Soil samples were collected from different rhizospheric soils and under decayed rice straw in Kyaukse district and around Patheingyi Township, Mandalay Region, Myanmar. Glucose Nitrogen Free Mineral Medium (GNFMM) was used to isolate nitrogen-fixing bacteria. The composition of the isolated medium was as follows (g/L): 1.0 K₂HPO₄, 1.0 CaCl₂, 0.5 NaCl, 0.25 MgSO₄·7H₂O, 0.01 FeSO₄·7H₂ or O, 0.01 Na₂MoO₄·2H₂O, 0.01 MnSO₄·5H₂O and a carbon source was glucose (7 g/L). Solid medium was produced by adding 2% agar.

The visual detection of nitrogen-fixing activity was observed by using NFMM with 0.7% Glucose. Bromothymol blue (BTB) was added to the media as an indicator. The media was prepared for both agar and broth. After 7 days incubation, changing the color of the medium was recorded. To detect nitrogen-fixing activity from the broth culture, the reagents of ammonium test kit (VISCOLOR Alpha Ammonium reagent, MACHEREY-NAGEL GmbH & Co. KG, Germany) which was kindly supported from Shibaura Institute of Technology, Japan, were added and the appeared color was noted by comparing the color chart on the test kit. There are three kinds of reagents of ammonium test kit. Two drops per 1 ml of ammonium test kit reagent (I) were added into the supernatant. And, one-fifth of reagent (II) was added and incubated for 5 minutes. Finally, one drop per 1 ml of reagent (III) was added and incubated again for 5 minutes. And then, the color development of supernatant was observed and compared with the color chart from the ammonium test kit.

The best nitrogen-fixing strain was chosen and identified by 16S rDNA sequencing. DNA extraction was performed using the PrepMan reagent (Thermofisher Scientific, California, US) and primers 10F (5'-AGTTTGATCCTGGCTC-3') and 800R (5'-CTACCAGGGTATCTAAT-3'). 16S rDNA was amplified by PCR 30 cycles using a universal primer. Each cycle consisted of denaturation for 1 min at 94^°C, annealing for 30 s at 56^°C, and extension for 45 sec at 72^°C and final extension for 45 sec at 72^°C. Sequencing reactions were performed using the Big Dye Terminator Cycle Sequencing Kit (v.3.1) (Thermofisher Scientific, California, US) and analyzed in Genetic Analyzer 3130 sequencer (Applied Biosystems, US). Nucleotide sequences were analyzed using BLAST on the NCBI BLASTN and Greengenes.lbl.gov.

Some phenotypic characteristics of A. agilis were studied and characterized such as Triple Sugar Iron Agar (TSI), Motility, Indole, Methyl Red, Citrate Utilization, Voges-proskauer, Starch hydrolysis, Gelatin liquification, Catalase, Cellulase, Urease, growth at different NaCl concentrations, some antibiotic sensitivity patterns and utilization of some sugars, glucose, sucrose, fructose, arabinose, cellulose, CMC (Carboxymethyl Cellulose), and mannitol.

A single colony of the bacterium was inoculated into 10 ml of NFGMM broth and incubated at 37^°C for 2-3 days before they were inoculated in nitrogen-free liquid medium with various carbon sources. 1ml of culture broth was inoculated into NFMM broth containing various carbon sources; glucose, sucrose, fructose, arabinose, cellulose, CMC and mannitol (7g/L for each sugar). After one week incubation, accumulated ammonium concentration in culture broth from various carbon sources was determined by Indophenol method [13]. Each experiment was replicated three times.

Various NaCl concentrations, 3%, 5%, 7%, 9% and 12% were used to screen the growth and ammonia accumulation of A. agilis using GNFMM. After inoculation in GNFMM broth containing various NaCl concentrations and incubation for one week, accumulated ammonium concentrations were measured by Indophenol method.

