WO2008037718A1 - Surface active compounds produced by gordonia strains, method of production and use thereof - Google Patents

Abstract

The present invention relates to a micro-organism belonging to the genius Gordonia able to produce and release extracellularly surface active compounds.

Description

Surface active compounds produced by Gordonia strains, method of production and use thereof

TECHNICAL FIELD OF THE INVENTION The present invention relates to surface active compounds (SAC) produced by three new bacterial strains belonging to the genus Gordonia, called BS25, BS29 and M22, the optimisation of their production and their use in various industrial activities such as oil, cosmetic, food and biomedical applications.

STATE OF THE ART

Surface active compounds interact with oil/water or air/water interfaces. They are amphiphilic molecules with both hydrophilic and hydrophobic (generally hydrocarbons) moieties that partition preferentially at the interface between fluid phases with different degrees of polarity and hydrogen bonding such as oil/water or air/water interfaces. SACs are classified by Neu (1996) in three different classes:

1. Biosurfactants: defined as low-molecular- weight SACs (e.g., glyco lipids and peptido lipids).

2. Amphiphilic polymers: defined as high-molecular- weight SACs with a hydrophobic region at one end of the molecule (e.g., lipopolysaccharides and lipoteichoic acids).

3. Polyphilic polymers: defined as high-molecular- weight SACs with hydrophobic groups distributed across the entire polymeric molecule (e.g. hydrophobic polysaccharides and emulsan).

SACs activities can be determined by measuring the changes in surface and interfacial tensions or stabilization of emulsions (Desai and Banat, 1997).

The surface tension (at the air/water interface) or interfacial tension (at oil/water interface) are used as a measure of efficiency of surface active compounds (Neu, 1996). Surface and interfacial tensions are measured with a tension-meter. The surface tension of distilled water is 72 mN/m, and addition of bio surfactant lowers this value to 30 mN/m (Desai and Banat, 1997). When a biosurfactant is added to air/water or oil/water systems at increasing concentrations, a reduction in surface tension is observed up to a critical level, above which amphiphilic molecules associate readily to form supra-molecular structures like micelles, bilayers, and vesicles. This value is known as the critical micelle concentration (CMC). CMC is commonly used to measure the efficiency of a biosurfactant (Neu, 1996). An emulsion is formed when one liquid phase is dispersed as microscopic droplets into another liquid continuous phase. The emulsification activity is the ability of the SAC to stabilize a water-oil emulsion (Cooper and Goldenberg, 1987).

The low molecular weight SACs reduce surface and interfacial tensions, whereas the high molecular weight SACs (also called bioemulsifϊers) are more effective at stabilizing oil-in- water emulsions (Rosenberg and Ron, 1999). The different properties of the two classes of SACs justify their different commercial applications. Surfactants constitute an important class of industrial chemicals widely used in almost every sector of modern industry. Most of the surfactants available on the market are chemically synthesised from petroleum.

SACs show a wide variety of chemical structures and surface properties; thus, they find application in many different industrial sectors. Many SACs exhibit activities equivalent to synthetic surfactants, but have the advantages of lower toxicity, higher biodegradability, higher foaming, high selectivity and specific activity at extreme temperatures, pH, and salinity, and of being produced from renewable and inexpensive substrates. These characteristics result in greater acceptability, especially in applications that cause the dispersion of SACs in the environment. Structurally different SACs can be produced by the same microbial strain in different fermentative conditions. Furthermore, some structural types of microbial SACs cannot easily be synthesized by chemical processes (Banat et al., 2000; Makkar and Cameotra, 2002).

Table 1. Properties, SAC-producing microorganisms and comparison with chemically synthesized surfactants (Desai and Banat, 1997; Christofi and Ivshina, 2002).

Pseudomonas sp 25-30 0.1-10 1

Trehalose lipids Rhodococcus 32-36 4 14 -17 erythropolis

Rhodococcus 26 15 <1 erythropolis

Mycobacterium sp. 38 0.3 15

Sophorose lipids Candida 33 82 1.8

(synonymous

Torulopsis) bombicola

Candida 30 0.9

(synonymous

Torulopsis) apicola

Lipopeptides and lipoproteins

Peptide lipids Bacillus 27 12 - 20 0.1 - 0.3 licheniformis

Serrawettin Serratia marcescens 28 - 33

Viscosin Pseudomonas 26.5 150 fluorescens

Surfactin Bacillus subtilis 27 - 32 23 -160 1

Fatty acid, neutral lipids, phospholipids fatty acid "Corynebacterium 30 150 2 lepus " neutral lipids "Nocardia 32 3 erythropolis " phospholipids Rhodococcus 30 <1 erythropolis

Low and high molecular weight SACs (biosurfactants) and/or microbial producers find application in the following technological sectors:

