Production and characterization of rhamnolipid biosurfactant from waste frying coconut oil using a novel Pseudomonas aeruginosa D

Authors


Correspondence

Kochupurackal Jayachandran, School of Biosciences, Mahatma Gandhi University, Kottayam 686 560, Kerala, India. E-mail: jayansbs@gmail.com

Abstract

Aim

To improve biosurfactant production economics by the utilization of potential low-cost materials.

Methods and Results

In an attempt to utilize cost-effective carbon sources in the fermentative production of biosurfactants, various pure and waste frying oils were screened by a standard biosurfactant producing strain. Considering the regional significance, easy availability and the economical advantages, waste frying coconut oil was selected as the substrate for further studies. On isolation of more competent strains that could use waste frying coconut oil efficiently as a carbon source, six bacterial strains were isolated on cetyltrimethyl ammonium bromide–methylene blue agar plate, from a soil sample collected from the premises of a coconut oil mill. Among these, Pseudomonas aeruginosa D was selected as the potential producer of rhamnolipid. Spectrophotometric method, TLC, methylene blue active substance assay, drop collapse technique, surface tension measurement by Du Nouy ring method and emulsifying test confirmed the rhamnolipid producing ability of the selected strain and various process parameters were optimized for the production of maximum amount of biosurfactant. Rhamnolipid components purified and separated by ethyl acetate extraction, preparative silica gel column chromatography, HPLC and TLC were characterized by fast atom bombardment mass spectrometry as a mixture of dirhamnolipids and monorhamnolipids. The rhamnolipid homologues detected were Rha-Rha-C10-C10, Rha-C12-C10 and Rha-C10-C8/Rha-C8-C10.

Conclusion

These results indicated the possibility of waste frying coconut oil to be used as a very effective alternate substrate for the economic production of rhamnolipid by a newly isolated Ps. aeruginosa D.

Significance and Impact of the Study

Results of this study throws light on the alternate use of already used cooking oil as high-energy source for producing a high value product like rhamnolipid. This would provide options for the food industry other than the recycling and reuse of waste frying oils in cooking and also furthering the value of oil nuts.

Introduction

Surfactants, the surface-active agents, are compounds capable of reducing surface and interfacial tension between liquids, solids and gases. Most of the surfactants currently in use are chemically synthesized by organic chemical reactions. Synthetic surfactants cause environmental problems due to their resistance to biodegradation and toxicity to natural ecosystems. Increasing environmental awareness and emphasis on a sustainable society in harmony with the global environment have led to serious consideration of biological surfactants as possible alternatives to synthetic surfactants (Kim et al. 1999).

Microbial compounds that exhibit high surface activity and emulsifying activity are classified as biosurfactants. Biosurfactants offer several advantages over their chemical counterparts – such as their ecological acceptance, biodegradability, low toxicity, potentially high activities, effectiveness and stability at extremes of temperature, pH and salinity and possibility of production from renewable substrates. Pseudomonas is known to produce low molecular mass rhamnolipid biosurfactants with excellent surfactant properties.

Evaluation of the low-cost substrates with the right balance of carbohydrates and lipids to support optimal bacterial growth and biosurfactant production has resulted in the wide screening of suitable agricultural and industrial residues. The promising future of biosurfactants appears to specifically depend upon the use of abundant and low-cost raw materials and the optimization of the operational cultivation conditions to achieve high yields (Makkar et al. 2011). In a previous work conducted in our laboratory, the fermentative production of rhamnolipid biosurfactant from Pseudomonas aeruginosa MTCC 2297 was found to be maximum with orange peel as substrate (George and Jayachandran 2009). Utilization of such type of agricultural waste materials for biosurfactant production could alleviate the waste management problem, while addressing the economic issues related to production costs.

Great quantities of waste are also generated by the oil and fat industries. Apart from studies utilizing agro-industrial wastes as substrates for biosurfactant production, many researchers have used variety of vegetable oils and vegetable oil industry wastes for economical biosurfactant production. Another raw material associated with vegetable oil industry is residual cooking or frying oil, which is a major source of nutrient rich low-cost fermentative waste (Makkar et al. 2011). The process of frying starchy food caused the formation of potentially harmful amounts of many chemicals listed as probable carcinogens. The longer the food was cooked and the higher the temperature, the more of these harmful compounds produced. But these oils can be considered as high-energy sources for microbial growth and transformations into high value products like biological surfactants and thus can better manage these wastes by recycling. In this study, we utilized waste frying coconut oil as an alternate low-cost carbon source for the fermentative production of rhamnolipid biosurfactant. Incorporation of these types of cheaply available oils and oil wastes in the industrial production media might improve biosurfactant production economics, making them challenging targets for future R&D activities.