An inoculum from an overnight grown culture in log phase was added to 100 ml NFMM containing cellulose powder (Sigma-Aldrich) as a carbon source in 250-ml Erlenmeyer flasks. After incubation for 48 h, at 37^°C, under shaking condition of 150 rpm, the culture was harvested and growth was measured at OD600 with UV-Vis spectrophotometer (Optima SP-3000 Plus, Slovakia). The cultures were centrifuged at 10,000 rpm for 10 min at 4^°C. After centrifugation, the supernatant (cell-free extract) was used as crude preparation to measure cellulase activity.

For enzyme assay, Cellulose and Sodium CMC were used as the substrates. Enzyme activity was determined by measuring the release of reducing sugars during the enzyme-substrate reaction using dinitrosalicylic acid method [14]. The values were determined from glucose standard curve. One unit (IU) of activity was defined as the amount of enzyme required to liberate 1μmol of glucose per minute under given assay conditions.

A single colony of the bacterium was inoculated into NFMM containing cellulose as a carbon source. After incubation at 37^°C for 48 h, under shaking condition of 150 rpm, the supernatant (after centrifugation) was prepared for cellulase activity using agricultural residues. Different agro-residues like rice straw, filter paper and sawdust were used.

The isolated nitrogen-fixing bacterium was identified as A. agilis by 16S rDNA sequencing with 99.34% similarity. Colonial and microscopic morphology of A. agilis was shown in Fig. (1). Some phenotypic characteristics of A. agilis were characterized in (Table 1).

Fig. (1). Colonial Morphology and Microscopic Morphology (x100) of A. agilis.

For preliminary screening of nitrogen-fixing activity by plate screening method, A. agilis caused color changing of the GNFMM from green to blue color after one-week incubation as shown in Fig. (2). Such kind of screening can be easily seen and select the bacteria having nitrogen fixing activity. Nitrogen-fixing bacteria that can excrete a significant amount of ammonia into their environment can change the color of the media. Nitrogen-fixing bacteria normally fixed atmospheric nitrogen and excreted ammonia from fixation into their surrounding environment. In this case, ammonia was accumulated in the media from the nitrogen-fixing activity of bacteria and increasing the ammonium concentration caused pH changing of the media. As a consequence, it caused color changing of the media. The finding was compatible with Gothwal et al., Iwata et al. who described that nitrogen activity of Azotobacter beijerinckii and Lysobacter enzymogenes DMS 2043T and rhizospheric bacterial isolates were screened by changing the color of BTB containing nitrogen free media, and these strains showed a color change in BTB containing media, suggesting excretion of ammonia [15, 16].

Fig. (2). Screening on Nitrogen Fixing Activity, (a) Control, (b) Color Changing of Culture of A. agilis and (c) Detection by Ammonia Test Kit (the yellow bottle is the control).

Bali et al. found that the rise of pH to approximately 8.5 occurred in ammonia excretion of Azotobacter vinelandii for both wild type and mutant strains during the period when ammonium was released.

On detection with ammonia test kit, the supernatant of A. agilis culture broth gave color development after harvesting the bacterial broth. The estimated amount of above 3 ppm of ammonium concentration was obtained from the comparison of standard color chart on the test kit Fig. (2). This detection could give estimate amount of ammonium concentration. Ammonium concentration was increased after once week incubation. It was also reported that free-living diazotrophs fix dinitrogen sufficient for their own needs and do not generally excrete significant amounts of ammonium into their environment; fixed nitrogen is released after death and lysis of bacteria [17].

A. agilis accumulated ammonia in NFMM broth in utilizing various carbon sources. Among seven carbon sources, the bacterium gave the highest amount of ammonium concentration 14.665 ppm from the utilization of sucrose as a carbon source and followed by fructose. It was found that ammonium concentration was the lowest in arabinose containing NFMM broth, which was only 0.326 ppm of ammonium concentration (Table 2).

