MATERIALS AND METHODS

Bacterial isolates: Three bacterial isolates were previously isolated from three PAH polluted sites in the Mediterranean Sea, Alexandria, Egypt, in January 2011. One isolate was able to degrade phenantherene as a sole carbon source and with an energy source up to 150 mg L^-1 and labeled as (P5). The other two isolates were able to degrade naphthalene and anthracene as a sole carbon source up to 450 and 110 mg L^-1 and were labeled as N7 and A2, respectively.

Bacterial cell immobilization
Inoculum preparation: The three bacterial isolates P5, N7 and A2 were cultivated at 30°C on a rotary shaker for 24 h in nutrient broth medium containing 0.15% (w/v) beef extract, 0.15% (w/v) yeast extract and 0.5% (w/v) NaCl (Konsoula and Liakopoulou-Kyriakides, 2006). Five mililiter of each overnight strain were used both for immobilization and the free cells experiments.

Immobilization of whole cells in calcium alginate: Immobilization was carried out under sterile conditions. 5 mL of overnight cultures were suspended in 12.5 mL of 2% (w/v) sodium alginate solution (Adinarayana et al., 2005). The mixture obtained was extruded drop-wise through a syringe into a 25 mL 3.5% (w/v) CaCl₂ solution. Alginate drops were solidified upon contact with CaCl₂, forming capsules and thus entrapping the bacterial cells. The capsules (mean volume and diameter 35 μL and 4 mm, respectively) were allowed to harden for 30 min and were then washed with a sterile saline solution (0.9% (w/v) NaCl) to remove any excess Ca⁺² ions and cells.

Immobilization of whole cells in agar-agar: A definite quantity of agar-agar (Difco) was dissolved in 18 mL of a 0.9% sodium chloride solution to get a final concentration of 2% and was sterilized by autoclaving (Adinarayana et al., 2005). The cell suspension (5 mL) was added to the molten agar-agar maintained at 40°C, shaken well for few seconds (without forming foam), poured into sterile flat bottom 4-inch Petri plates and allowed to solidify. The solidified agar block was cut into equal size cubes (4 mm³), added to a sterile 0.1 M phosphate buffer (pH 7.0) and kept in the refrigerator (1 h) for curing. After curing, the phosphate buffer was decanted and the cubes were washed with sterile distilled water three to four times.

Immobilization of whole cells in agarose: Entrapment of cells into agarose was performed by dissolving 1.2 g of agarose in a 20 mL physiological saline at 100°C, autoclaving was conducted and the solution was then cooled to 40-45°C (Adinarayana et al., 2005). Subsequently, the agarose solution was mixed with 5 mL cell suspension and poured into sterile flat bottom 4-inch Petri plates and allowed to solidify. The solidified agarose block was cut into equal size cubes (4 mm³) and kept in the refrigerator for 1 h.

Fermentation: Immobilized cells prepared by the above methods (beads and blocks) were added to 250 mL flasks containing 50 mL Bushnell-Haas (BH) broth medium supplemented with either 2% glucose or 50 mg L^-1 of tested PAH. For the free cell cultures 50 mL medium was inoculated with bacterial cells equivalent to those used in the immobilized cultures. Batch fermentations with free and immobilized cells were carried out at 30°C for ten days, on a rotary shaker (200 rpm). Samples were taken at regular intervals of two or three days and assayed for biosurfactant production.

Biosurfactant detection: The methylene blue analysis procedure was used for determining the concentrations of biosurfactants in the water samples (Pinzon and Ju, 2009). One mililiter of each sample was first adjusted to pH 2.3±0.2 using 1 N HCl. The acidified samples were then extracted with a five-fold volume of chloroform. Five mililiter of the chloroform extract was carefully removed and put in contact with a freshly prepared methylene blue solution containing 200 μL of methylene blue reagent (1 g L^-1 methylene blue in 10 mM borax buffer, pH 5.5±0.2) and 4.9 mL of distilled water. Fifteen μL of borax buffer (50 mM sodium tetraborate in distilled water, pH 10.5) was added to the mixture to adjust the pH to 8.6±0.2. The mixture was vigorously mixed for four min and the samples were left to stand for 15 min. The chloroform phase was transferred into a cuvette and the absorbance was measured at 638nm with a Shimadzu UV/Vis spectrophotometer (Model 1601) against chloroform blank. The absorbance values were converted to rhamnolipid concentrations using a calibration curve established by applying the same procedure to standard rhamnolipid solutions of different concentrations.