Materials and methods

Cost-effective carbon source and biosurfactant production media

Aliquots of liquid media for biosurfactant production were prepared from a basal solution having the following compositions per litre: KH2PO4 0·68 g, Na2HPO4 4·5 g, MgSO4·7H2O 0·1 g, Glycerol 30 g, NaNO3 6·5 g and Yeast Extract 0·5 g at a pH of 6·5 (Arino et al. 1996). Different oily substrates such as pure coconut oil, waste frying coconut oil, pure palm oil, waste frying palm oil and rubber seed oil (at 3% concentration in the place of glycerol) were tested for biosurfactant production by the standard strain Ps. aeruginosa MTCC 2297 (obtained from the Institute of Microbial Technology, Chandigarh, India). The waste frying oils were collected from a local bake house and filtered twice through a filtration assembly (containing a filter and muslin cloth) to remove solids. Vegetable oils were used as such without further purification. A 5% bacterial cell suspension of 1 OD concentration in 0·9% saline solution prepared from a 24-h culture in nutrient broth was used as inoculum (George and Jayachandran 2009). After inoculation and mixing, the flasks were incubated at 37°C under static condition for 96 h. All the experiments were conducted in three independent replicates, and controls were kept under similar conditions. Cells were removed from the culture by centrifugation at 10 000 rev min−1 for 20 min. Cell-free culture broth after deproteinization was used for analytical measurements. On the basis of maximum biosurfactant yield, the emulsification efficiency of the biosurfactant on kerosene and the availability, the most efficient carbon source was selected for further studies.

Micro-organism

Six bacterial strains were isolated from soil sample by serial dilution and plating from the premises of a coconut oil mill in Changanacherry, Kerala, India. Screening for biosurfactant producing strain was performed using methylene blue agar plates containing cetyltrimethyl ammonium bromide (CTAB) (Siegmund and Wagner 1991). The cultures were isolated and maintained on nutrient agar slants and were stored at 4°C. These isolates were tested for biosurfactant production under submerged fermentation using the selected cost-effective carbon source as substrate. The preparation of medium, process parameters for submerged fermentation and biosurfactant recovery steps were the same as described in previous section. The most efficient biosurfactant producing strain was selected for further studies on the basis of maximum biosurfactant yield on selected carbon source and the emulsification efficiency of the biosurfactant on kerosene. The selected bacterial strain was identified based on the morphological, biochemical and physiological characteristics according to Bergey's Manual of Systematic Bacteriology.

Optimization of the process parameters

The impact of various parameters influencing the production of biosurfactant by the selected strain under submerged fermentation using waste frying coconut oil was studied and optimized. For all these experiments, 100 ml of the medium in 250 ml flask was inoculated with 5% (v/v) of inoculum at 37°C under static condition. The effect of incubation time (1–10 days), pH (4·5–8) and waste frying oil concentration (1–5% concentration) on biosurfactant production was studied and the biosurfactant production was monitored at regular intervals of 24 h.

Rhamnolipid recovery

Produced rhamnolipid biosurfactant was recovered by solvent extraction. Cells were removed from the culture by centrifugation at 10 000 rev min−1 for 20 min. Cell-free culture broth was then deproteinized by heating at 110°C for 10 min. After cooling, it was acidified to pH 3·0 by the addition of 2 N HCl and the acidified supernatant was left overnight at 4°C for the complete precipitation of biosurfactant (Pansiripat et al. 2010). Rhamnolipids were separated by continuous extraction with equal amount of ethyl acetate in a separating funnel at room temperature. The ethyl acetate was completely evaporated from the sample using rotor evaporator at 40°C under reduced pressure. A partially purified viscous honey coloured rhamnolipid product was collected after solvent evaporation.