It was found that ammonia accumulation of A.agilis in NFMM broth was depended on carbon source. For significant ammonia excretion into their environment, it was needed that carbon source in the media should be their preference besides other physical parameters. Therefore, the nitrogen-fixing activity of A.agilis by using other nitrogen-free media should be studied. The dominant sugars in root exudates are glucose, fructose, galactose, arabinose, maltose, raffinose, rhamnose, sucrose and xylose [18]. Therefore, it can be assumed that isolated A.agilis can fix atmospheric nitrogen well and excrete ammonia into their environment when applied in field trial because sucrose is one of the dominant sugars in root exudates.

A.agilis can grow on NFGMM with various NaCl concentrations in the range from 3% to 12% but growth became poor with increasing NaCl concentrations. In this study, it was obviously seen that NaCl effects on ammonia accumulation of A. agilis in the culture broth. Normally, NFMM contains 0.005% NaCl concentration and in this condition, the bacterium excreted ammonia concentration of about 14 ppm. When 3% NaCl concentration as the lowest and 12% as the highest concentration were used, the ammonia concentration in culture broth decreased from lowest concentration to highest concentration of NaCl. Ammonia concentration was obviously decreased when compared with the culture broth in original NFMM media (control). At the lowest concentration of NaCl, it was found that only 0.918 ppm of ammonium concentration could be accumulated. This concentration was very low and not significant excretion (Fig. 3).

Fig. (3). Accumulated ammonium concentrations of A. agilis in GNFMM broth with various NaCl concentrations. With increasing NaCl concentration, it was seen that ammonium concentration was obviously decreased.

Nitrogen fixation is particularly sensitive to salt stress, the nitrogenase activity of Klebsiella pneumoniae was sharply decreased as soon as the osmolarity was increased, with no activity detected at NaCl concentrations higher than about 3% [19]. However, it was also stated that the nitrogen-fixing activity of salt-tolerant Azotobacter was maximal at 5–25% NaCl [20-22].

Cellulolytic activity was assayed in nitrogen-free media because this research work aims to improve soil fertility by the nitrogen-fixing activity of diazotroph from the utilization of various agricultural residues. Cellulose and CMC in powder form of 0.2% were used as a substrate for assay. Reducing sugar concentrations were almost the same from both substrates as shown in Fig. (4). After 4 days incubation in broth culture, cellulase activity was decreased. Cellulase activity of A.agilis was not high because nitrogen deficient condition may inhibit A. agilis to produce cellulase enzyme. Higher production of enzyme requires the presence of complex nitrogen sources [23].

Fig. (4). Reducing sugar concentrations converted from cellulose and CMC by cellulase activity of A. agilis in NFMM. This reducing sugar concentration was detected in nitrogen free condition for the bacterium. The concentrations were almost the same for both substrates.

A maximum cellulose activity of 0.043 U/ml has been reported from cell-free culture supernatants of Geobacillus sp. isolated from a deep goldmine environment [24]. Li et al. (2008) [25] reported maximum cellulase activity (0.26 U/ml) of a Bacillus sp. when the culture was grown in LB medium supplemented with 1% CMC. In another study, a cellulase activity of 0.0113 U/ml was observed under optimized conditions from Geobacillus sp [26].

Among three different substrates, cellulase activity of A. agilis was higher in using filter paper and rice straw than the other one. Cellulase activity was less on Sawdust. It can be seen that the bacterium can easily degrade filter paper and rice straw than sawdust. It may promote the cellulase activity of the bacterium for the substrate of sawdust when nitrogen source into media was added. Results were shown in Fig. (5). Rice straw and filter paper are abundant in our country. After harvesting rice in the field, it is needed to effectively degrade rice straw. Utilization of cellulosic wastes as cheap carbon and energy sources can serve as an alternative to our rapidly depleting finite sources of fossil fuels.

Fig. (5). Reducing sugar concentrations converted from three substrates (filter paper, rice straw and sawdust) by cellulase activity of A. agilis in NFMM broth.