In this study bacterium B. subtilis A1 was used, which was isolated and identified from an Indian crude oil reservoir also crude oil used in this study was collected from same oil reservoir, the sampling site was presented in Figure 1 (latitude: 10.6694 and longitude: 79.3155). This strain was identified by 16S rDNA sequencing and deposited under NCBI Genbank accession number KP895564. The strain was retrieved and sub-cultured in Luria–Bertani (LB) agar plates [g/l 10.0 tryptone, 5.0 yeast extract, 10.0 sodium chloride with 15.0 agar (Himedia, Mumbai, India)] and incubated at 37°C for 24 h. Further optimized conditions were applied to culture preparations by single colony inoculation method using LB broth (pH: 7.0) and incubated in an orbital shaker (150 rpm) for 24 h at 37°C.

Biosurfactant production was aerobically carried out in 500 ml Erlenmeyer flask containing 200 ml of sterile Minimal Salt Medium (MSM) (g/l: 0.2 MgSo₄, 0.02 CaCl₂, 1.0 KH₂PO₄, 1.0 K₂HPO₄, 1.0 NH₄NO₃, and 0.5 FeCl₃ Himedia, Mumbai, India), supplemented with 1% (v/v) sterile crude oil (0.22 μm syringe filtered). In triplicate flasks, the pre-culture of B. subtilis A1 was inoculated (1.6 × 10⁴ CFU ml^-1) and incubated at 37°C in an orbital shaker at 200 rpm for 7 days. At the end of the incubation, the biosurfactant was extracted by centrifugation (refrigerated centrifuge, Remi-India: R-248) of culture medium at 4°C for 20 min at 3400 × g and the resultant supernatant was utilized for screening purposes. All the assays were performed in triplicate and sterile distilled water was used as the control.

Drop collapse test was performed by following the procedure described by Jain et al. (1991) and Patowary et al. (2016) with slight modifications. A drop of crude oil was applied to the glass slide, after that a drop of cell free culture broth was added onto crude oil drop and drop collapse activity was noted. Biosurfactant-producing culture gave flat drops. Deionized water was used as negative control and Triton X-100 (a chemical surfactant) solution used as positive control (1 mg/ml) (Thavasi et al., 2011).

Oil displacement technique was carried out as described previously Hassanshahian (2014) 50 ml of distilled water was added to petri dishes followed by addition of 100 μl of sterile crude oil to the surface of the water. Then, 10 μl of the culture filtrate was put on the crude oil surface. The diameter of the clear zone on the oil surface was measured. A negative control was maintained with distilled water (without surfactant), in which no oil displacement or clear zone was observed. Triton X-100 was used as the positive control (Thavasi et al., 2011).

The emulsification activity of the biosurfactant solutions was determined by measuring the emulsion index (E24) at 25°C as described by Wang et al. (2014). In general, 4 ml of crude oil was poured separately into a test tube containing 4 ml of biosurfactant solution. After being vigorously vortexed for 2 min, the test tube was kept for 24 h and the heights of emulsion, oil and aqueous zones were measured. The emulsion index (E24) was determined as the percentage of height of the emulsified layer (mm) divided by the total height of the liquid column (mm).

At end of each optimization studies, bacterial cells were removed from surfactant-containing medium by centrifugation using refrigerated centrifuge (Remi-India: R-248) for 20 min at 13,500 × g at 4°C and the supernatant was used for the emulsification activity. The optimal growth conditions of the strain were confirmed by emulsification activity and bacterial biomass of each parameter. Bacterial biomass was obtained as described in Santos et al. (2014). Cell free supernatant collected from the optimized study was used for quantify biosurfactant. Crude biosurfactant was obtained as described in Gudiña et al. (2015). In brief, supernatant was acidified to pH 2 using HCl and left for precipitation, precipitated biosurfactant was pooled by centrifugation [refrigerated centrifuge (Remi-India: R-248)] at 7,600 × g for 20 min at 4°C. Obtained crude biosurfactant was suspended in double-deionized water and pH was adjusted to 7.0. The biosurfactant solutions were freeze-dried and the products obtained were weighed and stored at -20°C. The surfactant collected in this method was considered as partially purified biosurfactant and used for the characterization purposes.

