Application of nitrogen-fixing purple non-sulfur bacteria in improving nitrogen uptake, growth, and yield of rice grown on extremely saline soil under greenhouse conditions

The experiment was conducted from September 2023 to December 2023 at the greenhouse of the Faculty of Crop Science, College of Agriculture, Can Tho University [10°01′43.2″N 105°45′58.9″E].

The experimental soil was collected from the topsoil layer of 0–20 cm of rice-shrimp cultivation land in My Xuyen district, Soc Trang province. In brief, the soil collection followed a diagonal cross with five locations in the field. The soil was let to dry naturally after the removal of residue and autoclaved for 50 min twice (24 h interval). Then, a soil mass (8 kg) was added to each pot. Next the soil was flooded for 2 days and mud was mixed before sowing. Each pot was sown with ten seeds, then eight uniform plants were selected 5 days after sowing (DAS). The experimental soil characteristics are shown in Table 1.

The diameters of mouth, bottom, and height of the plastic pots were 25, 21, and 21 cm, respectively.

The experimental rice variety was OM5451, high tillering, wide adaptability, and moderate salinity tolerance [39]. The variety originated from the Cuu Long Rice Research Institute, Vietnam.

The water source was clean water from tap.

The fertilizers used were Phu My urea (46% N), Long Thanh superphosphate (16% P₂O₅, 20% CaO), and Phu My potassium chloride fertilizer (60% K₂O).

Bacterial source: N-fixing bacterial strains, Rhodobacter sphaeroides S01 and S06, were selected from saline soil of the rice-shrimp model in Thanh Phu district, Ben Tre province [40], stored at the laboratory of the Faculty of Crop Science, College of Agriculture, Can Tho University.

The two-factor experiment was arranged in a completely randomized block design with 16 treatments and 4 replications (Factor A was the four recommended N-applying rates of 100, 75, 50, and 0%; Factor B was the N-fixing PNSB levels including no bacterial addition, single bacterial strain of S01, S06, and two mixed bacterial strains of S01 and S06) under greenhouse conditions. The experiments in seasons 1 and 2 shared the same design. In season 1, the temperature was 37.8°C, the humidity was 60.3%, and the light and dark hours per day were 11 and 13, while those in season 2 were 34.5°C, 65.1%, and 11.5 and 12.5 h, respectively.

The fertilization followed the recommended fertilizer formula for rice in the Mekong Delta: 100 N- 60 P₂O₅- 30 K₂O (kg ha⁻¹) [41] corresponding to 217.4 kg urea, 375 kg superphosphate, and 50 kg potassium chloride. P fertilizer was applied 100% before sowing. N fertilizer was used at the rate of 30, 40, and 30% at 10, 20, and 45 DAS, respectively, for each pot. K fertilizer was applied 50% at each time of 10 and 45 DAS.

Bacteria were prepared in advance and added to the soil at times of 7, 14, 21, 28, 35, and 42 DAS with a volume of 4 mL pot⁻¹ with a density of 1 × 10¹⁰ CFU mL⁻¹ (for mixed solution, each strain was used at 2 mL). Four milliliters of bacterial inoculant was added to 8 kg of soil evenly in the treatments with bacteria, On the other hand, in the treatment without bacteria, the inoculant was replaced by clean water. The density of bacteria added to the soil was 8 × 10⁷ CFU g soil⁻¹.

Irrigation was done by using 4‰ NaCl at 20, 40, and 75 DAS with a volume of 10 mL pot⁻¹.

In each pot, ten random mature leaves at two-third position from leaf stem to leaf tip were collected. The index in leaves was calculated by averaging three measurements on the leaves using a chlorophyll meter (Chlorophyll Meter SPAD) at 21, 28, 35, and 42 DAS. Fresh leaf samples at 42 DAS were extracted with N,N-Dimethyl Formamide, and chlorophyll content was measured at 664.0 and 647.0 nm wavelengths [42].

Proline was determined at 42 DAS by the Ninhydrin method and measured at 520 nm wavelength [43].