Bacterial genomic DNA extraction: The genomic DNA of each strain was isolated according to Griffiths et al. (2006). Cells were collected from overnight LB cultures by centrifugation and re-suspended in a 500 μL TE buffer. Twenty five microliter of 10 mg mL^-1 of lysozyme were added and the tubes were incubated at 37°C for 30 min, followed by the addition of 75 μL of 10% SDS and the tubes were inverted gently several times until complete lysis had occurred. Three μL of 20 mg mL^-1 of Proteinase K were added and the tubes were incubated at 37°C for 1 h. After incubation, 100 μL of NaCl (5M) were added followed by 800 μL of phenol/chloroform: isoamyl alcohol (24:1) and the tubes were inverted several times and then centrifuged for 10 min. The upper phase was transferred to a fresh sterile tube and extracted once with chloroform. The upper phase was again transferred to a fresh tube and a 0.7 volume of isopropanol was added and mixed gently and centrifuged for 10 min.

Table 1:	Code and sequence of the five different random RAPD primers

The supernatant was removed carefully and the pellets were washed with 1 mL of 70% ethanol. The pellets were collected by centrifugation for 5 min. The DNA was dried and dissolved in 100 μL TE buffer and stored at -20°C.

Fingerprint using random amplification of polymorphic DNA-polymerase chain reaction (RAPD-PCR) analysis: Fingerprinting using an RAPD-PCR technique was carried out according to (Jabbarzadeh et al., 2010). To detect the molecular variation among the bacterial isolates, P5, N7 and A2, five single arbitrary (10 base) primers were used for RAPD-PCR amplification; their names and sequences are summarized in Table 1. RAPD-PCR was carried out using a Ready-To-Go RAPD analysis kit (Amersham, GE Healthcare, UK). The amplification was performed using a thermocycler (PRIMUS 96) which was programmed as follows; initial denaturation step 95°C for 5 min followed by 35 cycles with 95°C for 1 min for DNA denaturation, annealing at 30°C for 1 min, extension at 72°C for 1 min and final extension at 72°C for 10 min, followed by cooling to 4°C. Visualization of amplified DNA fragments was carried out on 2.0% agarose gel in a 1X TBE buffer and stained with ethidium bromide. Moreover, electrophoresis was carried out for 20 min at 150 V and photographed using a gel documentation system (BIO-RAD GeL Doc, 2000).

Data analysis: Patterns of the studied genotypes using RAPD primers were scored as present (1), or absent (0) bands by using the Phoretix 1D image analysis system (Phoretix International, London) to integrate the data. Similarity indices were calculated and a consensus tree was developed based on the obtained banding patterns of the tested genotypes using the SPSS statistical analysis program (Version 10). The genetic relationships among tested genotypes were determined by constructing the phylogenetic tree.

Small-scale preparation of plasmid DNA: The plasmid DNA of each strain was isolated according to Griffiths et al. (2006) from overnight LB cultures (16 h). Cells were collected by centrifugation at 12.000 rpm for 30 sec and the supernatant was discarded. The pellets were resuspended in 100 μL of solution I (50 mM glucose, 25 mM Tris.Cl (pH 8.0) and 10 mM EDTA pH 8.0) by vigorous vortexing. Two hundred μL of freshly prepared solution II (0.2 N NaOH (freshly diluted from a 10 N stock and 1% SDS) were added and the contents were mixed by inverting the tubes several times. The tubes were stored on ice for 2 min and 150 μL of ice-cold solution III (60 mL of 5 M potassium acetate, 11.5 mL of glacial acetic acid and 28.5 mL of H₂O) were added. The tubes were gently inverted for 10 s to disperse solution III through the viscous bacterial lysate and the tubes were stored on ice for 3-5 min. The tubes were centrifuged at 12,000 rpm for 5 min and the supernatant was transferred to a fresh tube. The plasmid DNA was precipitated with 0.7 volume of isopropanol, washed with 70% cold ethanol, dried and dissolved in 50 μL of TE buffer (10 mM Tris base (pH 8.0) and 1 mM EDTA pH 8.0).