Rhamnolipid purification

Column chromatography

Liquid column chromatography was used for the separation of rhamnolipids. The polar lipids were separated in a 26 × 3·3 cm column containing 50 g of activated silica gel 60 – chloroform slurry. The column was loaded with a 5 g sample of crude rhamnolipid prepared in 10 ml CHCl3 and then the column was washed with chloroform until the neutral lipids were completely eluted. Then chloroform/methanol mobile phases were applied in sequence; 50 : 3 v/v (1000 ml), 50 : 5 v/v (200 ml) and 50 : 50 v/v (100 ml) at a flow rate of 1 ml min−1 and 20 ml fractions were collected (Sim et al. 1997; Sanchez et al. 2007). A final wash with 50 : 50 chloroform/methanol removed any remaining rhamnolipid from the column. Fractions were combined and solvent of active fractions containing biosurfactant were evaporated to dryness under vacuum with a rotor evaporator at 40°C under reduced pressure.

Thin-layer chromatography

The fractions from the liquid column chromatography were then subjected to preparative thin layer chromatography (TLC). The samples were dissolved in 1 ml CHCl3. One hundred microlitres of each sample was applied to a 20 × 20 silica gel TLC plate and developed in a chloroform/methanol/acetic acid (65 : 15 : 2 v/v/v) solvent system. The separated spots on these preparative TLC plates were carefully scrapped and collected separately. From these silica gel scrapings, the rhamnolipids were extracted thrice with 8 ml of CHCl3/CH3OH (1 : 2 v/v). The solvent scraping mixture was vortexed for 1 min, centrifuged down the silica gel for 10 min and pipetted off the solvent.

High-performance liquid chromatography

The purity of the separated components were tested by gradient elution high-performance liquid chromatography (Laboratory for polymer analysis, Sree Chitra Tirunal Institute for Medical Sciences and Technology, Thiruvananthapuram, India) using a Waters C 18 column (4·6 × 250 mm) with a Waters 717 plus auto sampler and 2487 refractive index detector. The flow rate was 0·4 ml min−1, and the mobile phase used was acetone/acetonitrile (30 : 70 v/v). The injected sample volume was 20·0 μl.

Rhamnolipid quantification

Rhamnolipid concentration in the cell-free culture broth was estimated by a calorimetric method for the determination of rhamnose concentration proposed by Chandrasekaran and Bemiller (1980). One ml of the cell-free culture broth was mixed with 4·5 ml of dilute sulphuric acid and heated at 100°C for 10 min. It was cooled to room temperature, mixed with 0·1 ml of freshly prepared 3% thioglycolic acid and was incubated in darkness for 3 h. Absorbance was measured at 420 nm, and the rhamnolipid concentration was calculated using a standard curve prepared using different concentrations of l-rhamnose. Rhamnolipid values were determined by multiplying rhamnose values by a coefficient of 3·4 obtained from the correlation [y = (0·0139 × −0·0058) × 0·68] of pure rhamnolipids/rhamnose (Benincasa et al. 2002, 2004; Benincasa and Accorsini 2008). As the rhamnose moiety represents only part of the rhamnolipid molecule, it is necessary to multiply the mass of rhamnose by a correction factor. This factor has been calculated as ranging between 3·0 (Itoh et al. 1971) and 3·4 (Benincasa et al. 2002). Note that the number is not exact as the rhamnolipid biosurfactant is not composed of a single molecule but rather of family of congeners that have different molecular masses (Monteiro et al. 2007; Neto et al. 2008). All experiments were conducted in three independent replicates.

Qualitative analytical methods for rhamnolipid production

The rhamnolipid biosurfactant production was qualitatively analysed by various techniques such as (i) thin-layer chromatography (George and Jayachandran 2009) (ii) methylene blue active substance (MBAS) assay (Hayashi 1975) (iii) drop collapse method (Bodour and Miller-Maier 1998) and (iv) measurement of emulsification activity (Nitschke et al. 2005). An indirect method used for the qualitative analysis of rhamnolipid production was the surface tension measurements. The surface tension of cell-free culture broth was measured by Du Nouy ring method (Harkins and Alexander 1959; Bodour and Miller-Maier 1998; Rahman et al. 2002; Nitschke et al. 2005; Rivera et al. 2007), and the instrument used was a KRUSS GmbH, Germany, model K100MKII tensiometer (Orbit Research Associates Pvt. Ltd, Delhi, India).