Growth determinations included plant height and panicle length. Plant height was measured from the ground to the tip of the highest leaf, measuring four plants in each pot. Panicle length was determined from the panicle neck to the tip of the panicle of 8 panicles per pot at 90 DAS [44].

Yield components were determined as described below. The number of panicles per pot: Counting the total number of panicles per pot; the number of seeds per panicle: Total number of seeds/total number of panicles, counting the number of seeds of eight panicles per pot; filled seed rate: Total number of filled seeds/total number of seeds × 100%; 1,000-seed weight: Weighing the weight of 1,000 filled seeds of each treatment [44].

Actual yield was the seed yield weighed and measured at harvest time of each pot and converted to 14% moisture content [44].

Analysis of soil properties at the end of the season was carried out according to the method of Sparks et al. [45]. pH_H2O and EC were extracted from the soil with distilled water at a soil: water ratio (1:5) and measured with a pH meter and an EC meter, respectively. pH_KCl was performed similar to pH_H2O but H₂O was replaced by KCl.

Total nitrogen: The soil was mineralized with a mixture of concentrated H₂SO₄, salicylic acid, and a catalyst mixture of CuSO₄: Na₂SO₄: Se. The total N content in the sample was determined by the Kjeldahl distillation method.

NH₄ ⁺ content in soil was determined by extracting the soil with 2.0M KCl, the resulting solution was developed with a mixture of sodium nitroprusside, sodium salicylate, sodium citrate, and sodium tartrate and a mixture of sodium hydroxide and sodium hypochlorite solutions. NH₄ ⁺ was quantified by measuring using a spectrophotometer at 650 nm.

The soluble P content was determined by the Bray II method by extracting soil with 0.1N HCl and 0.03N NH₄F, the soil:extract ratio was 1:7, developed with ascorbic acid and measured on a spectrophotometer at 880 nm.

The total P content of the soil was digested with a mixture of 5 mL concentrated H₂SO₄ and 1 mL HClO₄. The color was developed with a reducing mixture containing sulfuric acid, ammonium molybdate, ascorbic acid, and antimony ammonium tartrate and measured spectrophotometrically at 880 nm.

The cation exchange capacity (CEC) was extracted with 0.02 M MgSO₄ and titrated with 0.01 M EDTA using Eriochrome Black, Hydroxylamine-HCl in 96% Ethanol as indicators.

The cations were extracted with 0.1 M BaCl₂ and the solution was measured on an atomic absorption spectrophotometer at 422.7 nm (Ca²⁺), 285.2 nm (Mg²⁺), 589.0 nm (Na⁺), and 766.5 nm (K⁺).

The Al-P, Fe-P, and Ca-P contents were determined according to the method of Chang and Jackson [46]. Fe-P was determined by extracting the soil with 0.1 M NaOH, Al-P was extracted with 0.5 M NH₄F (pH = 8.2), and Ca-P was extracted with 2.5 M H₂SO₄, the extracts were colorized with ascorbic acid and measured on a spectrophotometer at 880 nm.

Total N and sodium (Na) content in rice stems, leaves, and seeds were determined according to the method of Walinga et al. [47]. Plant samples were digested with H₂SO₄ and H₂O₂. N content was determined by Kjeldahl distillation and titrated with 0.01 N H₂SO₄. Na content was measured on an atomic absorption spectrophotometer at 589.0 nm.

Data were processed using Microsoft Excel 2013 software and SPSS 13.0 software. In brief, the raw data were input into Microsoft Excel 2013 software. These data were then analyzed and compared by the SPSS software version 13.0 according to the Duncan’s test at 5% significance level.

Early-season soil characteristics of rice-shrimp soil collected in My Xuyen-Soc Trang had high salinity but slightly poor nutrient content. In brief, it had pH_H2O and pH_KCl at neutral levels (6.59 and 6.93, respectively), while the EC value was recorded as very high at 6.80 mS cm⁻¹. Total N content was assessed at a low level and NH₄ ⁺ content was at an optimal level. However, available P content was at an average level (Table 1).