Amplification and sequencing of the 16S rRNA gene: The 16S rRNA gene was amplified by a Polymerase Chain Reaction (PCR) using specific universal primers (Poppert et al., 2005). The sequence for the 16S primer for 16S350F was 5'-AACTGGAGGAAGGTGGGGAT-3' and the sequence for the 16S primer for 16S350R was 5'-AGGAGGTGATCCAACCGCA-3'. The PCR mixture consisted of 30 picomoles of each primer, 10 ng of chromosomal DNA, 200 μM dNTPs and 2.5 units of Taq polymerase with 10 μL of polymerase buffer containing MgCl₂. The PCR was carried out for 30 cycles of 94°C for 1 min, 55°C for 1 min and 72°C for 2 min and final extension at 72°C for 10 min. After completion, a fraction of the PCR mixture was examined using 1.5% of agarose gel in a TBE buffer (25 g L^-1 Tris base, 27.5 g L^-1 boric acid, 20 mL of 0.5 M EDTA PH 8.0). Electrophoresis was carried out for 20 min at 150 V.

Sequence similarities and phylogenetic analysis: The Blast program (www.blast.ncbi.nlm.nih.gov) was used to assess the DNA similarities. Multiple sequence alignment and molecular phylogeny were performed using BioEdit software (Hall, 1999).

Statistical analysis: Analysis of variance was used to compare the data from the different treatments. All analyses were performed at p≤0.05 using minitab, version 13.1.

RESULTS AND DISCUSSION

Immobilization is a useful tool to study the ability of a specific microorganism to overproduce specific product at specific conditions. In this study, our bacterial isolates were able to grow and produce biosurfactants in different immobilization matrices. The preferred matrix was chosen according to the type of organism and the type of carbon source.

Production of biosurfactants with immobilized cells in various matrices by entrapment technique using glucose as substrate: Cell immobilization has become one of the most common techniques used for increasing overall cell concentration and productivity. The separation of products from immobilized cells is easier compared with suspended cell systems. In addition, immobilization of cells may allow continuous operation of cultivation processes at high dilution rates (Beshay et al., 2002). Furthermore, immobilization is a strategy for protecting cells from shear forces. Many different techniques for immobilizing cells have been proposed (Abd-El-Haleem et al., 2003).

The applicability of several polymeric natural or synthetic polymers as matrices for immobilization of viable cells motivated the study of the use of different gels such as alginate, agar-agar and agarose. The effects of these types of polymer supports used for immobilization on the biosurfactant production by free and immobilized cells are demonstrated in Fig. 1-3. The tested bacterial strains (P5, N7 and A2) whether free or immobilized were tested for biosurfactant production using BH broth medium supplemented with 2% glucose.

According to the results of this study, the rate of biosurfactant production was high in cases with immobilized cells compared with free cells. However, the three strains tested showed different production rates with a different preferable matrix. For example, P5 strain (Fig. 1) showed the highest biosurfactant production rate when immobilized with alginate (136 mg L^-1), followed by agar (110 mg L^-1) and then agarose (101 mg L^-1).

Fig. 1:	Effect of different immobilization matrices on biosurfactant production of P5 isolate grown at BH medium supplemented with 2% glucose

Fig. 2:	Effect of different immobilization matrices on biosurfactant production of N7 isolate grown at BH medium supplemented with 2% glucose

Fig. 3:	Effect of different immobilization matrices on biosurfactant production of A2 isolate grown at BH medium supplemented with 2% glucose

While in cases with the N7 isolate (Fig. 2), the highest biosurfactant concentration was obtained when the cells were immobilized with alginate (165.3 mg L^-1) followed by agarose (141 mg L^-1) and then agar gels (140 mg L^-1). The immobilization of the A2 strain with agarose gel displayed the highest production rate of biosurfactant (132.4 mg L^-1) compared to alginate and agar (97.4 and 70.6 mg L^-1, respectively). Therefore, it can be concluded that the alginate gel showed higher production efficiency than the other two gels. This might be due to the differences in the porous structures of the matrices which allow a better growth of the cells in the alginate matrix (Dias et al., 2000) and thus a higher production of biosurfactant.