The extracted biosurfactant was further characterized by Fourier transform infrared spectrum (FT-IR) and gas chromatographic mass spectrum (GC-MS) methods. The functional groups of the surfactant collected from B. subtilis A1 was qualitatively characterized by FT-IR (Perkin–Elmer, Nicolet Nexus – 470). The dried biosurfactant was ground with the addition of potassium bromide in the ratio of 1:100 and the pellet was fixed in the sample container, and analyzed in the mid IR region 400–4000 cm^-1. For GC-MS analysis, ∼10 mg of biosurfactant was mixed with 5% HCl-methanol reagent. After the reaction was quenched with addition 1 ml of sterile H₂O, the sample was recovered with methanol and 1 μl of sample was injected into a gas chromatograph [Shimadzu QP2010 Ultra, Rtx-5Sil MS (30 m × 0.25 mm ID × 0.25 μm)]. The carrier gas used was Helium, the flow rate was set as 1.5 ml min^-1 and the working temperature of the GC injector was 260°C. The gradient temperature was set as range from 60 to 260°C at a speed of 5°C min^-1, through an isothermal phase of 10 min at the end of the analysis. The electron impact ion source was sustained at 200°C. Mass spectra were recorded at 70 keV. The mass spectra were obtained with a m/z range: 40–700 ultra-high resolution mode with an acquisition speed of 6 spectra/second. The identification of components was done in scan mode by using NIST11 and Wiley8 library and the target mass spectra obtained from sample were compared with mass spectra obtained from the library.

Biodegradation of crude oil was tested as described by Rahman et al. (2002) with slight changes in incubation period. Pre-cultured B. subtilis A1 culture was transferred (initial load about 2.1 × 10⁴ CFU ml^-1) to a 250 ml of Erlenmeyer screw cap flask, containing 100 ml of MSM with 1% (v/v) filter sterilized crude oil as carbon resource. An un-inoculated control flask was used for monitor abiotic loss of the crude oil substrate. The flasks were incubated at 37°C for 7 days at 200 rpm. Both the experiments were performed in triplicate. Triplicate flasks were recovered from both (inoculated and uninoculated control) systems for every day to measure the growth of total bacterial population, i.e., by conventional serial dilution method using pour plate technique with plate count agar (Himedia, Mumbai, India). For determination of enzyme activity, cells were harvested every day by centrifugation [refrigerated centrifuge (Remi-India: R-248)] at 6000 × g at 4°C for 10 min and then used for both enzyme assays.

Alkane hydroxylase activity during the biodegradation study was confirmed as described in Jauhari et al. (2014). In brief, the collected bacterial cells were rinsed twice and then re-suspended in 2 ml of 20 mM Tris–HCl buffer (pH = 7.4). Bacterial cells were disrupted using sonicator and centrifuged [refrigerated centrifuge (Remi-India: R-248)] at 6000 × g at 4°C for 10 min. The cell free supernatant was utilized for testing of alkane hydroxylase activity and absorbance was measured at 340 nm using UV–Vis spectrophotometer (JASCO V-630). 1 ml of testing solution contained 20 mM Tris–Hydrochloride and 0.15% CHAPS buffer (pH 7.4), 0.1 mM of Nicotinamide adenine dinucleotide (NADH), 10 μl of hexadecane mixture (1% hexadecane diluted with 80% DMSO) and 50 μl of crude extract in 1 ml quantity. The reaction was started by adding of 10 μl of hexadecane mixture. The activity of the alkane hydroxylase was expressed as 1 mmol of NADH oxidized per minute.

Alcohol dehydrogenase activity during the biodegradation was measured as mentioned in Jauhari et al. (2014). In brief, cell free supernatant was used for the assay and absorbance was measured at 340 nm using UV–Vis spectrophotometer. 1 ml of reaction solution contained 1 M of Tris– Hydrochloride buffer (pH 8.8), 4 mM of NAD⁺, 100 μl of ethanol (99% pure) and 50 μl of crude extract. Activity of the enzyme alcohol dehydrogenase was recorded as 1 mM of NADH formed per minute.

Biodegradation of crude oil hydrocarbons was examined by GC-MS analysis. After 7 days of incubation the remaining crude oil present in the culture flask was extracted twice with an equal volume of n-hexane (Adebusoye et al., 2007) and the solvent phase was dried in a vacuum oven at 60°C. 10 μl of resultant crude oil was dissolved in 990 μl of n-hexane. GC-MS model Perkin Elmer, clarus 680, Elite-5MS (30m × 0.25 mm ID × 0.25 μm) was used and 1 μl of sample was injected by split mode at 10:1 ratio. The carrier gas used was Helium, the flow rate was set at 1 ml min^-1 and the working temperature of the GC injector was 250°C. The gradient temperature was set as range from 60 to 300°C at a speed of 10°C min^-1, through an isothermal phase of 6 min at the end of the analysis. The mass spectra were obtained with an m/z range: 50–600 ultra-high resolution mode with an acquisition speed of 6 spectra/second. The identification of components was done in scan mode by using NIST08 library and the target mass spectra obtained from sample are compared with mass spectra obtained from the library. The biodegradation of crude oil hydrocarbon was expressed as the percentage (%) of crude oil degraded relative to the quantity of the remaining fractions in the suitable abiotic control samples. The biodegradation efficiency percentage (BE) based on the degradation of hydrocarbons, was calculated as described in Michaud et al. (2004) and Rajasekar et al. (2007). Changes in functional groups of crude oil hydrocarbon during biodegradation were characterized by FT-IR spectroscopy as described in the biosurfactant analysis section.