In the first season, PNSB addition affected pH_H2O, EC, P_soluble, NH₄ ⁺, CEC, Mg²⁺, K⁺, and Na⁺ as shown in Table 2. The N-applying rates affected EC, P_soluble, and K⁺ in the soil at harvest. In the second season, PNSB addition affected pH_H2O, EC, P_soluble, NH₄ ⁺, Mg²⁺, Ca²⁺, K⁺, and Na⁺. The N applying rates affected NH₄ ⁺. In the first season and the second season, PNSB addition increased pH_H2O, P_soluble, NH₄ ⁺, and decreased EC and P_soluble in the soil. Significantly, after two consecutive crops, the nutrient contents increased while the EC values decreased. This shows the work of the N-fixing PNSB in highly saline soil.

The NH₄ ⁺ content in the soil in the treatment without added bacteria was the lowest (7.38 mg kg⁻¹), increased in the treatments supplemented with single strain S01 (9.03 mg kg⁻¹), S06 (10.1 mg kg⁻¹), and mixed S01 and S06 (14.9 mg kg⁻¹) in the first season. Similarly, the NH₄ ⁺ content in the treatments with no bacteria < S06 < S01 < S01 + S06 was 85.7 < 115.3 ∼ 111.9 < 125.6 mg kg⁻¹ in the second season (Table 2).

According to Table 2, the P_soluble content increased in the treatment supplemented with PNSB. In the first season, the highest P_soluble content was in the treatment supplemented with single strain S06 (25.6 mg kg⁻¹) and the lowest content was in the treatments without added bacteria (19.3 mg kg⁻¹). In the second season, the highest P_soluble content was in the treatments supplemented with mixed strain S01–S06 (14.5 mg kg⁻¹) and the lowest P_soluble content was in the treatment without added bacteria (12.0 mg kg⁻¹). The highest Fe-P, Al-P, and Ca-P contents in the treatments without added bacteria were 119.6, 29.2, and 114.2 mg kg⁻¹ in the first season; and 485.0, 152.6, and 150.6 mg kg⁻¹ in the second season, respectively. The Fe-P, Al-P, and Ca-P contents decreased in the treatments supplemented with single-strain PNSB or a mixture of two bacterial strains, ranging from 110.4–114.0, 24.4–26.6, and 80.53–104.2 mg kg⁻¹ in the first season; 394.3–427.8, 140.5–143.8, and 72.4–100.0 mg kg⁻¹ in the second season. However, the N_total and P_total contents in the soil were not statistically different in both seasons (Table 2).

Based on Table 2, the CEC in the first season was highest in the treatments without bacterial supplementation. However, the CEC between different N levels was not statistically different, ranging from 14.4–14.8 meq 100 g⁻¹ (the first season). Similarly, CEC values in the second season were equivalent, with 12.9–13.3 meq 100 g⁻¹ (N application rates) and 12.9–13.4 meq 100 g⁻¹ (PNSB). Treatments supplemented with PNSB had statistically significant differences in Mg²⁺, K⁺, and Na⁺ while different N applying rates had stable cation concentrations. Specifically, the addition of single-strain bacteria S01, S06, or mixed S01–S06 resulted in increased Mg²⁺ and K⁺ and decreased Na⁺ concentrations in the first season. In the second season, bacterial addition increased Ca²⁺ concentrations and had lower Na⁺ concentrations than no bacterial addition, while single-strain bacteria S06 and mixed S01 and S06 increased Mg²⁺ and K⁺ concentrations. Bacterial supplementation and combined N fertilization interacted with each other as shown by EC, P_total, P_soluble, Fe-P, Ca-P, NH₄ ⁺, and K⁺ in the first season; pH_H2O, Al-P, and Ca-P in the second season (Table 2).

There were significant interactions between the two factors in the NH₄ ⁺ content. In particular, in the first season, in the treatments supplemented with a mixture of two bacterial strains S01 and S06 combined with 75 and 50% N fertilizer, the NH₄ ⁺ content was equivalent, with 15.1 and 15.0 mg NH₄ ⁺ kg⁻¹, higher than the treatment without bacterial supplementation combined with 100% N fertilizer (7.58 mg NH₄ ⁺ kg⁻¹) (Figure 1).

