Swine manure was obtained from a swine farm in Kimpo, Korea. Sludge was removed from the swine manure, and liquid waste was sprayed into the filtering system made up of thinning-out tree wood chips piled in a container (10 m width×10 m length×6 m height). The wood chips were used to pack the filter, which was then utilized in the microbial-degradation phase of the composting. The filtering system was spontaneously self-heated in aerobic condition and kept at approximately 60°C for several days. Wood chip temperatures reached a peak of 66°C during the second week of composting and were maintained at that level until d 21 for the thermophilic stage of composting. The average temperature of the filtering system during this process was 60.7°C. Temperatures in the composting piles decreased gradually after the thermophilic stage until they reached ambient temperatures of approximately 27 to 30°C. The pH of the final wood chip compost was 8.3.

Composted wood chips (1 g) were mixed with 5 mL LB media, stirred for 5 min, and incubated at room temperature for 30 min. Sample supernatants (100 μL) were plated on 1.5% LB agar plates and incubated at 55°C for 24 h in a sealed humidified jar to prevent plate drying. Three strains were selected for further investigation based on the colony color, odor, and morphological features.

The bacteria were identified by 16S rRNA sequence homology analysis as follows. Genomic DNA was isolated from these strains and used for the amplification of the 16S rRNA gene fragment as previously described (Sogin, 1990). Briefly, 16S rRNA gene fragment was amplified by the polymerase chain reaction (PCR) using the eubacterial universal primers 27F (5′-AGAGTTTGATCCTGGCTCAG-3′) and 1492R (5′-GGTTACCTTGTTACGACTT-3′). Polymerase chain reaction (PCR) were performed using a thermal cycler (Takara Bio Inc. Japan) with an initial denaturation step of 94°C for 3 min, followed by 35 amplification cycles consisting of denaturation at 94°C for 60 s, annealing at 60°C for 60 s, and elongation at 72°C for 60 s. The reaction was terminated with a terminal elongation step of 72°C for 3 min followed by cooling at 4°C. The resulting PCR band was used for nucleotide sequence analysis. Nucleotide sequences of approximately 1,450 bp length were analyzed for similarity to the 16S rRNA sequence of the isolated bacteria by using the BLAST program (Altschul et al., 1997). The nucleotide sequences of 16S rRNA were analyzed three times for each strains and aligned using the ClustalW program for multiple sequence alignment (Thompson et al., 1994). The evolution distance matrix was calculated using the Kimura 2-parameter evolution model (Kimura, 1983). The resulting phylogenetic tree was constructed using neighbor-joining analysis (Saitou and Nei, 1987; Felsenstein, 1993). Biochemical identification was also performed using API 50CHB (BioMerieux, Hazelwood, MO) according to the manufacturer’s instructions.

The optimum NaCl concentration for bacterial growth was determined using previously described methods (DeFlaun et al., 2007). The effect of temperature on growth was tested in LB broth. An additional study was performed to determine the optimal pH for growth in a buffered system. The LB broth was adjusted to pH values ranging from 5.0 to 9.5 (room temperature) by using the following buffers: 50 mM citrate-Na₂HPO₄ buffer (pH 5.0 to 6.8), 50 mM Tris-HCl buffer (pH 7.0 to 8.0), and 50 mM glycine-NaOH buffer (pH 8.5 to 9.5). Each LB broth was inoculated with an SG-01 culture that had been grown overnight. Optical density at 600 nm was analyzed by measuring absorbance values every 6 h. The initial optical density of each culture was 0.04. The Specific growth rate is defined as the increase in cell mass per unit time, e.g., grams cells (g) per gram cells (g) per hour: (g·g⁻¹·h⁻¹).

