Four Pseudomonas fluorescens strains (B10, B2-10, B2-11 and B4-6) isolated from the rhizosphere of barley plants in Bushehr province, Iran, were obtained from a culture collection of Biological Control Lab., Persian Gulf University, Iran. Bacteria were routinely cultivated at 27°C on Kingʼs B medium agar (KB) (peptone 20 g, K ₂HPO ₄ 1.5 g, MgSO ₄-7 H ₂O 1.5 g, glycerol 20 g, and agar 15 g, pH 7.0) (King et al., 1954) and Luria-Bertani broth (LB; Difco, Detroit, MI, USA). Glycerol stock (20% w/v) of bacterial strains was prepared and stored at -80°C until further use. Seeds of barley ( Hordeum vulgare L.) cvs. 'Desert', 'Mid-day', 'Zehak', 'South' and 'Karun', obtained from the Seed and Plant Impro­vement Institute (Karaj, Iran), were surface-disinfected in ethanol 70% (v/v), followed by 10 min in 10% (v/v) H ₂O ₂, rinsed thoroughly with sterile double-distilled water, and pre-germinated for three days on 0.85% water agar at 25°C in darkness.

The ability of bacterial strains to utilize ACC was evaluated based on Dell'Amico et al. (2005) with slight modifications. Bacteria were grown on tryptic soy broth medium (TSB) (TSB, Merck) for 48 hours. Thereafter, 50 µL of bacterial suspension was transferred to 20 mL of DF (Dworkin and Faster) minimal medium (Dworkin & Faster, 1958) containing 3 mM L ⁻¹ ACC as the selective medium, DF containing 2 g L ⁻¹ (NH4) ₂SO ₄ as positive control and DF minimal medium with no amendments (as negative control). After 48 h incubation at 27 ºC in shaker incubator with 120 rpm, an optical density at 405 nm was determined using spectrophotometer for each suspension. The ability of a strain to utilize ACC was verified by comparing the optical density (OD) of each bacterial suspension with that of negative control. The absence of growth in negative control confirmed the utilization of ACC as N source.

Salt tolerance test was conducted based on Qin et al. (2014) with slight modifications. Bacteria were grown on modified mineral-based nutrient agar (CaSO ₄ 0.1 g, K ₂HPO ₄ 0.2 g, peptone 1 g, MgSO ₄·7H ₂O 0.2 g, agar 15 g in 1000 mL distilled water) amended with increasing concentrations of NaCl (0–10%, w/v) at intervals of 1% at 27 °C.

Isolates were screened for siderophore and IAA production and solubilization of phosphate based on Islam et al. (2009). Siderophores were detected on Chrome Azural S (CAS) agar plates by the formation of orange halos around bacterial colonies after incubation for 24 h at 27 °C. P-solubilization activity was tested on Pikovaskaya's agar medium containing 2.5% tricalcium phosphate. The formation of transparent halo zone around bacterial colonies after incubation for 24-48 h at 28 °C was observed. Ability to produce IAA was detected using the colorimetric method of Bric et al. (1991). The IAA secreted in the culture was determined using a calibration curve of pure IAA (Sigma-Aldrich, St. Louis, MO, USA) as a standard following the regression analysis. IAA production assay was conducted in a completely randomized design with three replications. The experimental data were analyzed statistically using SAS program and means were separated by Duncan's Multiple Range Test (DMRT).

The impacts of four effective Pseudomonas strains were evaluated on barley yield and growth under greenhouse conditions. Bacterial strains were grown in LB broth overnight. Cultures were washed with sterile 0.9% NaCl solution and adjusted to an OD at 600 nm of 0.125 that corresponds to cell density of about 10 ⁸ colony forming units (CFU) mL ⁻¹. Before performing greenhouse experiment, root tip colonization of bacterial strains on barley cultivars under salinity (50, 100 and 150 mM NaCl) was assessed under gnotobiotic conditions as described by Egamberdieva (2011) with some modifications. It was revealed that our bacterial strains were capable of colonizing the rhizosphere of barley cultivars at the highest saline conditions (150 mM) (unpublished data).

