In-Vitro Plant Growth Promotion of Rhizobium Strains Isolated from Lentil Root Nodules under Abiotic Stresses

Plant growth-promoting rhizobia are known to improve crop performance by multiple mechanisms. However, the interaction between host plants and Rhizobium strains is highly influenced by growing conditions, e.g., heat, cold, drought, soil salinity, nutrient scarcity, etc. The present study was undertaken to assess the use of Rhizobium as plant growth promoters under abiotic stress conditions. Fifteen Rhizobium strains isolated from lentil root nodules were tested for phosphate solubilization activity (PSA) and phytohormones production under salt and drought conditions. The results showed that 15 Rhizobium strains were significant phosphate solubilizers, and indole acedic acid (IAA) and gibberellic acid (GA3) producers based on least significant difference (LSD) analysis (p ≤ 0.05). The highest rate of PSA was attributed to three strains namely, 1145N5, 1159N11, and 1159N32 with a range of 144.6 to 205.6 P2O5 (µg/mL). The highest IAA production was recorded in the strain 686N5 with 57.68 ± 4.25 µg/mL as compared to 50.8667 ± 1.41 µg/mL and 37.32 ± 12.59 µg/mL for Rhizobium tropici CIAT 899 and Azospirillum brasilense DSM-1690, respectively. Strain 318N2111 produced 329.24 ± 7.84 µg/mL of GA3 as against 259.84 ± 25.55 µg/mL for A. brasilense DSM-1690. R. tropici CIAT 899 showed tolerance to salt (5% NaCl) and drought (ψ = −2.6 MPa) stress, whereas strain 686N5 showed an extremely high level of salt-tolerance (5% NaCl) and moderate level of drought tolerance (ψ = −0.75 MPa). These results indicate different pathways for drought and salt tolerance mechanisms. The assessment of plant growth promoting (PGP) activities of Rhizobium showed differences between bacterial viability and bacterial PGP activity in terms of abiotic stress tolerance where bacterial PGP activity is interrupted before reaching the bacterial tolerance threshold. These results integrate a new concept of PGPR screening based on PGP activity under abiotic stress.


Introduction
In the arid and semiarid regions, crops are seriously affected by drought and salinity which affect negatively plant growth. These stresses reduce crop yields by disrupting the biochemical, physiological, ICARDA genebank and selected based on the PCR-RFLP (Polymerase Chain Reaction-Restriction Fragment Length Polymorphism) analysis of 16S rRNA nodD genes diversity analysis (prepared for another publications). Rhizobium tropici CIAT 899 and Azospirillum brasilense DSM-1690, obtained from the German collection of microorganisms (Leibniz-Institut DSMZ-Deutsche Sammlung von Mikroorganismen und Zellkulturen Gmbh, Inhoffenstraße 7B38124 Braunschweig Germany) were used as checks for their respective high abiotic stress tolerance and plant growth promotion [21,22]. The bacterial growth study was carried out following the method described by Harley and Prescott [23] using R. tropici CIAT 899 as a reference strain. The overnight rhizobial suspensions were grown at 28 • C and 150 rpm in Yeast-Extract Mannitol Broth (YMAB) medium [24]. The bacterial suspensions were adjusted to the same final concentration of 10 8 CFU/mL which corresponds to a cell density of OD (600) ≈ 1.0 using the spectrophotometer and inoculated in the modified liquid medium BIII (g/L) at 1% (V/V) [25].

Qualitative Test
Strains were tested for their ability to solubilize phosphate by using modified Pikovskaya's agar medium [26] containing the Hydroxyapatite as a source of insoluble phosphate. Pure cultures of strains were inoculated on the modified Pikovskaya medium and incubated at 30 • C for 7-10 days. Phosphate solubilization efficiency was calculated using the formula: where SE: solubilization efficiency, Z: solubilization zone (mm), and C: colony diameter (mm).