Characterization of rhamnolipid by fast atom bombardment mass spectrometry

The characterization of the separated rhamnolipid components was carried out by fast atom bombardment mass spectral (FABMS) analysis. The purified fractions obtained from the preparative TLC were then subjected to FABMS analysis. Fast atom bombardment mass spectra were recorded on a JEOL JMS 600 H mass spectrometer (Mass Spectrum Laboratory, National Institute for Interdisciplinary Science and Technology, Thiruvananthapuram, India). FAB positive ion mode was used, and the matrix used was 4-nitro benzyl alcohol.

Statistical analysis

The data represent the arithmetical averages of at least three replicates. One-way analysis of variance with post-test (P < 0·05) was performed with all the experimental results to test the significance of the mean values.

Results

Selection of cost-effective carbon source

The search for a regionally significant cost-effective substrate for extracellular rhamnolipid biosurfactant production, led to the evaluation of five oils, viz. coconut oil, palm oil, waste frying coconut oil, waste frying palm oil, and rubber seed oil as carbon sources for the submerged fermentation by the standard strain Ps. aeruginosa MTCC 2297. Among these oils tested as substrates for biosurfactant production, used coconut oil showed the higher rhamnolipid production as well as the higher emulsification efficiency on kerosene than the pure oils (Table 1).

Table 1. Rhamnolipid concentration and emulsification index of the biosurfactant produced from Pseudomonas aeruginosa MTCC 2297 with different oily substrates as the sole carbon source
SubstratesRhamnolipid concentrationa (g l−1)Emulsification indexa (EI) (%)
  1. a

    Estimated after 192 h of incubation.

  2. b

    Mean ± SD, = 4.

  3. c

    Selected substrate for further work.

Pure coconut oil0·54 ± 0·024b24·10
Pure palm oil1·02 ± 0·59642·17
Used coconut oilc2·26 ± 0·06762·35
Used palm oil0·20 ± 0·186 0
Rubber seed oil0·77 ± 0·4796·02

Screening, selection and identification of potential strain for rhamnolipid production

Six biosurfactant producing bacterial strains were isolated from soil sample collected from the premises of a coconut oil mill. The strains were isolated on blue agar plates containing CTAB and methylene blue as proposed and developed by Siegmund and Wagner (1991) for the specific screening of anionic biosurfactant producing bacterial strains. Extracellular anionic biosurfactants developed an insoluble ion pair with the cationic tenside CTAB, and the basic dye methylene blue with the productive bacterial colonies appearing as colonies surrounded by dark blue halos on the light blue agar plates (Fig. 1a,b). Six bacterial colonies producing dark blue halos against a light blue background were taken as biosurfactant producing strains and were designated as strain A, B, C, D, E and F. All these six strains and the standard strain Ps. aeruginosa MTCC 2297 were tested for biosurfactant production under submerged fermentation using the selected cost-effective carbon source as substrate. The newly isolated strain D was selected for the further studies with used coconut oil as the sole carbon source as it showed a comparatively higher biosurfactant production and emulsifying index (Table 2) and was identified as Ps. aeruginosa. We designated this strain as Ps. aeruginosa D.

Table 2. Rhamnolipid concentration and emulsification index of the biosurfactant produced from Pseudomonas aeruginosa MTCC 2297 and isolated strains with used coconut oil as the substrate
Bacterial strain designationRhamnolipid concentration (g l−1)Emulsification index (EI) (%)
  1. a

    Mean ± SD, = 4.

Ps. aeruginosa MTCC 22971·975 ± 0·007a60
Isolate A0·415 ± 0·035 0
Isolate B0·905 ± 0·41821·41
Isolate C0·615 ± 0·049 0
Isolate D3·550 ± 0·35471·7
Isolate E1·096 ± 0·05653
Isolate F0·440 ± 0·09932
Figure 1.

The halos exhibited by extracellular anionic biosurfactants producing bacterial colonies on the methylene blue agar plates.