1) Oil industry a. Microbial enhanced oil recovery (MEOR) b. Oil transportation and pipelining c. Oil storage tank and oily equipment cleaning d. Emulsifiers

2) Remediation (hydrocarbon removal) a. Remediation of soils and aquifers b. Marine remediation

2') Remediation (heavy metal removal) 3) Mining

4) Agriculture

5) Food industry

6) Cosmetics 7) Biomedical and pharmaceutical applications a. Detergents, disinfectants or biocides b. Anti-adhesive biological coating c. Anti-thrombotic agents

Applications of low molecular weight SACs (biosurfactants) 1) Oil industry

Low molecular weight SACs (biosurfactants) and/or microbial producers find a wide range of applications in oil industry such as microbial enhanced oil recovery, oil transportation and pipelining, and oil storage tank cleaning (International patent application WO02062495).

Biosurfactants also find application within chemico -physical technologies for soil and aquifer remediation and in biological remediation processes (bioremediation) of soils, aquifers and marine ecosystems (Banat et al., 2000; Ron and Rosemberg; 2002; Christofi and Ivshina, 2002; Mulligan, 2005). Microbial enhanced oil recovery (MEOR)

MEOR is an area of considerable potential for biosurfactants and microbial producers. It is an important tertiary recovery technology, which utilizes microorganisms and/or their metabolites for residual oil recovery remaining in reservoirs after primary and secondary recovery procedures. The effects of microbial metabolism include gas and acid production, reduction in oil viscosity, plugging by biomass accumulation, and degradation of large organic molecules. Microorganisms also produce biosurfactants which reduce the interfacial tension at the oil-rock interface. This reduces the capillary forces preventing oil from moving through rock pores. Bacterial strains able to growth and to produce biosurfactants in the extreme conditions of the reservoir have been isolated. Different biosurfactants have been compared in laboratory study of MEOR (Table 2). Table 2. Effectiveness of biosurfactants used in microbial enhanced oil recovery (MEOR) (Banat et al, 2000).

SAC Time (h) Oil removal %

Control 114 81

Sophorose lipid 75 97

Rhamno lipid 77 94 trehalose-6,6'-dimycolate 71 93

Cello bio se lipid 79 99

Oil transportation and pipelining

Surfactants have been studied for use in reducing the viscosity of heavy oils, thereby facilitating transportation and pipelining (Banat et al., 2000).

Oil storage tank and oily equipment cleaning

Biosurfactants find applications in oil storage tank cleaning, in removing surface stains caused by petroleum products, in cleaning oily equipment, vehicles and tools (BT 100™) and in treating bilge and ballast water (Bertrand et al., 1994).

In a pilot field investigation, the ability of biosurfactant produced by a bacterial strain (Pet 1006) to clean oil storage tanks and to recover hydrocarbons from the emulsified sludge was tested. Two tonnes of bio surfactant-containing whole-cell culture were used to mobilize and clean 850 m³ oil sludge. Approximately 91% (774 m³) of this sludge was recovered as re-sellable crude oil and 76 m³ non-hydrocarbon materials remained as impurities to be manually cleaned. The value of the recovered crude covered the cost of the cleaning operation. Such a clean-up process is therefore economically rewarding and less hazardous to persons involved in the process compared to conventional processes (Banat et al., 1991). Emulsifiers

Biosurfactants have been used for production of stable hydrocarbon/water emulsifiers. The rhamno lipid produced by P. aerugimnosa has been patented as stabilizer for emulsions of high viscosity hydrocarbon (U.S. Patent 5,866,376).

2) Remediation (hydrocarbon removal) Remediation of soils and aquifers

Biosurfactants have been evaluated in bench and field-scale experimentations as substitutes of chemical synthesised surfactants to improve rate of oil removal in soils and aquifers. Biosurfactants can solubilize and mobilize organics adsorbed onto soil particles. Thus, they find applications within chemico -physical technologies, such as in situ soil flushing and ex situ soil washing for remediation of unsaturated zone, pump and treatment for aquifer remediation (Banat et al., 2000, Christofi and Ivshina, 2002; Mulligan, 2005). Biosurfactants also find applications in bioremediation technologies to improve the biodegradation rate of organic compounds in soils (Mulligan, 2005). In recent years, bioavailability is being considered one of the most important limiting factors of biodegradation rate in soil environments. Several studies have shown that the mass transfer from ab/adsorbed phase to liquid phase is the controlling mechanism of biodegradation rate (Weber, 1999; Banat et al., 2000, Christofi and Ivshina, 2002; Mulligan, 2005).