Figure 1

Effect of nitrogen fertilizer supplemented with nitrogen-fixing PNSB on NH₄ ⁺ content of saline soil in My Xuyen district, Soc Trang province under greenhouse conditions. Note: Bars followed by different letters are statistically different at the 5% level (*). S01: Single strain of bacteria S01 added, S06: Single strain of bacteria S06 added, S01 + S06: Mixed addition of two strains of bacteria S01 and S06, 100%: 100% nitrogen fertilizer as recommended, 75%: 75% nitrogen fertilizer supplement as recommended, 50%: 50% nitrogen fertilizer supplement as recommended, and 0%: No nitrogen fertilizer.

From the changes in soil properties, such as increased NH₄ ⁺ content and decreased Na⁺ content under the applications of N-fixing PNSB, the nutrient uptake and content within rice plants were affected as well. The results of Table 3 show that the N content in the stem, leaves, and seeds, Na content in the stem, leaves, and seeds, N uptake in the stem, leaves, and seeds, total N uptake, and total Na uptake were statistically significantly different by 5% at all PNSB supplementation levels, N applying rates, in both seasons. Supplementing the mixture of two strains S01 and S06 combined with 50% N fertilizer, the total N uptake reached the same value as the treatment with only 100% N fertilizer in both seasons (Figure 2).

Figure 2

Effect of nitrogen fertilizer supplemented with nitrogen-fixing PNSB on total N uptake of rice grown on saline soil in My Xuyen district, Soc Trang province under greenhouse conditions. Note: Bars followed by different letters are statistically different at the 5% level (*). S01: Single strain of bacteria S01 added, S06: Single strain of bacteria S06 added, S01 + S06: mixed addition of two strains of bacteria S01 and S06, 100%: 100% nitrogen fertilizer as recommended, 75%: 75% nitrogen fertilizer supplement as recommended, 50%: 50% nitrogen fertilizer supplement as recommended, and 0%: No nitrogen fertilizer.

The dry biomass of stems, leaves, and seeds increased in the PNSB supplementation treatments. In which, supplementing the mixture of two strains PNSB had stem and leaf biomass of 14.2 g pot⁻¹, seed biomass of 17.1 g pot⁻¹ (the first season), and stem and leaf biomass of 25.0 g pot⁻¹ and seed biomass of 24.6 g pot⁻¹ (the second season) (Table 3).

According to Table 3, Na content, Na uptake in leaf stems, seeds, and total Na uptake decreased in the treatment supplemented with mixed S01–S06 (143.2 g pot⁻¹) in the first season. In addition, in the second season, the addition of single-strain S01, S06, or mixed S01–S06 all effectively reduced Na uptake in leaves, stems, seeds, and total Na uptake. Of which, Na uptake in leaves, stems, and seeds and total Na uptake decreased the most in the treatment supplemented with single-strain S01 in the second season, respectively, at 199.3, 57.5, and 256.8 g pot⁻¹. The results of Table 3 show that in the first season, Na content in seeds, leaf and stem biomass, seed biomass, and seed Na uptake, and in the second season, Na content in leaves and seeds, leaf and stem biomass, seed biomass, leaf and stem Na uptake, seed Na uptake, and total Na uptake had a statistically significant interaction of 5%.

By the improvements of N uptake and reduction in Na uptake, the biochemical traits of rice plants were improved. Particularly, based on Table 4, the SPAD index in the treatments supplemented with PNSB at 21, 28, and 35 DAS was higher than that in the treatments not supplemented with PNSB, with 31.5–32.4 > 30.5, 33.6–34.3 > 32.5, and 34.2–34.5 > 32.6, respectively. In the second season, the SPAD index also achieved similar results, with 32.3–33.6 > 31.7, 32.0–33.0 > 31.1, and 31.1–31.2 > 30.3. In addition, the SPAD index decreased in the case of reducing the amount of N fertilizer in both seasons in the order of 100 > 75 > 50 > 0%.