Quantitative Test
The measurement of phosphate solubilization potential of these strains was carried out using the method described by Kothamasi et al. [27] with some modifications. The one mL strains (cell ≈ 10 8 CFU/mL) that showed phosphate solubilization activity were grown in 100 mL Pikovskaya broth containing 1000 µg P/mL. The flasks were incubated at 28 • C at 150 rpm for 11 days. Then, five mL of each bacterial culture was retrieved on the 3rd, 5th, 7th, and 11th day, and filtered through a Whatman No. 1 paper to remove the undissolved phosphate and centrifuged at 10,000× g for 20 min for P 2 O 5 and pH determination. To the one mL of the supernatant, 2.5 mL of Barton's reagent [28] was added and the final volume was adjusted to 50 mL. After 10 min, the absorbance of the solution was read at 430 nm in a spectrophotometer, and phosphate solubilization was calculated by referring to a standard curve of K 2 HPO 4 expressed by µg/mL. Three replicates were performed for each strain and an uninoculated Pikovskaya broth served as control.

IAA (Indole Acetic Acid) Detection
Qualitative Test Each strain was spot inoculated by a sterilized toothpick in the middle of petri dishes containing YMA agar media amended with L-tryptophan (5 mM) and incubated at 28 • C for 11 days. The detection of indole acedic acid (IAA) production was carried out by soaking discs of Whatman paper in Salkowski's reagent (0.5 M FeCl 3 :70% perchloric acid/water (2:49 ratio)) and followed by the addition of a few drops of orthophosphoric acid over the bacterial colonies. Development of pink color indicates positive for IAA production. A. brasilense DSM-1690 was used as a positive control.
Quantitative Test IAA production was quantified using Gordon and Weber method [29]. One percent strains (cell ≈ 108 CFU/mL) inoculated in Yeast Extract-Mannitol Broth (YMA) + 5 mM Tryptophan at 28 • C for 3-4 days. Cultures were centrifuged at 10,000× g for 20 min. Two mL of the supernatant was mixed with 2 drops of phosphoric acid and 4 mL of Salkowski's reagent and incubated at room temperature for 25 min. The pink-auxin complex developed was read at 530 nm in spectrophotometer. The quantity of auxin in the cultures was estimated from a calibration curve using a standard IAA (Fluka) and values were expressed in µg/mL.

Gibberellic Acid Detection
Gibberellic acid was detected using Berríos et al. [30] method. Strains were inoculated to YMA broth and incubated at 28 • C for 48 h [31]. The bacterial suspension was centrifuged at 10,000× g for 10-15 min. Then, two mL of supernatant was added to 5N HCl solution (1:2). The acidified solution was extracted using ethyl acetate solution (1:3). To this, 2 mL of Potassium solution 1 M of zinc acetate solution was added. This mixture was centrifuged at 15,000 rpm for 15 min. To the 5 mL of supernatant, equal volume of 30% HCl was added and incubated at 20 • C for 2 min [32]. The absorbance was read at 254 nm [33]. Gibberellic acid was calculated by referring to a standard curve and activity was expressed in µg/mL.

Measurement of Siderophore Production
The siderophore production was estimated by the universal chemical test (chrome azurol S assay) as described by Schwyn and Neilands [34]. Glassware was prewashed by 6 M hydrochloric acid and then rinsed by distilled water. Overnight grown rhizobia cultures were spot inoculated onto a chrome azurol S (CAS) agar plate and incubated at 28 • C for 3-4 days. Colonies surrounded by yellow to light orange halo indicate the production of siderophore. The intensity of siderophore production was calculated by using the formula PI% = (Z − C)/C × 100 where, PI: production index, Z: production zone (mm), and C: colony diameter (mm). The quantitative test was performed 3 times with 3 replicates for each strain.