Qualitative analysis for the production of rhamnolipid biosurfactant

All the qualitative methods used to detect biosurfactant production gave positive results and showed that the strain, Ps. aeruginosa D was able to grow and produce rhamnolipid when cultivated in the used coconut oil. On spraying with anisaldehyde reagent, blue spots indicative of carbohydrate units were detected in TLC plates loaded with crude extract of the biosurfactant produced by Ps. aeruginosa D using the substrate used coconut oil. When exposing the similar plates with iodine vapour, yellow spots indicative of lipids giving same Rf value as that of glycosyl units were obtained on the same region. Another spot with a lower Rf value 0·26 was seen with this sample, which also showed the presence of glycosyl and carbohydrate moiety. These results showed that Ps. aeruginosa D was able to grow and produce rhamnolipid biosurfactant when cultivated in the selected cost-effective carbon source. The optical density (655 nm) by MBAS assay corresponding to the anionic surfactant concentration in 100 μl cell-free culture broth was 0·292. The qualitative drop collapse method was carried out using cell-free culture broth produced by Ps. aeruginosa D. In this case, the droplet was collapsed, and the diameter of the drop was observed. The difference in the oil droplet diameter (2·5 mm) by drop collapse test was 4 mm. Surface tension measurements was used as an indirect measure of surfactant production and to evaluate the efficiency of the produced biosurfactant. The biosurfactant produced by Ps. aeruginosa D reduced the surface tension of culture broth, and the final surface tension reached from a value of 53 mN m−1 to a level up to 24·020 mN m−1. Percentage of decrease in the surface tension of the cell-free culture broth was calculated as 54·68%. The percentage of emulsification was checked with kerosene and the kerosene-emulsifying ability of rhamnolipid was found to be 71·7%. It was also observed that the culture broth containing the rhamnolipid biosurfactant was able to form stable emulsions for up to 1 month when kept at room temperature.

Optimization of process parameters for rhamnolipid production

The optimum incubation time for rhamnolipid production by Ps. aeruginosa D from used coconut oil was found to be 7 days. The optimum concentration of used frying coconut oil as sole carbon source for the maximum rhamnolipid yield was found to be 2%, and the best pH for maximum biosurfactant yield by Ps. aeruginosa D using used coconut oil was found to be 7.

Purification and characterization of rhamnolipid biosurfactant

Neutral lipids were eluted in the first few fractions and showed one major TLC spot at Rf value 0·96. Three main spots with Rf values 0·28, 0·37 (Fig. 2a) and 0·6 were detected on TLC from fractions 3–15 of 50 : 3 CHCl3/CH3OH elutions. The combined fractions of 50 : 25 chloroform/methanol elutions also gave the same spot having 0·6 Rf value on TLC detection. The other fractions of 50 : 5 and 50 : 50 elutions did not give prominent spots on TLC. The column loaded with extracted sample was compared with another column loaded with a blank sample. As explained before, in this case, the biosurfactant production medium with waste frying coconut oil and without bacterial inoculation kept under similar fermentation conditions was used as blank. The fractions with blank sample gave only neutral lipid spots on TLC analysis (Fig. 2b).

Figure 2.

(a) Thin-layer chromatogram of monorhamnolipid and dirhamnolipid (Rf values 0·28 and 0·37) produced by Pseudomonas aeruginosa D using used coconut oil as sole carbon source. (b) Thin-layer chromatogram of blank sample (medium with waste frying coconut oil and without inoculation) showing neutral lipids.

The mixture of two distinct rhamnolipid spots (Rf values 0·28 and 0·37) were then checked further for purity by HPLC. Two components were observed in retention times of 4·118 and 4·5 min (Fig. 3a). The HPLC chromatogram of the blank sample was shown for the comparison of this result (Fig. 3b).

Figure 3.

(a) HPLC chromatogram of the mixture of rhamnolipid spots (Rf values 0·28 and 0·37) using a Waters C 18 column (4·6 × 250 mm) with Waters 717 plus auto sampler and 2414 refractive index detector. (b) HPLC chromatogram of the blank sample using a Waters C 18 column (4·6 × 250 mm) with Waters 717 plus auto sampler and 2414 refractive index detector

The component with Rf value 0·28 corresponding a dirhamnolipid gave a molecular ion at m/z 674 (M + Na+) together with a characteristic fragment ion at m/z 528 (math formula, loss of terminal lipid) (Fig. 4). The monorhamnolipid with Rf value 0·37 gave a molecular ion at m/z 534 (M + Na+) and a fragment ion at m/z 381 (Fig 5). The component with Rf value 0·6 gave a molecular ion at m/z 496 (Fig. 6).

Figure 4.

Fast atom bombardment mass spectrum for Rha2-C10-C10 (m/z 674·20).

Figure 5.

Fast atom bombardment mass spectrum in the positive ion mode for Rha-C10-C12 (m/z 534).