The chlorophyll a content of the PNSB-supplemented treatments was higher than that of the non-PNSB-supplemented treatments, with 2.95–3.68 > 2.84 µg mL⁻¹ (the first season) and 2.34–2.48 > 2.10 µg mL⁻¹ (the second season). For the N fertilization levels, the lower the N level was, the lower the chlorophyll a content became, 100 > 75 > 50∼0% (the first season) and 100 > 75 > 50 > 0% (the second season), respectively. In addition, the chlorophyll b content was the highest in the non-PNSB-supplemented treatments in the first season, with 1.98 µg mL⁻¹. In addition, in the second season, the chlorophyll b content of the treatments without PNSB supplementation was equivalent to that of the single-strain S06 supplementation but higher than that of the single-strain S01 and mixed S01 and S06 supplementation treatments, with 0.808∼0.758 > 0.667–0.729 µg mL⁻¹. In addition, the chlorophyll a + b content between the different N fertilizer levels was statistically significant at 5%, ranging from 4.19–5.88 µg mL⁻¹ (the first season) and 2.03–3.90 µg mL⁻¹ (the second season) (Table 4).

The proline content of the treatments with PNSB supplementation was lower than that of the treatments without PNSB supplementation. Specifically, in the first season, the single strain S01 and single strain S06 had similarly low proline content, followed by the mixture of two strains S01 and S06, and both were lower than without bacteria supplementation, with 16.4–16.6 > 18.1 > 23.9 µmol g⁻¹ DW. In addition, the single strain S01 had the lowest proline content in the second season, with 4.20 µmol g⁻¹ DW. In addition, the higher the chemical fertilizer application was, the higher the proline content became, respectively, 100∼75 > 50 > 0% (the first season) and 75 > 100 > 50 > 0% (the second season). The interaction was statistically significant at 5% between the two factors PNSB addition and N fertilization, for chlorophyll index at 21, 35 DAS and proline content of the first season, chlorophyll index at 21, 28, 35, and 42 DAS, chlorophyll a, b, a + b, and proline in the second season.

From the improvement in chlorophyll contents, SPAD indices, and proline indicator, the plants growth and yield were improved by the application of N chemical fertilizers and N-fixing PNSB. In the first season, plant height, panicle length, the number of seeds per panicle, filled seed rate, and rice yield increased in the treatments supplemented with single strains S01, S06, or the mixture of two strains S01 and S06. It gradually decreased when reducing fertilizer levels of 100, 75, 50, and 0%. However, the weight of 1,000 seeds was not statistically significant, ranging from 22.7–23.0 g in the PNSB supplement factor and 22.6–23.4 g in the N fertilizer level factor. Besides, in the second season, there were similar results, plant height, the number of panicles per pot, the number of seeds per panicle, filled seed rate, and rice yield of the bacterial supplement treatment were higher than the treatment without PNSB. However, bacterial supplementation did not affect panicle length in the second season (Table 5). It is showed that plant height, filled seed rate, and yield in the first season, plant height, panicle length, yield components, and yield in the second season interacted significantly between the additional factors PNSB and N applying rates, except for the number of seeds per panicle and 1,000-seed weight.

In addition, 50–75% N fertilization with 2 strains S01 and S06 or single strain S01 had seed yield equivalent to the treatment with only 100% N fertilization and higher than the treatment with only 50–75% N fertilization in the first season. On the other hand, no N fertilization (0% N) with bacteria supplementation had a higher yield than 0% N and no bacteria supplementation (Figure 3).

Figure 3

Effect of nitrogen fertilizer supplemented with nitrogen-fixing PNSB on rice yield grown on saline soil in My Xuyen district, Soc Trang province under greenhouse conditions. Note: Bars followed by different letters are statistically different at the 5% level (*). S01: Single strain of bacteria S01 added, S06: Single strain of bacteria S06 added, S01 + S06: Mixed addition of two strains of bacteria S01 and S06, 100%: 100% nitrogen fertilizer as recommended, 75%: 75% nitrogen fertilizer supplement as recommended, 50%: 50% nitrogen fertilizer supplement as recommended, 0%: No nitrogen fertilizer.