For greenhouse experiment, after soaking pre-germinated barley seeds in the bacterial suspensions for 1 h, 25 seeds were transferred to plastic pots filled with 8 kg of sterilized soil. Soil sample for the study was collected from the experimental field at Persian Gulf University, Bushehr, Iran, air-dried, sieved (2 mm, 10 mesh) and analyzed for physico-chemical characteristics. The soil was clay loam having pH 6.2, organic matter 1.15%, total nitrogen 0.1%, total P 18 mg kg ⁻¹, and electrical conductivity 5.45 mS m ⁻¹.

Plants were watered when needed after emergence, and no fertilizers were used. Salinity conditions at four-leaf stage were established by adding 50, 100 and 150 mM NaCl into the irrigation water (sterile distilled water). Electrical conductivity (EC) of these solutions was 5 dS m ⁻¹ for 50 mM NaCl, 10 dS m ⁻¹ for 100 mM NaCl and 15 dS m ⁻¹ for 150 mM NaCl. Sterile distilled water with no addition of NaCl was considered as non-stressed controls. Plants were grown in the greenhouse at 25ºC and at a 16/8 day/night regime. After 15 days, thinning was done to leave 15 uniform seedlings in each pot. Morphological parameters like plant height, peduncle length, spike length, spike weight, number of grains per spike, 1000-grain weight (MGW) and grain yield were measured at physiological maturity stage.

An experiment was conducted to test the performance of the hybrid PSO, GA and ANN approach. Six factors and three levels for each one were selected following Taguchi approach, as presented in Table 1.

The effect of salinity levels on the survival of inoculated strains in the rhizosphere of barley cultivars was determined under greenhouse conditions which were described above.

At harvest, plants were removed from the pots and gently shaken to discard loosely adhering soil. Roots with tightly adhering soil were weighted, placed in 10 mL sterile saline solution (0.9%), and shaken for 20 min at 420 rpm. The number of bacteria in the resulting suspensions was determined as CFU by plating serial dilutions on KB agar plates.

A pot experiment was performed in a split-split plot design based on randomized complete block with three replicate pots per treatment. Main plots, subplots and sub-sub plots were consisted of salinity (at three levels of 50, 100 and 150 mM), five varieties ('Karun', 'Zehak', 'Mid-day', 'Desert' and 'South'') and four bacterial strains, respectively. In the greenhouse experiment, controls were consisted of plants either not watered with NaCl solutions or not inoculated with bacterial strains. Each experimental unit was a pot consisted of 15 plants. Data were subjected to statistical analysis using SAS 9.1 (SAS Inst., Inc., Cary, NY, USA). Means were subsequently compared using the DMRT.

All selected strains from the rhizosphere of barley were able to grow on DF minimal medium amended with ACC, indicating that they have the ACC deaminase activities. Optical densities of Pseudomonas sus­pensions in minimal medium without nitrogen source (negative control), amended with 2 g of (NH ₄) ₂SO ₄ L ⁻¹ (positive control), or with ACC (3 mM L ⁻¹) are listed in Table 1.

The bacterial strains were screened for their plant growth-promoting (PGP) traits and the results are summarized in Table 2. All the four isolates produced IAA and siderophores and also had phosphate solubilizing activity. The NaCl tolerance of bacterial strains was tested, and the results revealed that all strains could tolerate 8% of NaCl stress after which no growth was observed. Results also showed that as the salt concentration increased, the bacterial growth decreased. Since the four strains showed high tolerance to 8% NaCl, they were selected to study their effects on barley yield and growth under saline stress.

Data analysis of variance showed that the main effects of salinity, variety and bacterial strains and the interaction between them were significant on most measured traits ( p≤0.05) (Table 3). Generally, as salinity increased, all measured traits decreased; however, inoculation of seeds with bacterial strains ameliorated salinity stress and resulted in an increase in barley growth parameters and yield comparing to non-inoculated control.

Under normal condition and salinity of 50 mM, isolate B2-10 (with 10% increase), at 100 mM NaCl stress, all bacterial strains and at 150 mM salinity B4-6 and B2-10 elevated plant height in comparison with other treatments (Table 4). Data regarding peduncle length and spike length showed that treatment with B2-10 strain significantly promoted these traits by 26-46% and 26-30%, respectively, under normal condition and salt stress (Table 4). Similarly, inoculation with B2-10 strain caused the higher number of grains per spike (with 12-14% increase) under all salinity levels. Under normal condition and at 50 and 150 mM, B2-10 strain increased spike number (20, 19 and 12%, respectively) significantly over untreated control; however, at 100 mM, B10 and B4-6 strains performed better (Table 4). B2-10 and B4-6 were the most effective isolates which enhanced MGW up to about 14%. The maximum spike weight was obtained under normal condition following inoculation of seeds with B2-10 strain comparing to other treatments. Under normal condition and all salinity levels, B2-10 strain had the maximum impact on increasing grain yield, with no significant difference to B2-11 under normal condition and 100 mM salinity (Table 4).