Fig. 4:	Production of biosurfactant using different immobilization matrices using P5 isolate grown at BH medium+50 mg L-1 phenantherene

Fig. 5:	Production of biosurfactant using different immobilization matrices using N7 isolate grown at BH medium+50 mg L-1 naphthalene

Production of biosurfactants with immobilized cells in various matrices by entrapment technique using PAHs as substrate: The effect of different immobilization matrices on biosurfactant production was investigated in the presence of PAH as the sole source of carbon and energy. The tested bacterial isolates (P5, N7 and A2) were immobilized in different matrices including alginate, agarose and agar. The immobilized microorganisms were then grown in BH medium supplemented with 50 mg L^-1 of the tested hydrocarbons (anthracene, naphthalene and phenanthrene). The three bacterial isolates showed different production levels of biosurfactant according to the matrices used in cell immobilization (Fig. 4-6).

Fig. 6:	Production of biosurfactant using different immobilization matrices using A2 isolate grown at BH medium supplemented with 50 mg L-1 anthracene

In the case of agar immobilized cells, isolates P5 and A2 produced a high amount of biosurfactant (195.9 and 121 mg L^-1, respectively), whereas the free cells of isolate N7 produce a higher amount of biosurfactant (133.7 mg L^-1) compared with immobilized cells.

It is assumed that the high production level of biosurfactant obtained by the agar immobilized cells of isolates P5 and A2 may be due to the fact that the agar caves protect the bacterial cells from the toxicity of the PAH residue (Adinarayana et al., 2005). This kind of protection enables the bacterial cells to produce the biosurfactant that directly facilitates the availability of PAH to be consumed by the cells. We believe that agar immobilized cells of isolates P5 and A2 produced a high amount of biosurfactant when they utilized the phenanthrene and anthracene as the sole carbon source. However, a high production of biosurfactant was obtained when N7 bacterial cells were free and grown on naphthalene. This might be due to the fact that naphthalene is more readily available in the medium inoculated with free cells but this availability decreased when the cells were immobilized. This result is in harmony with the findings of other researchers who have suggested that immobilized biocatalysts produce lower levels of enzyme in comparison to free cells due to diffusion barriers and reduced oxygen availability for immobilized aerobic cells (Mamo and Gessesse, 1997; Konsula and Liakopoulou-Kyriakides, 2004)

Molecular identification of the three selected bacterial isolates
RAPD-PCR analysis: The RAPD-PCR technique was used to differentiate between the three selected isolates P5, A2 and N7 using five arbitrary primers. Different band patterns were observed with each primer (Fig. 7). Notably, high similarity was observed between A2 and N7. The RAPD results revealed that about 107 reproducible fragments were obtained and a total of 51 bands out the 107 were monomorphic (47.6%) and the rest were polymorphic. Similar studies were performed to study the genetic diversity between different bacterial species. For example, Istock et al. (2001) used RAPD-PCR for studying the geographical diversity of genomic lineages in Bacillus subtilis. In addition, Drake et al. (1996) used the single primer G1 that could distinguish 16 commercial strains of Lactobacillus helveticus by RAPD-PCR profiling. Moreover, Matar et al. (2009) used RAPD-PCR to study genetic diversity among 14 antagonistic Bacillus subtilis isolates obtained from different Egyptian sites. Josic et al. (2008) have also reported the ability of RAPD fingerprinting to differentiate between the two isolates of Lysinibacillus fusiformis obtained from Oil-Polluted Soil. In addition, Michelim et al. (2008) studied the molecular epidemiology of infection due to Proteus species using a variety of methods, including RAPD-PCR. They reported that the RAPD technique is one of the most suitable methods for local Proteus epidemiological studies. Moreover, Abdo-Hasan et al. (2008) have used the RAPD technique to differentiate between 40 isolates of Xanthomonas axonopodis pv. malvacearum the causal agent of cotton angular leaf spot disease. They demonstrated that the RAPD technique revealed high degrees of polymorphisms among the studied races. Pooler et al. (1996) used three PCR-based techniques which rely on different amplification priming strategies, RAPD, Repetitive Extragenic Palindromic (REP) and Enterobacterial Repetitive Intergenic Consensus (ERIC), to study the genetic relationships among 25 isolates of Xanthomonas fragariae from diverse geographic regions. In conclusion, unlike biochemical and serological methods, molecular tools for genotyping strains are more accurate in terms of identifying sub-lineages of microorganism (Kumar et al., 2007).