Similarly, in the second season, the treatment with 75% N supplemented with single strain S01, single strain S06, and mixed S01–S06 was higher than the treatment with only 75% N. In addition, the treatment without N fertilization but added with single strain S06 or mixed S01–S06 had a higher yield than the treatment without N fertilization and without bacteria supplement (Figure 3).

PNSB supplementation contributed to an increase in pH_H2O compared to no bacteria supplementation in saline soil (Table 2). PNSB released metabolites such as NH₄ ⁺, 5-aminolevulinic acid (ALA), exopolymeric substances (EPS), IAA, and siderophores. In EPS secreted from PNSB, functional groups such as –OH and –COOH bind H⁺ to reduce the H⁺ concentration in the soil, helping to increase the pH value [34,41]. According to Chowdhury et al. [48], pH plays an important role in supporting rice growth in saline soil. In addition, pH affects the availability of nutrients [49]. PNSB supplementation resulted in lower EC concentrations, from 0.35 to 0.52 mS cm⁻¹ compared to the treatment without bacterial supplementation (Table 2). According to Sundar et al. [50], EC is one of the main factors determining the growth and development of rice. EC in the soil directly affects the availability of nutrients, CEC, and soil health [51].

The CEC of the PNSB-supplemented treatment was lower than that of the non-PNSB-supplemented treatment. This resulted in higher Mg²⁺ and K⁺ contents in the soil in the PNSB-supplemented treatment than in the non-PNSB-supplemented treatment [52]. In addition, PNSB reduced Na⁺ content in soil from 16.0% in the first season to 25.7% in the second season. The reason is that EPS contains functional groups, including –OH and –HOOC, which are capable of binding to Na⁺ ions [34].

The content of NH₄ ⁺ and P_soluble increased in the treatment supplemented with R. sphaeroides S01 and S06, and reduced insoluble P in both the seasons because R. sphaeroides S01 and S06 can solubilize insoluble P forms [40]. PNSB solubilizes P in soil, releasing P in soluble form [41,53], leading to improved soil health [50]. The results of this study are consistent with that in the study by Khuong et al. [41] where adding PNSB including Luteovulum sphaeroides W03 and W11 to solubilize P from Al-P, Fe-P, and Ca-P on saline soil resulted in higher soluble P content in soil compared to no bacteria added. Therefore, adding PNSB reduced 50% of the recommended amount of chemical P fertilizer for rice. In addition, PNSB converts molecular N₂ in soil pores into NH₃ or NH₄ ⁺ by nitrogenase enzymes for plants to easily absorb [34]. Ultimately, although, the soil was highly salinized, the N-fixing PNSB still performed properly at fixing N₂ to NH₄ ⁺. This is because these N-fixers were isolated from a highly saline condition [40]. Thus, the current soil condition was suitable for the work of the N-fixing PNSB strains.

Due to the increased availability of soil nutrient and decreased soil salinity, the uptake and biomass of rice plants were improved by the N-fixing PNSB application. Therein, supplementation of PNSB increased both leaf and stem biomass and dry seed biomass compared to treatments without bacterial supplementation (Table 3). According to Irakoze et al. [54], rice plants exposed to salt stress had reduced growth and fresh and dry biomass. This was due to a decrease in water potential in the cells, causing stomatal closure and leading to limited carbon dioxide assimilation [55]. High salinity reduced N content and N uptake in the stems [41,55]. Therefore, supplementing a mixture of two strains S01 and S06, N content in leaf, stem, and seeds and N uptake were all higher than treatments without PNSB supplementation in the two seasons. According to Marag and Suman [56], PNSB play an important role in the N fixation process, dissolving macronutrients and micronutrients into available forms for plants. In addition, PNSB supplementation decreased Na accumulation in plant parts because PNSB produced EPS to immobilize Na to prevent Na transport [57]. Therefore, PNSB supplementation reduced Na content in rice stems, leaves, and seeds in the two seasons.