Statistical Analysis and Graphic Presentation
Two experiments were conducted, one to assess the effect of drought and the other to assess the effect of salinity. Each experiment consists of three replications (petri dishes or tubes) for each level of treatment. For each replicate, three readings were done for each variable. Data were arranged in excel files (Excel 2013: Microsoft, Redmond, WA, USA) and analyzed using SPSS 20 (IBM Corp., Chicago, IL, USA), to run the test of homogeneity of variances, one-way analysis of variance and comparison of means using Duncan test. The one-way analysis of variance was conducted to evaluate the null hypothesis that there is no difference in the amount of the bacterial growth and plant growth promotion activity (phytohormones production and inorganic phosphate solubilization) under salt and drought stress conditions. The bacterial strains were considered as independent variables.

Qualitative Test
The qualitative test of PGP traits showed diverse response among the studied strains. For instance, strains 686N5 and 996N5 were IAA producers, siderophore producers, and phosphate solubilizers whereas some of the strains, such as 115N2 and 1574N4 did not express any PGP traits (Table 1). Seven Rhizobium strains were siderophore producers with activity rate ranging between 1.18 cm and 2.07 cm ( Table 1).

Rhizobium Growth under Drought Stress
The results presented in Figure 5 showed significant osmotic stress effect on bacterial growth. Bacterial growth decreased significantly with increasing osmotic pressures. For instance, the growth of strain 996N2 did not exceed 0.17 ± 0.001 (OD) at the osmotic pressure ψ = −0.75 MPa ( Figure 5). Strain 1145N1 showed moderate osmotic-tolerance with a growth rate of 0.62 ±0.005 (OD) under ψ = −1.2 MPa. A high osmotic-tolerance was expressed by strain 1159N32 with a growth rate of 0.42 ± 0.04

Rhizobium Growth under Drought Stress
The results presented in Figure 5 showed significant osmotic stress effect on bacterial growth. Bacterial growth decreased significantly with increasing osmotic pressures. For instance, the growth of strain 996N2 did not exceed 0.17 ± 0.001 (OD) at the osmotic pressure ψ = −0.75 MPa ( Figure 5). Strain 1145N1 showed moderate osmotic-tolerance with a growth rate of 0.62 ± 0.005 (OD) under ψ = −1.2 MPa. A high osmotic-tolerance was expressed by strain 1159N32 with a growth rate of 0.42 ± 0.04 (OD) at ψ = −2.6 MPa, whereas R. tropici CIAT 899 showed an extremely high osmotic-tolerance with 0.31 ± 0.05 (OD) at ψ = −3.7 MPa ( Figure 5).

Rhizobium Growth under Salt Stress
The qualitative test of salt tolerance showed diverse response among the studied strains. For instance, strains 1159N11, 322N32, and 686N5 expressed high sensitivity towards salinity whereas an extremely high salt-tolerance ability was expressed by R. tropici CIAT 899, 686N5, and 318N2111 ( Figure 6). These results were confirmed by the quantitative salinity test where the Rhizobium strains

Rhizobium Growth under Salt Stress
The qualitative test of salt tolerance showed diverse response among the studied strains. For instance, strains 1159N11, 322N32, and 686N5 expressed high sensitivity towards salinity whereas an extremely high salt-tolerance ability was expressed by R. tropici CIAT 899, 686N5, and 318N2111 ( Figure 6). These results were confirmed by the quantitative salinity test where the Rhizobium strains were divided into four categories in terms of salt-tolerance. The highest salt-sensitive strains were 318N211, 1574N4, 996N2, 322N32, 996N5, 1159N11, and 1159N41 with an extremely low growth rate under 1% NaCl. Three strains 115N2, 1159N5, and 1145N1 showed moderate salt-tolerance under 3% of NaCl with the growth rate of 0.56 ± 0.011 OD, 0.67 ± 0.01 OD, and 0.54 ± 0.01 OD, respectively. Strain 1145N5 showed high salt-tolerance with the growth rate of 0.58 ± 0.011OD. Under 5% of NaCl, we identified extremely high halotolerant strains of R. tropici CIAT 899, 686N5, and 318N2111 with the growth rate of 0.30 ± 0.005 OD, 0.36 ± 0.005 OD, and 0.40 ± 0.005 OD, respectively ( Figure 6).

