensis J12 and B. cereus J156 promote the highest rate of wheat seeds germination and seedlings growth

The treatment of wheat seeds by the nine PSB strains had a significant effect (P < 0.05) on the germination rate and wheat vigor index, as compared to the control (Figs. 5A et 5E). However, these effects varied depending on the PSB isolates. For instance, both P. moraviensis J12 and B. cereus J156 strains were the most efficient in promoting wheat germination as represented by vigor index (Fig. 5E). Results revealed that, compared to non-inoculated control, seeds inoculated by each of the nine strains showed a considerable impact on different growth parameters (Fig. 6).

Regarding shoot and root length after 7 days of growth, seeds inoculated with all strains, especially B. megaterium J11 significantly enhanced shoot and root length (p < 0.05). Maximum root length was seen upon inoculation with P. moraviensis J12 (Fig. 5B). We also noticed, whatever the nature of the inoculum was, seeds root dry weights remain unchanged. In contrast, a significant increase in shoot dry weight was detected (Fig. 5D). Furthermore, wheat seeds inoculation significantly affected shoot fresh weight, but not root dry weight except for B. megaterium J11, P. moraviensis J12, and J15 strains (Fig. 5C).