With the improvements in nutrient uptake and content in rice plants, the biochemical traits were ameliorated. In particular, SPAD index, chlorophyll a, b, and a + b content decreased at different levels of N fertilizer application in both seasons from 21 to 42 DAS. However, the SPAD index, chlorophyll a, b, and a + b contents increased in rice leaves in the PNSB-supplemented treatment (Table 4). N helps increase growth because N is a component of chlorophyll, amino acids, and nucleic acids [58]. Chlorophyll content in leaves reflects the growth status of crops and helps guide the appropriate use of N fertilizer [59]. This result is consistent with Yen et al. [60], where supplementing a mixture of two bacterial strains resulted in a higher SPAD index than using a single strain and no bacterial supplementation. According to Sundar et al. [50], chlorophyll content in leaves increased due to increased δ-ALA content. PNSB produces δ-ALA to increase the presence of protoporphyrin IX, an important precursor for chlorophyll synthesis. Salt stress directly affects photosynthesis and chlorophyll synthesis in plants [61]. Rice cultivation on saline soil supplemented with PNSB resulted in increased chlorophyll content, potentially increasing photosynthesis, increasing energy production, and improving the growth and yield of rice plants [62].

Proline content in plants at 42 DAS was high in the absence of bacterial supplementation. Proline content decreased gradually in the treatment supplemented with PNSB or reduced N fertilization (Table 4). According to Salinas et al. [63], proline is an indicator to assess salt damage to plants. According to the research by Ábrahám et al. [64] in saline soil, the proline content in plants increased, because proline is accumulated in plant cells [65]. The predominant endogenous proline accumulated in plants in response to salinity helps rice plants increase their salt tolerance [66]. According to Khuong et al. [67], PNSB produces EPS to immobilize Na⁺, reducing soil salinity. Therefore, in treatments without PNSB supplementation, the proline content was higher than in treatments with PNSB supplementation. In other words, because the N-fixing PNSB eased the soil salinity, the salt stress of rice was lessened, leading to lower proline content in rice [68].

Because the SPAD indices were improved, the chlorophyll content was enhanced, and the salt stress was lessened, the growth and yield of rice were elevated. Reducing N fertilization levels resulted in a decrease in plant height, panicle length, and yield and yield components. However, single-strain S01, S06, or mixed S01 and S06 supplementation increased the growth and yield components of rice (Table 5). N is one of the macronutrients that play an important role in determining rice yield [69]. According to Javed et al. [70], adequate N fertilization is considered an important and economical measure to address salt stress and minimize the adverse effects of salt on plant growth. Ijaz et al. [71] demonstrated that when rice plants are exposed to salt, cell elongation and division are affected, leading to reduced root and leaf growth. Supplementation of R. sphaeroides S01 and S06 can fix N, solubilize P, and produce growth stimulants IAA, EPS, siderophores, and ALA under rice cultivation conditions [40]. This contributes to an 86.8% increase in rice seed yield compared to no PNSB supplementation [32]. According to Sundar et al. [72], supplementation of R. palustris contributes to a 65% increase in rice seed yield. In the current study, the N-fixing PNSB was able to replace 25% of the N chemical fertilizer. This is in accordance with some other N-fixers such as, Azotobacter chroococcum, Azospirillum brasilense, and Bacillus megaterium [73].

However, the current liquid biofertilizer should be further investigated for performance under field conditions. Therefore, a suitable carrier should be selected. A carrier can not only lessen the influences of the environment on the microorganisms [74] but also promote the colonization of the microbes under adverse conditions [75]. Some suitable carriers for PNSB can be named as rubber wood ash, decanter cake, rice husk ash, and spent coffee grounds [75]. However, despite a number of previous field trials testing the effectiveness of PNSB on rice [50,76,77], the combination of chemical fertilizer and PNSB on rice has not been focused yet, because of the fact that chemical fertilizers cannot be 100% altered due to their main influences on crop attributes [78]. Thus, a field study is oriented after this greenhouse experiment.