Biosurfactants enhance growth on bound substrates by desorbing them from surfaces or by increasing their apparent water solubility (Ron and Rosemberg; 2002). Surfactants help degradation by solubilising or emulsifying oils, by releasing hydrocarbons adsorbed to soil organic matter and by increasing the aqueous concentrations of hydrophobic compounds, thus resulting in higher mass transfer rates. It is evident in bioremediation studies that the correct biosurfactants and surfactant-producing microorganisms must be used to treat the particular pollutants (Banat et al., 2000; Christofi and Ivshina, 2002). Marine remediation In recent years, biosurfactants have been evaluated in bench and field-scale experimentations in order to enhance dispersion of oil in water (fragmentation up to solubilisation) and to remove oil from solid surfaces (e.g. sand, rock) during marine remediation processes (Bertrand et al., 1994; Banat et al., 2000; Christofi and Ivshina, 2002; Mulligan, 2005). Harvey et al. (1990) tested a biosurfactant from P. aeruginosa for its ability to remove oil from contaminated Alaskan gravel samples under various conditions, such as different concentrations of surfactant, time of contact, temperature of the wash and presence or absence of gum. They reported increased oil displacement (about 2-3 fold) in comparison to water alone. The necessary contact time for maximum effect was also reduced from 1.5- 2 min for water to 1 min. These results demonstrated the capacity of biosurfactants to remove oil from a naturally occurring substrate. The Environmental Technology Laboratory at the University of Alaska, Fairbanks, conducted a field trial in July 1993 in Sleepy Bay on LaTouche Island in Prince William Sound to test the effectiveness of a biosurfactant in removing weathered crude oil from subsurface beach material. They reported complete removal of diesel-range petroleum hydrocarbons (to the limit of 0.5 mg kg)l) while semi- volatile petroleum hydrocarbons were reduced to the 70% level, a removal of 30% (Tumeo et al. 1994). All of these studies are laboratory-based and successful bioremediation of exposed marine open sites using bio surfactants remains a challenge.

There are three main strategies for the use of bio surfactants in MEOR, in soil or marine bioremediation: i) Addition of bio surfactant-producing micro-organisms into a reservoir or to the soil. ii) Addition of selected nutrients into a reservoir or to the soil, thus stimulating the growth of indigenous biosurfactant-producing micro-organisms. iii) Addition of partially purified biosurfactants into a reservoir or to the soil.

2') Remediation (heavy metal removal)

Some biosurfactants bind to heavy metals decreasing their toxicity and enhancing metal desorption from solid surfaces (e.g. sand, soil particles, see Table 3).

The potential of rhamno lipid biosurfactants produced by P. aeruginosa in the removal of cadmium, zinc and lead from contaminated soils was investigated (Tan et al., 1994; Herman et al., 1995; Sandrin et al., 2000).

Surfactants can remove metals from surfaces in a number of ways. First, metals in a nonionic form can complex with biosurfactants, enhancing surface removal by Le

Chatelier's Principle Additionally, the use of anionic surfactants which contact metals can lead to their desorption from surfaces. The surfactant-metal union would then need to be removed from the soil matrix. Cationic surfactants can also act to reduce the association of metals by competition for some but not all negatively charged surfaces (Miller, 1995; Christofϊ and Ivshina, 2002).

Table 3. Metal extraction using biosurfactants (Christofi and Ivshina, 2002).

SAC Zn removal Cu removal

Rhamno lipid, 12% 19.5 35.1

Rhamnolipid, 0 .5% (1 .0% NaOH) 5.6 25.1

Rhamnolipid, 2% (1 .0% NaOH) 2.8 28.3

Sophorose lipids, 4% (0 .7% HCl) 15.8 37.2

Sophorose lipids, 4% (1 .0% NaOH) 7.3 36.0

Surfactin 0 .25% (1 .0% NaOH) 6.0 24.8

NaOH only 2 5

Water only Negligible Negligible Thanks to their properties, future applications of biosurfactants as enhancers in chemico- physical and biological remediation of heavy metal and radioactive contaminants can be envisaged.

3) Mining Kao Chemical Corporation (Japan) used Pseudomonas, Corynebacterium, Nocardia, Arthrobacter, Bacillus and Alcaligenes to produce biosurfactants for the stabilization of coal slurries to aid the transportation of coal (Kao 1984. Australian Patent 8317-8555). Biosurfactants, produced by Candida bombicola, have been tested for solubilisation of coal and achieved partial solubilisation of North Dakota Beulah Zap lignite coal (Banat et al., 2000).

4) Agriculture

Biosurfactant-producing bacteria and microbial surfactants are already exploited as bio fungicide (Serenade^®; ZONIX™: Stanghellini et al., 1998) in the control of main diseases of economically important crop plant pathogens (Thimon et al., 1995; Banat et al., 2000). The proposed mechanism for the lytic activity against zoosporic plant pathogens (Pythium, Phytophthora, Plasmopara) of rhamnolipid is the loss of membrane integrity upon exposure to the biosurfactants (Stanghellini and Miller, 1997).