Phosphate Solubilization Activity under Salt Stress
Phosphate solubilization activity of the studied strains was maintained at a high level under 0.5%, 1%, 2%, and 3% of NaCl ranging from 136.5 ± 0.001 to 141.83 ± 1.15 P 2 O 5 µg/mL for R. tropici CIAT 899, from 132.83 ± 0.28 to 161.50 ± 0.5 P 2 O 5 µg/mL for strain 1145N5, and from 145.50 ± 0.86 to 159.0 ± 5.63 P 2 O 5 µg/mL for 1159N11 with no significant difference between them (Figure 9). Phosphate solubilization activity started declining at 4% and 5% NaCl with 82.66 ± 0.28 P 2 O 5 µg/mL for 1159N11 as the highest rate of P 2 O 5 and 28.50 ± 0.86 P 2 O 5 µg/mL for 1145N5 as the lowest rate of P 2 O 5 among the strains (Figure 9). In 0.5% NaCl, no notable decrease of pH in the studied strain cultures was observed except for R. tropici CIAT 899 with pH ≈ 3.8 ± 0.2 reached on the 11th day ( Figure 10). At 1% NaCl, pH started decreasing to reach 4.3 ± 0.2 on the 5th day, then started increasing to reach 5.03 ± 0.2 on the 11th day. However, the pH decreased notably to 3.0 ± 0.6 on the 3rd day and then started increasing to reach the pH ≈ 4.1 ± 0.2 in the 11th day of the experiment under 2%, 3% 4%, and 5% ( Figure 10).

Potential of Rhizobium as Plant Growth Promoters
In this study, 15 out of 16 Rhizobium strains isolated from lentil crop were phosphate solubilizing bacteria. The highest rate of phosphate solubilization activity between 144.6 and 205.6 P 2 O 5 µg/mL was recorded with three Rhizobium strains, namely 1145N5, 1159N11, and 1159N32. Phosphorus is the second most important element after Nitrogen [37]. Although agricultural soils might contain high phosphate content, much of this element is available under insoluble forms that plants cannot take advantage of [9]. In fact, only 0.1% [38] representing 0.01-3.0 P 2 O 5 mg/L of p is available which does not meet all the needs of a plant [37]. Thereby, the remaining soluble phosphate is acquired mainly through phosphate solubilization activity of microbes, including Rhizobia [39,40]. Past studies reported the potential of Rhizobium in terms of phosphate solubilization activity [39,41,42]. In the present study, lentil Rhizobium strains showed a high rate of phosphate solubilization activity compared to what was reported in previous studies [41,42]. Saghafi et al. [41] reported phosphate solubilization activity of two Rhizobium strains (Rlp281 and Sm29) with 128 and 155 P 2 O 5 µg/mL. Alikhani et al. [42] reported that the Iranian phosphate solubilizing Rhizobia released P 2 O 5 between 88.66 and 197.10 µg/mL whereas Bacillus sp. and Pseudomonas fluorescence, which were taken as positive controls, released on an average of 268.6 and 205.6 P 2 O 5 µg/mL, respectively. Further, the same study reported the same shape of the pH curve as found in the present study, where the pH value averaged at~4. Among the tested strains, some of them (for example: 1159N24) showed contradictory results with regard to qualitative (in plate assay) and the quantitative (in broth) tests of phosphate solubilization activity. Same results were reported where many strains do not show their PSA activity on plate while they can solubilize inorganic Phosphate in liquid medium [43,44]. This was explained by the nature of the used selective media [45][46][47][48].
Indole acetic acid (IAA) and gibberellic acid (GA3) were significantly produced by most of the Rhizobium strains compared to the control. The highest level of IAA was produced by strain 686N5 with 57.68 ± 4.25 µg/mL against 50.8667 ± 1.41 and 37.32 ± 12.59 µg/mL produced by R. tropici CIAT 899 and A. brasilense DSM-1690, respectively. Strain 318N2111 produced the highest level of GA3 (329.24 ± 7.84 µg/mL) as against 259.84 ± 25.55 µg/mL for A. brasilense DSM-1690 which was taken as a positive control. These results showed high performance of lentil Rhizobium strains in terms of phytohormones production. Saghafi et al. [49] reported two Rhizobia strains (R281 and R307) with an average of IAA production not exceeding 10.2 µg/mL. Nearly the same rate of IAA production (10.3 ± 1.5) was reported with Pseudomonas fluorescencens Ms-01 [50]. Bacillus cereus, considered one of the best PGP bacteria, showed maximum GA3 production of 205.58 µg/mL under the same conditions [51]. Phytohormones play an important role in the regulation of plant growth development as well as in abiotic stress tolerance [2,3]. IAA increases the plant root system by triggering the development of high number of root tips resulting in better uptake of water and nutrients [3,20,52] whereas GA3 is responsible for several physiological mechanisms such as stem elongation, seed germination and sex expression [39,53]. It is known that most of the bacteria colonizing the rhizosphere are phytohormone producers [54]. Furthermore, many studies have demonstrated the efficiency of the exogenous implementation of phytohormones through the inoculation of the plant with phytohormones producing Rhizobacteria [55][56][57].