Barley varieties showed variations in plant height during NaCl induced salinity stress (Table 5). Under all conditions, 'Karun' showed the maximum plant height comparing to other varieties. 'South' showed the maximum spike length and peduncle length under all salinity levels. 'Desert' had the highest spike number at all salinity levels which was not significantly different to 'Mid-day' at 50 mM. 'Mid-day' had the highest number of grains per spike at all salinity levels comparing to other varieties. In all salinity levels, 'Zehak' had the highest MGW, which was not significantly different to 'Karun' and 'Mid-day' varieties at 50 and 100 mM salinity. Considering spike number, under normal condition, 'Karun' had the highest spike number which was not significantly different to 'Desert'. Under salinity levels of 50, 100 and 150 mM, 'Desert' showed the maximum number of spikes, which was not significantly different to 'Mid-day'. Considering grain yield, under normal condition and salinity of 50 and 150 mM, 'Desert' had the highest grain yield. Under salinity level of 150 mM, 'Mid-day' showed the maximum grain yield, which was not significantly different to 'Desert' and 'Karun' varieties (Table 5).

Inoculation of barley varieties with tested PGPR strains caused an increase in different measured traits comparing to uninoculated control; however, varieties did not show identical response to inoculation with different strains (Table 6). Inoculation of 'Karun' with B2-11 resulted in the maximum plant height as compared to other treatments. Seed inoculation of 'Zehak' with B2-10 (with 46% increase), and 'South' with B4-6 and B2-10 (with 27 and 25.6% increase, respectively) increased peduncle length more significantly than the other treatments. The maximum spike length was observed following the inoculation of 'South' with B2-10, which was not significantly different to the treatment of 'Desert' with B4-6 strain. Inoculation of 'Mid-day' and 'Desert' with B2-10 resulted in the highest spike number as compared to other treatment. The highest increase in the number of grain per spike was observed in 'Mid-day' following seed inoculation with B2-11 strain. Inoculation of 'Zehak' with B10 caused the maximum MGW, which was significantly different to all treatments. The maximum increase in spike weight was seen in 'Mid-day' and 'Karun' after seed inoculation with B2-10 strain. Inoculation of 'Karun' with B2-10, B4-6 and B10 strains, 'Zehak', 'Mid-day' and 'South' with B2-10 and B2-11 and 'Desert' with B2-10, B2-11 and B4-6 caused the highest increase in grain yield of barley cultivars as compared to other treatments (Table 6).

The survival of bacterial strains in the rhizosphere of five barley cultivars grown in different salinity levels was determined under greenhouse conditions. Analysis of variance revealed that the main effects of salinity, variety and bacterial strains and the interaction between them were significant on rhizosphere colonization ( p≤0.05) (Table 3). All four bacterial strains were able to colonize and survive in the rhizosphere of barley cultivars. However, their colonization level was partly inhibited under salt stress. Under normal condition isolate B2-10, at 50 mM, isolates B2-10 and B2-11, at 100 mM NaCl stress, isolate B2-11 and at 150 mM salinity B4-6, B2-11 and B2-10 had higher rhizosphere colonization in comparison to other treatments (Table 4). Barley varieties showed variations in rhizosphere colonization by bacterial strains during NaCl induced salinity stress (Table 5). Under normal conditions, 'Mid-day' at 50 mM, 'South' at 100 mM and 150 mM all barley varieties, except 'Zehak', showed higher rhizosphere colonization. Barley genotype at the cultivar level had significant impact on rhizosphere colonization by bacterial strains under greenhouse conditions (Table 6). In 'Karun' the rhizosphere colonization by B2-10, and in 'Mid-day' the colonization by B2-11 were higher than by other strains. In 'Zehak', bacterial strains were not significantly different regarding rhizosphere colonization. In 'Desert' and 'South', B2-11 strain colonized the rhizosphere better than the others; however, the colonization level was not significantly different to B2-10 and B4-6 (in 'Desert') and B2-10 (in 'South').