The biosurfactant production of the B. subtilis A1 was confirmed at the end of the repeated sub-culturing and screening methods and identified B. subtilis A1 as an excellent biosurfactant producer. In particular, the strain used in this study gave quick positive results for all biosurfactant screening methods. Specifically, drops collapsed within 30 s indicating higher amount of the biosurfactant present in the solution. Emulsification index was recorded as 76% for initial screening. The results are found consistent with the recent report by Freitas de Oliveira et al. (2013). Biosurfactants produced by different microorganisms are substrate specific, emulsifying diverse hydrocarbons at various rates (Ilori et al., 2005). The present results indicate that biosurfactant produced by B. subtilis A1 possess emulsifying activity. In the oil displacement test, a clear zone of ∼2.4 cm was visualized, followed by addition of surfactant solution in the crude oil layer. These results confirmed the presence of biosurfactant in the cell free culture supernatant. After the confirmation of biosurfactant synthesizing capability of the strain B. subtilis A1 culture condition was further optimized.

After the initial screening for biosurfactant producing capabilities of the bacterium, B. subtilis A1 was further subjected to optimization studies. Five different parameters were selected for optimization studies including pH, temperature and concentration of carbon and nitrogen sources. Figure 2 shows the optimal parameters obtained for biosurfactant production by strain A1 at the different conditions assessed (Korayem et al., 2015). The synthesis of biosurfactant level was reported in terms of emulsification index (E24%) and cellular activity reported as biomass of the bacterial cells.

Many physiochemical factors such as pH, temperature, growth conditions, and agitation have been shown to strongly influence microbial growth and metabolism (Khopade et al., 2012). Among them, pH of the production medium is a key factor for microbial growth. The optimum pH for the strain A1 was confirmed as 7.0 (E24: 70%), subsequently pH 8.0 showed a considerable effect (Figure 2A). Similarly, the role of temperature on biosurfactant production is presented in Figure 2B. The optimum temperature was confirmed as 40°C (E24: 76%). Strain A1 is mesophilic bacterium, which indicates this strain exhibits effective production level at moderate temperature (30–40°C).

As represented in the Figure 2C, eight carbon sources were screened for biosurfactant production. Among the carbon sources, sucrose was found the most favorable for strain A1 (E24: 78%) (Makkar and Cameotra, 1998; Khopade et al., 2012) followed by crude oil (E24: 76%). Similarly, the effect of the different nitrogen sources on biosurfactant production by strain A1 is presented in Figure 2D. Among the eight nitrogen sources, yeast extract showed the highest E24 value (68%) (Kiran et al., 2009; Khopade et al., 2012) followed by Urea (E24: 59%). All the optimized conditions were used to design the production medium. Utilizing the optimal substrate concentrations was essential to determine biosurfactant production. As demonstrated in Figures 2E,F, among the given 1–5% of the carbon and nitrogen sources, 2% of the sucrose (E24:.82%) and 3% of the yeast extract (E24: 84%) were found to be the optimum concentrations for biosurfactant production by strain A1. Due to the application of optimized conditions including pH, temperature and carbon and nitrogen sources, E24 values were gradually increased to maximum level in the substrates concentration optimization compared to individual optimization conditions.

Overall, optimized conditions were used for final biosurfactant production as described earlier. Biosurfactant produced by the strain A1 was measured as 4.85 g l^-1. This found to be maximum and comparable with other literature (Das and Mukherjee, 2007b; De Franca et al., 2015).

Fourier transform infrared spectrum (FT-IR) was recorded for the biosurfactant and revealed the functional groups present (Figure 3). The distinctive bands at 3138 cm^-1 designate the occurrence of -OH bonds (Aparna et al., 2012). The assimilation peak positioned at 1646 and 1168 cm^-1 states the existence of ester carbonyl groups (-C = O bond in -COOH) (Aparna et al., 2012). The presence of peaks at 2391 cm^-1 was likely due to the P-H₂ stretch of phosphines in phosphoserine (Bayoumi et al., 2010). The peak at 1406 cm^-1 corresponds to aliphatic chains (-CH₃, -CH₂-) of the fraction (Pornsunthorntawee et al., 2008). Medium peak was renowned at 970 cm^-1 shows that presences of O–H bend (carboxylic acids). The absorption peak at 637 cm^-1 specifies that the presence of -CH₂ group (Aparna et al., 2012). FT-IR spectra revealed a peak at 598 cm^-1 arising from C–I (Carbon–Iodine) bond. Based on this observation biosurfactant produced by B. subtilis A1 was categorized as lipopeptide in nature (Rodrigues et al., 2006).