Figure 6.

Fast atom bombardment mass spectrum in the positive ion mode for Rha-C8-C10 (m/z 496).

Discussion

Fermentation medium can represent almost 30% of the overall cost of a microbial fermentation, and the success of biosurfactant production depends on the development of cheaper processes and the use of low-cost raw materials (Rodrigues et al. 2006). In this study, we focused on the potential utilization of alternative low-cost substrates or waste materials as relatively inexpensive carbon sources, in the fermentative medium formulations for the cost-efficient production of biosurfactants. Biosurfactants, produced from cost-free by-products, have the twin advantages of waste minimization with economic potential biosurfactant production. Here, in our study, we evaluated the use of various oils as alternative carbon sources in the fermentative production of rhamnolipid biosurfactant by a standard bacterium Ps. aeruginosa MTCC 2297. Among these, the higher rhamnolipid production (2·26 g l−1) as well as the higher emulsification efficiency (62·35%) was shown by used coconut oil (Table 1). Considering its regional significance, easy availability and the economical advantages in the utilization of such a relatively inexpensive waste material as carbon source for biosurfactant production, used coconut oil was selected for further studies as an alternative substrate for the production of rhamnolipid biosurfactant. To the best of our knowledge, waste frying coconut oil as a cost-effective carbon source have not been reported anywhere in the available literature, for the production of biosurfactant.

In our previous publication (George and Jayachandran 2009), we reported that the standard bacterial strain Ps. aeruginosa MTCC 2297 was able to produce 9·18 g l−1 of rhamnolipid biosurfactant with a high emulsification ability of 73·3% using orange peel as the sole carbon source. In comparison with this result, the rhamnolipid productivity by the standard strain Ps. aeruginosa MTCC 2297 using used coconut oil was disappointingly too low (2·26 g l−1) such that the process of utilizing used coconut oil for rhamnolipid production was not so economic for large-scale production. To overcome this, we opted for the screening and isolation of competent bacterial strains that could use waste frying coconut oil more efficiently as carbon source for biosurfactant production. For this purpose and to isolate bacterial strains that produce improved amounts of rhamnolipids, our strategy was to isolate strains from the soil sample collected from the premises of a coconut oil mill and to subsequently screen those strains for rhamnolipid production using used coconut oil as the sole carbon source. Six bacterial strains were confirmed for the production of anionic biosurfactants as they represented dark blue halos against a light blue background on methylene blue agar plates (Fig. 1a,b) due to the specific reaction between the cationic CTAB-methylene blue complex and the anionic glycolipid biosurfactants (Perfumo et al. 2006). This method has been widely accepted for the screening of anionic biosurfactant producing micro-organisms (Perfumo et al. 2006; Razaa et al. 2007). The six biosurfactant producers, designated as strain A, B, C, D, E and F along with standard strain Ps. aeruginosa MTCC 2297 were screened for rhamnolipid productivity using waste frying coconut oil as carbon source. The new isolate strain D gave the highest rhamnolipid production (3·550 g l−1) and emulsifying activity (71·7%) with waste frying coconut oil as the sole carbon source (Table 2). The morphological and biochemical characteristics of strain D according to Bergey's Manual of Systematic Bacteriology (Buchanan and Gibbons 1974) confirmed its identity as a strain of Ps. aeruginosa and we designated it as Ps. aeruginosa D.

The results of various analytical techniques confirmed the ability of the strain Ps. aeruginosa D to grow and produce rhamnolipid biosurfactant when cultivated in waste frying coconut oil. The presence of both glycosyl units and lipid moieties on the same spots on TLC plates indicated that the biosurfactant produced by Ps. aeruginosa D using the substrate waste frying coconut oil was a glycolipid. Our result is in agreement with Rendell et al. (1990) who reported an anthrone positive spot on TLC plate having 0·24 Rf value as a dirhamnolipid. The presence and concentration of an anionic surfactant in the crude extract sample was confirmed by the modified version of MBAS assay as described by Hayashi (1975). The blue tint observed at the bottom layer of chloroform was formed from the formation of a complex between the anionic surfactant present in the crude extracts and the cationic dye methylene blue. The collapsed droplets of oil observed during the drop collapse technique also proved the ability of the selected Pseudomonas strain to produce biosurfactants. The increase in the diameter of drops may have been proportional to the amount of biosurfactant produced by the micro-organism.