A jar fermenter was used to investigate the SG-01 growth curve and to perform nitrate and nitrite reduction assays. The pH changes were also monitored during this process. Bacterial cells (1%) were inoculated into 100 mL LB broth containing 2.0% NaCl and supplemented with 100 ppm nitrate (1.18 mM NaNO₃) or nitrite (1.45 mM NaNO₂), and then incubated at 55°C for 48 h. The modified brucine-sulfanilic acid method (APHA, 1989) was used to measure nitrate concentrations. Nitrite concentrations were measured using the Griess reaction (Green et al., 1982) Color absorbance was measured at 540 nm by using a spectrophotometer (U-2000, Hitachi, Japan). Un-inoculated samples were used as controls for the nitrate and nitrite reduction assays. Experiments were performed in triplicate, and standard curves were generated for each experiment by using known concentrations of sodium nitrate and sodium nitrite as references.

Nine different colonies from the samples of the wood chips composted with swine manure were isolated and grouped based on colony morphological characteristics, color, and odor. Biochemical characteristics were analyzed in 3 rod-shaped spore-forming candidate bacteria, SG-01, SG-02, and SG-03.

The PCR products for 16S rRNA (approximately 1,450 bp) were isolated and sequenced. Sequence homology analysis revealed that SG-01 had 98 to 99% homology with the Geobacillus genus. However, the sequences of the 3 strains were not-identical and the strain SG-01 was phylogenetically related to Geobacillus thermodenitrificans species (99% similarity with 16S rRNA of the strain 465^T). The phylogenetic position of the studied strains among other Geobacillus species is shown in Figure 1.

The results of biochemical analyses using the API 50CHB kit revealed that the SG-01, SG-02, and SG-03 strains were biochemically unique based on the utilization of L-arabinose, fructose, inositol, melibiose, sucrose, raffinose, and glycogen (Table 1). The SG-02 and SG-03 strains used L-arabinose, melibiose, and sucrose, whereas the SG-01 strain used glycogen. It was also found that the SG-02 strain used inositol as a carbon and energy source.

Manachini et al. (2000) reported that G. thermodenitrificans type strain DSM 465T grows well at 50 to 65°C, with optimum pH values of 6 to 8 and in the presence of 3% NaCl. It is capable of reducing nitrates and nitrites to gas and of anaerobic gas production from nitrate. A common feature between SG-01 strain and DSM 465T strain was observed but specific growth rate of SG-01 strain was not as high as those for other Geobacillus sp. (>0.1). More than 70% of the isolates from the wood chip compost piles were classified as SG-01 strains based on the colony morphological characteristics, growth patterns, and 16S rDNA sequence data. This suggests that Geobacillus thermodenitrificans is a common inhabitant of composts at 60°C under aerobic conditions, as earlier reported (Hatsu et al., 2002).

Optimal conditions for the growth of strain SG-01, including temperature, pH, and NaCl concentration, were investigated. The specific SG-01 strain growth rates were 0.024 at 50°C, 0.027 at 55°C, 0.026 at 60°C, and 0.025 at 65°C. Poor growth was observed at 70°C and 37°C. The optimal temperature for the growth of SG-01 was 55°C. The optimum pH for SG-01 growth was 7.0. No growth was detected at pH values greater than 9.5. The strain grew in liquid media at NaCl concentrations of 0 to 3%. The optimal NaCl concentration was 2.0%, showing a specific growth rate of 0.025. The specific growth rate of strain SG-01 decreased at NaCl concentrations greater than 3.2%.

The SG-01 growth curve was investigated using a jar fermenter at 55°C and 2.0% NaCl for 48 h. The maximum cell density of 2.87×10⁸ CFU/mL of the strain SG-01 was reached in the LB broth at 12 h (Figure 2A). The pH was not changed during the fermentation period (pH 7.0). Nitrogen reduction by SG-01 in nitrate-and-nitrite-containing LB broth was investigated. Nitrate was rapidly reduced within 12 h, and approximately 91.8% of the nitrate was removed within 48 h (Figure 3A, 3B). The nitrate values in uninoculated controls did not decrease. These results suggest that strain SG-01 may reduce nitrate to nitrogen under aerobic conditions. Approximately 99% of the nitrite was reduced in 18 h, and nitrites were no longer detected at 48 h. Nitrite values of the inoculated controls did not decrease. The denitrifying ability of strain SG-01 was stronger than those of other microorganisms (Kariminiaae-Hamedaani et al., 2004). These results suggest that SG-01 could effectively remove nitrates and nitrites from wastewater.