Effect of Abiotic Stresses on Rhizobium Growth
In this investigation, seven Rhizobium strains (318N211, 1574N4, 996N2, 322N32, 996N5, 1159N11, and 1159N41) were salt-sensitive with an extremely low growth rate under 1% NaCl. Only three Rhizobium strains (115N2, 1159N52, and 1145N1) grew under 3% of NaCl. Only one strain 1145N5 was able to grow at 4% of NaCl and three Rhizobium strains (R. tropici CIAT 899, 686N5 and 318N2111) at 5% of NaCl. Rhizobia are considered very diverse when it comes to their response to salinity [33]. For example, most of Rhizobium leguminosarum strains are salt-sensitive and fail to grow at 2% of NaCl [13] whereas Rhizobium meliloti strains are categorized as salt-tolerant [58]. However, Benidire et al. [59] isolated a highly salt-tolerant (428 Mm ≈ 2.5% of NaCl) Rhizobium sp. strain (RHOF53) that related closely to Rhizobium leguminosarum. Interestingly, some Rhizobium leguminosarum strains isolated from Egyptian soil can grow at 7% NaCl medium [58].
In the present study, Rhizobium strains were also selected for their ability to tolerate drought stress. Nine out of 16 strains showed drought tolerance at ψ = −0.75 MPa osmotic potential whereas only two strains (1159N32 and R. tropici CIAT 899) showed high drought tolerance above ψ = −2.6 MPa. Sandhya et al. [60] considered ψ = −0.75 MPa as the threshold for the screening of drought-tolerant bacteria whereas Alikhani and Mohamadi [61] considered 0.4 turbidities (OD) the minimum bacterial growth for drought-sensitive bacteria. In order to avoid harsh conditions, soil bacteria employ diverse physiological mechanisms such as compatible solutes (proline, glycine betaine, trehalose, polyamines) accumulation [20] and exopolysaccharides production [62]. Bacterial drought tolerance is so closely related to exopolysaccharides production that their production is triggered by the increase of the water potential of the bacterial growth medium [63]. Moreover, bacterial exopolysaccharides play a significant role in increasing salt stress tolerance for both bacteria and plants. In fact, exopolysaccharides make water available and protect the bacterial environment from desiccation and cations Na+ [3,64,65]. This study showed that salt stress tolerance did not necessarily coincide with drought stress tolerance in the same strains. This is the case of the strain 686N5 which showed an extremely high salt-tolerance by growing under 5% NaCl and moderate drought stress tolerance under ψ = −0.75 MPa. Previously, Mohammad et al. [66] found Rhizobium meliloti accessions able to grow even at 616 mM NaCl and ψ = −1.0 MPa, explaining that salt stress and drought stress tolerance could involve different mechanisms. Indeed, many studies reported the expression and the repression of different genes when the bacteria were exposed to drought stress or salt stress [67][68][69][70]. Rüberg et al. [71] reported that many genes were induced by exposing Sinorhizobium meliloti only by osmotic upshifting. Jiang et al. [67] identified five salt-tolerance genes within Sinorhizobium fredii RT19 genome including phaD2, phaD2, phaF2, phaG2 which are mainly involved in the Na+ efflux.