Another indirect measure of surfactant production was surface tension measurement. It was also used to evaluate the efficiency of the produced biosurfactant. Reduction of surface tension observed in the culture broth of Ps. aeruginosa D indicated the production of surface-active compounds. The final surface tension of culture broth reached from a value of 53 mN m−1 to a level up to 24·020 mN m−1 (reduction of 28·98 mN m−1). For ease of use, we converted and expressed this reduction as a percentage of decrease in surface tension, and it was calculated as 54·68%. A similar higher reduction on the culture broth surface tension (a reduction of 20 mN m−1) was reported by Benincasa et al. (2002). The ability to form and stabilize emulsions is one of the most important features to be considered for the practical application of a surfactant (Nitschke et al. 2010). The emulsifying ability of the crude extracts was therefore a clear proof of the biosurfactant producing capability of the selected Ps. aeruginosa strains. The kerosene-emulsifying ability of rhamnolipid produced by Ps. aeruginosa D from used coconut oil was found to be 71·7%. It should be noted that the emulsifying activity was measured using a kerosene/water ratio of 6 : 4, which means that the kerosene phase constitutes 60% of the total volume. Therefore, E24 values ≥60% entail a complete emulsification of the kerosene phase (Abdel-Mawgoud et al. 2009). Such a strong emulsifying activity may be highly productive for environmental applications such as bioremediation and enhanced oil recovery applications in oil tank cleaning (Perfumo et al. 2006). Above all, the culture broth containing the rhamnolipid biosurfactant was able to form stable emulsion for up to 1 month, suggesting great potential for pharmaceutical and cosmetic industrial applications. From both Tables 1 and 2, it was clear that the emulsification index was not at all proportional to rhamnolipid concentration. In our previous study (George and Jayachandran 2009), we compared this observation with previous studies and reported that the variations observed in surface-active properties, such as the emulsifying ability of biosurfactants, were probably due to differences in the individual homologue concentration in the rhamnolipid biosurfactant.

The optimum incubation time for rhamnolipid production by Ps. aeruginosa D from used coconut oil was found to be 7 days. During this study, we observed a diauxic profile of surfactant accumulation and this had been reported elsewhere (Manresa et al. 1991; Mercade et al. 1993; Benincasa et al. 2002). The Rhamnolipid production started after 24 h of incubation and was completed in two stages. This result is in agreement with Abalos et al. (2001) and the diauxic profile of rhamnolipid accumulation and cell growth explained by them as due to the initial consumption of the most volatile fraction, accumulation of surface-active compound and emulsification of medium fatty acids to allow their assimilation by the cell and a subsequent stage of cell growth. Rhamnolipid production by Ps. aeruginosa D was studied with varying concentrations of waste frying coconut oil as the sole carbon source in the respective media. The optimum concentration of used frying coconut oil for the maximum rhamnolipid yield was found to be 2% with Ps. aeruginosa D and showed a general trend that rhamnolipid production initially increased with increasing carbon substrate concentration, until it reached a maximum value and then levelled off (Nitschke et al. 2010).

The optimum pH range of the rhamnolipid production medium for Ps. aeruginosa D was found to be 7. The biosurfactant production decreased with the increase in pH beyond the optimum value. A similar report (Guerra-Santos et al. 1984) stated that rhamnolipid production by Pseudomonas was at its maximum in a pH range from 6 to 6·5 and decreased sharply above 7.

During all experiments, the rhamnolipid production was best achieved through incubation of inoculated Erlenmeyer flasks at 37°C under static growth. It is in contrast with usual optimal conditions for Ps. aeruginosa, which includes rapid mechanical agitation at 37°C (Chen et al. 2007; Razab et al. 2007; Zeng et al. 2007; Wu et al. 2008). While studying rhamnolipid production by Ps. aeruginosa 44T1 on glucose, Robert et al. (1989) found that the best temperature for the product formation was 37°C. The optimal temperature for rhamnolipid production for Ps. aeruginosa J4 strain, reported by Wei et al. (2005) was in the range of 30–37°C. During the study on the effect of agitation rate on rhamnolipid production, the same authors suggested that the heavy foaming caused by the emulsification of rhamnolipid during vigorous shaking may decrease the transfer efficiency of oxygen gas into the liquid medium and is thus unsuitable for rhamnolipid production in shake-flask cultures. It is also reported that an increase in agitation speed caused a decrease in surfactant production by Nocardia erythropolis due to a shear rate effect on the growth kinetics of the micro-organism (Syldatk and Wagner 1987). Here, substitution of the shake-flask culture for better production did not give a significant positive difference in our result. Such static production conditions, therefore, as stated by Gunther et al. (2005), should save energy costs during the rhamnolipid production process.