The gas chromatography and mass spectrum determination further revealed (Figures 4A–C) that the biosurfactant extracted from B. subtilis A1 was a lipopeptide. Most of the compounds were fatty acids in nature such as hexadecanoic acid, methyl ester (Figure 4A) [retention time (RT): 23.14 & 25.48, MW: 270, chemical formula (CF): C₁₇H₃₄O₂)] (Kuyukina et al., 2001; Kiran et al., 2010), 9, 12-octadecadienoic acid (Z, Z)-, methyl ester (Figure 4B) (RT: 23.94, MW: 294, CF: C₁₉H₃₄O₂) (Sadouk et al., 2008), 9-octadecenoic acid, 12-hydroxy-, methyl ester (Figure 4C) (RT: 24.71 & 26.35, MW: 312, CF: C₁₉H₃₆O₃) (Akintunde et al., 2015). Deshmukh et al. (2012) summarized that the biosurfactant produced by B. subtilis was basically lipopeptide in nature.

Figure 5 illustrates the growth status of the isolates in the presence of crude oil as sole carbon source. Utilization of crude oil by biosurfactant producing bacteria was continuously monitored at the time of the biodegradation process. It was visible that the inoculation of B. subtilis A1 in MSM broth with crude oil as the sole carbon source turns the medium more turbid within 2 days of incubation. The turbidity of the medium was increased with incubation time. At the end of the incubation period the residual crude oil was recovered and used for further characterization to understand the degradation products. Different functional groups present in the residual crude oil were confirmed by FT-IR spectrum analysis. Both degraded crude oil spectrum and abiotic control spectrum are presented in Figure 6. Crude oil in abiotic control (Figure 6A), showed distinctive bands at 2922 and 2852 cm^-1 which belong to C–H aliphatic stretch, a strong peak at 1707 cm^-1 which is a C = C stretch in aromatic nuclei, medium peaks at 1455 and 1360 cm^-1 represents the C-H bend for alkanes, sharp and small peaks present at 1220 and 1092 cm^-1, respectively, correspond to C–N stretch aliphatic amines, presence of peaks at 898 and 745 cm^-1 is due to the presence of C–H “oop” 2° aromatics. On the other hand, the FT-IR spectrum of degraded crude oil with B. subtilis A1 (Figure 6B) showed a decrease in the intensity in bands at 1707, 1360, 1220, 1092, 898, and 745 cm^-1 which indicates degradation of the respective aliphatic and aromatic compounds present in the crude oil.

Further degraded sample and abiotic control samples were qualitatively analyzed by GC-MS and were compared in Figure 7. The best structural matches of GC retention data of crude oil and mass spectrum interpretation was presented in Table 1. Figure 7A shows the gas chromatogram of the abiotic crude oil samples. Higher peaks were present in the control and almost all the peaks from control chromatogram disappeared in the experimental samples inoculated with B. subtilis A1 (Figure 7B). Based on the primary observation it is confirmed that bacteria were capable of utilizing all these hydrocarbon components from the crude oil. The biodegradation efficiency (BE) of the crude oil in presence of B. subtilis A1 was calculated to be 87% which was achieved within 7 days. Based on the gas chromatogram analysis it is revealed that crude oil was used as a major carbon source at the hydrocarbon ranges between C₁₀–C₂₉. More accurately, this bacterial strain completely degraded some of the low MW compounds between C₁₀–C₁₄. Compounds with ranges of C₁₅–C₁₉ were degraded nearly 97%, other high molecular compounds are degraded about 78%. This shows that B. subtilis A1 has a high capability to degrade the different ranges of alkanes compounds present in the crude oil.

The alkane hydroxylase enzyme was induced in presence of B. subtilis A1 during the crude oil degradation (Figure 8). Activity of the alkane hydroxylase was increased with incubation period; the maximum activity was recorded as 188 μmol min^-1 mg^-1 protein at 3rd day. After that enzyme activity was slowly decreased with incubation period. This level of enzyme activity was much higher than reported in previous studies (Mishra and Singh, 2012). Alkane hydroxylase begins the degradation of alkanes by introducing the oxygen atoms at various sites of alkane terminus (Ji et al., 2013).

As reported by the Mishra and Singh (2012), activity of the alcohol dehydrogenase was not found as high as for alkane hydroxylase during the biodegradation. Activity of the alcohol dehydrogenase was gradually increased toward incubation period (Figure 8). The maximum enzyme activity 88 μmol min^-1 mg^-1 protein was attained at 3rd day of incubation as similarly recorded for alkane hydroxylase enzyme. After reaching the maximum activity then enzyme production was slowly declined as a decrease of the bacterial growth was observed as mentioned in the Figure 5. Similarly Pirog et al. (2010) also reported higher alkane hydroxylase activity compared to alcohol dehydrogenase during biodegradation of hexadecane by Rhodococcus erythropolis EK-1.