Effect of Abiotic Stresses on Rhizobium PGP Activities
We also investigated the phosphate solubilizing and IAA production activities under salt and drought stress conditions. The studied strains were able to produce IAA and solubilize phosphate significantly under both stresses. Egamberdieva et al. [9] reported the same behavior regarding the IAA production under salt stress where two strains (Pseudomonas putida 1T1 and Strenoytophomonas rhizophila ep10) produced IAA under 1.5% NaCl. Kadmiri et al. [50] reported phosphate solubilization activity in two strains (Azospirillum brasilense DSM-1690 and Pseudomonas fluorescens Ms-01) which surprisingly increased under hypersaline conditions. Microbes play a major role when it comes to alleviating plant abiotic stresses and soil nutrient-deficiency [3,21,[70][71][72]. Auxin production brings balance to plant indigenous hormones caused by the harsh environments [4]. Bacterial phosphate solubilization is reported to impact positively on plant growth knowing that Phosphorus is the key to many plant-microbe interactions including nitrogen fixation [73].
Our study showed a relevant difference between the bacterial growth rate and their phosphate solubilization and IAA production activities under drought and salt stress conditions. PGP traits of Rhizobium strains decreased and/or stopped before the limit of bacterial growth tolerance. For example, R. tropici CIAT 899 was able to grow at 5% NaCl and ψ = −3.7 MPa, however, its phosphate solubilization activity decreased significantly and IAA production activity reduced drastically at 2% NaCl and ψ = −1.77 MPa. The similar results were reported by Egamberdieva et al. [9] where Pseudomonas putida 1T1 maintained its IAA production activity up to 1.5% NaCl while the strain was able to grow under 3% NaCl. This difference between bacterial viability and bacterial activity might be explained by the disruption of the community genetic regulation mechanism called Quorum Sensing (QS) that controls many bacterial functions including PGP traits [74]. The bacterial activity is mainly related to bacterial population density through the synthesis of QS signal molecules [74]. Once the threshold of the signal molecules is reached, the communication between the same population is triggered and the expression or the repression of PGP regulated genes occur. Thus, the decline or the interruption of the PGP activity of Rhizobium strains might be due to the decrease in the density of the bacterial population under drought as well as salt stress.

Conclusions
In this present study, we report the ability of Rhizobium to be used not only for legumes as nitrogen fixers but also as plant growth promoters. The challenge to use the rhizobacteria as bio-fertilizers for different crops reside in keeping them in plant growth promotion active status under harsh conditions. This study showed that the screening of inoculants based only on plant growth promotion and stress tolerance performance is not enough. Thus, we propose a new concept of screening based on PGP activity stress tolerance. Isolates have showed plant growth promoting potential under in-vitro conditions only. However, their use as PGPR in lentil and other crops needs to be tested by conducting pot experiments. Morphological (such as relative water loss, stomatal conductance, etc.), physiological (such as proline content, lipid peroxidase, etc.), and molecular abiotic stress markers of the plant (such as stress-induced genes, stress-related genes, etc.) should be considered to confirm the plant growth promoting activity of the studied strains.