In this study, the rhamnolipid mixture was recovered from the batch cultures by centrifugation followed by acid precipitation and solvent extraction. The organic solvent used for the extraction was ethyl acetate (Deziel et al. 1999; Abdel-Mawgoud et al. 2009; Sarachat et al. 2010). The resulting partially purified rhamnolipid product was a viscous honey coloured product (Wu et al. 2008; Pansiripat et al. 2010), and this partially purified product was further purified by different chromatographical methods. The distinct rhamnolipid spots on preparative TLC plates were then checked further for purity by high-performance liquid chromatographical analysis. The FABMS analysis gave confirmation to the presence and structure of the major rhamnolipid components.

The identity of the four main spots on TLC plates of the rhamnolipid mixture produced by Ps. aeruginosa D using waste frying coconut oil as substrate was evident from previously reported TLC results of rhamnolipid mixtures. The spot having 0·96 Rf value was neutral lipid, and the other three lower spots with Rf values 0·28, 0·37 (Fig. 2a) and 0·6 were detected as various rhamnolipid forms according to the previous reports. The less mobile homologues were the more polar ones that appeared as the lower spots. However, the more mobile spots were the less polar ones that appeared as the upper spots on TLC plates (Abdel-Mawgoud et al. 2009). Rf values 0·24 (Rendell et al. 1990), 0·20 (Thanomsub et al. 2007) 0·28 and 0·32 (Arino et al. 1996; Matsufuji et al. 1997) represented dirhamnolipids and the higher spot with Rf value 0·37 represented monorhamnolipid (Gunther et al. 2005). A TLC chromatogram of blank sample which was given for comparison gave only neutral lipids on analysis (Fig. 2b). The additional purity checking of combined lower fractions by HPLC gave compounds having retention times 4·118 and 4·5 min (Fig. 3a), which again provided evidence to the presence of dirhamnolipid and monorhamnolipid. This result is consistent with the data obtained in our previous work (George and Jayachandran 2009) and those provided by Sim et al. (1997) and Abalos et al. (2001). Mass spectrometric analysis of the component which gave 0·28 Rf value showed a signal at m/z 674·20 with a characteristic fragment ion at m/z 528 (Fig. 4) confirming the presence of the C10·C10 dirhamnolipid (Rha2C10C10) (Rendell et al. 1990). The monorhamnolipid with Rf value 0·37 gave a molecular ion at m/z 534 and a fragment ion at m/z 381 (Fig. 5) confirming the presence of a monorhamnolipid form. These data are consistent with the structure previously reported for the Rha-C12-C10 monorhamnolipid (Haba et al. 2003). Hence, the structural characterization of biosurfactants, produced by Ps. aeruginosa D from waste frying coconut oil, revealed by FABMS shows the presence of one dirhamnolipid Rha-Rha-C10-C10 and two monorhamnolipid forms Rha-C12-C10 and Rha-C10-C8/Rha-C8-C1 as major components.

Conclusions

A novel Ps. aeruginosa strain that could use waste frying coconut oil effectively as substrate for rhamnolipid biosurfactant production was isolated from soil collected from the premises of a coconut oil mill. The utilization of used/waste frying coconut oil in this study is truly significant as it is easily available, regionally significant and inexpensive waste material generated from food industry. The chemical composition of produced rhamnolipid, determined by FABMS, showed the product as a mixture of three rhamnolipid congeners which include both mono and dirhamnolipids. Incorporation of these types of abundant and cheaply available oils and oil wastes, in the industrial production media, might improve biosurfactant production economics, making them challenging targets for future R&D activities. Moreover, the application of used coconut oil as an alternate carbon source for biosurfactant production would provide an advantage for the development of the productive chain of the coconut, an important industrial segment in tropical coastal regions.

Acknowledgements

The authors are thankful to the School of Biosciences, Mahatma Gandhi University, Kottayam, Kerala, India for the facilities provided for this work.

Ancillary