According to the data obtained using the RAPD-PCR technique, all the primers succeeded to distinguish between the three isolates. Data shown in Fig. 8 revealed that the three isolates were grouped into two main groups. The first group contained isolates A2 and N7 which are highly similar to each other. The second group contained only the P5 isolate which showed less similarity with the other two isolates. These results indicated that the P5 isolate belongs to a genus that is completely different from the genus of the A2 and N7 isolates.

Plasmid profile of the three bacterial isolates: Plasmids are extrachomosomal circular double stranded DNA found in most bacteria. Each bacterium may contain one or several plasmids.

Fig. 7:	RAPD-PCR banding patterns for the three bacterial genotypes using primer I, III, IV, V and VI, M: 1 kb ladder, P5, A2 and N7: Bacterial isolates

Fig. 8:	Consensus tree for three bacterial genotypes developed on the basis of their banding patterns with five RAPD primers

Plasmid profile analysis involves the study of the size and number of plasmids (Busch and Nitschko, 1999). This gives the size and number of plasmids present in the cells.

Fig. 9:	Plasmid profile of the isolates P5, A2 and N7, M: 1 kbp marker, P5, A2 and N7: Bacterial isolates

Some species may contain a variable number of plasmids and even unrelated bacteria may harbor a similar number of plasmids (Li et al., 2002). In this study, two different plasmids were detected in the three bacterial isolates P5, N7 and A2 (Fig. 9).

Fig. 10:	PCR amplification for the 16S rRNA gene from the three selected bacterial isolates degrading the polyaromatic hydrocarbons, M: 50 bp ladder

Accordingly, the estimated molecular sizes of the two plasmids were 2 and 1 kbp, respectively. It was noticed that, the two isolated molecular weight plasmids were presented in the three tested isolates.

Since, plasmid profile analysis is one of the methods used for differentiation between bacterial strains, many authors have investigated the ability of plasmid profile analysis to discriminate bacterial strains from each other. Appuhamy et al. (1998) have reported that plasmid profile analysis has added a further level of discrimination between Histophilus ovis strains and is a useful supplement to the PCR typing methods. In addition, it was demonstrated that plasmid profile typing is a useful method for the rapid differentiation of Salmonella enteritidis phage type 4 strains isolated from humans, poultry and eggs (Threlfall et al., 1994). The plasmid profile in this investigation was not suitable to discriminate between the three tested isolates. This might be due to the fact that the number of examined strains was not large enough to give a considerable variation.

Molecular identification of the selected bacterial isolates using 16S rRNA gene: For bacterial identification, the genomic DNA was subjected to PCR amplification using universal primers. These primers were able to amplify about 350 bp from the 16S rRNA gene (conservative region). The PCR product was separated on 1.5% agarose gel and data presented in Fig. 10 reveals that amplicone with a molecular size 350bp were observed with the three examined bacterial isolates.

The result of 16S rRNA gene alignment of the three isolates with the Gene Bank revealed the P5 isolate was identified as Lysinibacillus fusiformis with a similarity of 98%, whereas N7 and A2 were identified as Cronobacter turicensis with 99% similarity. The 16S rRNA sequences of the bacterial isolates were deposited in Gene Bank (NCBI).

Table 2:	Strain codes, strain name, accession number and similarity percentage of experimental isolates

The accession number of each isolate and percentage similarity to Gene Bank deposited strains are shown in Table 2. The bacterial isolate Lysinibacillus fusiformis strain had been previously isolated from oil polluted soil and showed an ability to degrade toluene, xylole and crude oil (Josic et al., 2008) in addition to other different hydrocarbons (Bento et al., 2003).

CONCLUSION

This study demonstrates that PAH degrading bacterial isolates can produce biosurfactant as a potent method to overcome bioavailability problem. In addition, the bacterial isolates prefer to be immobilized than to be free during the degradation process. In addition, the type of preferred immobilization matrix depends on the used carbon source; where, in general, calcium alginate was preferred at the presence of glucose while agar was preferred at the presence of hydrocarbons.