5) Food industry

In the food industry, biosurfactants are used as emulsifϊers for the processing of raw materials. Emulsifϊcation plays an important role in forming the right consistency and texture as well as in phase dispersion. Applications of SACs are in bakery and meat products, where they influence the rheological characteristics of flour and the emulsifϊcation of partially broken fat tissue (Banat et al., 2000).

6) Cosmetics Biosurfactants have found a niche in the personal care market because of their low moisturizing properties and skin compatibility (Banat et al., 2000). Sophorose lipids, produced by yeast belonging to the genus Candida (such as C bombicola and C apicola), are the most exploited biosurfactants in food industry. Kao Chemical Corporation at present uses sophorose lipids commercially as humectants for cosmetic makeup brands such as Sofϊna. There are currently patents for the use of rhamno lipids to make liposomes and emulsions, both important in the cosmetic industry (Ishigami et al. 1988). Threalolipids produced by genus Rhodococcus are suited for application in cream, pastes, sticks and films (Lang and Philip, 1998). 7) Biomedical and pharmaceutical applications

Some microbial surfactants have been described for their potential to act as biologically active compounds, having activities such as antiviral activity against a wide spectrum of DNA and RNA viruses, antibacterial, antimycoplasma, antimycobacterial activities and antifungal activity against important human {Candida albicans) and plant pathogens (Pythium, Phytophthora, Plasmopara). Biologically active bio surfactants belong to different structural types: lipopeptides (iturin, surfactin), glyco lipids (rhamno lipid, mannosylerythritol lipid, treahalose lipid), cyclic despeptides (viscosinamide) (Stanghellini and Miller, 1997; Singh and Cameotra, 2004; Rodrigues et al, 2006). The inactivation by surfactin of enveloped viruses, in particular herpes viruses and retroviruses, is significantly more efficient than that of non-enveloped viruses, suggesting that the antiviral action of surfactant is primarily due to a physicochemical interaction between the membrane-active surfactant and the outer part of the virus lipid membrane bilayer (Vollenbroich et al., 1997). Detergents, disinfectants or biocides

In the detergent and cleaning industries it is important to remove fatty materials. Biosurfactants have three major advantages for these applications: the environmental consequences of their use are minimal and their addition can reduce the concentration of harmful chemical detergents needed. In addition, the biosurfactants are compatible with the variety of enzymes that are used in the biological-detergents, which are often inactivated by the chemical detergents. Biosurfactants can also be used as substitutes for chlorinated solvents for cleaning electronic boards, cutting devices and delicate instruments that can be damaged by standard detergents (Rosenberg and Ron, 1999). Thanks to their antimicrobial and antiviral activities, future applications of biosurfactants as disinfectants (or biocides) in industrial, domestic, veterinarian sectors, in food industry and in personal care can be envisaged. Anti-adhesive biological coating

Biosurfactants have been found to inhibit the adhesion of pathogenic organisms to solid surfaces or to infection sites, limiting bio film formation or breaking down existing bio films (Kuiper et al., 2004). Swarming motility and biofilm formation are the key actions in the colonization of a surface by bacteria and favour nosocomial infections. The pre-coating of catheters with surfactin showed to decrease the amount of biofilm formed by Salmonella typhimurium, Salmonella enterica, Escherichia coli and Proteus mirabilis (Mireles et al.,

2001).

The lactobacilli species hold promise for application in many human sites where the pathogens attach, colonize and confer disease, including the urinary tract (Boris and Barbes, 2000). There are reports of inhibition of biofϊlm formed by uropathogens and yeast on silicone rubber by biosurfactants produced by L. acidophilus (Velraeds et al., 1996;

Busscher et al.,1997; Velraeds et al., 1998; Reid G., 2000).

Biosurfactants have also several promising applications in the food industry as anti- adhesives. In order to limit the microbial contamination in the finished products, the common disinfection procedures are combined with preventive measures such as bacterial adhesion reduction. It has been recently observed that the precoating of stainless steel surfaces with a biosurfactant produced by Pseudomonas fluorescens significantly inhibited the adhesion of Listeria monocytogenes (Meylheuc et al., 2001).

Thanks to their activities, future applications of biosurfactants in biological anti-adhesive coating (e.g catheters and medical equipments) in order to prevent microbial colonisation can be envisaged.

Anti-thrombotic agents

Surfactins produced by Bacillus spp show several biological activities, such as the inhibition of fibrin clot formation, the induction of ion channel formation in lipid bilayer membranes, the inhibition of cyclic AMP, the selective inhibition of platelet and spleen cytosolic phospholipase A2 (PLA2) and anti- inflammatory activity (Kim et al., 1998).

Surfactin C is able to enhance the activation of prourokinase (plasminogen activator) and the conformational change in plasminogen, leading to increased fibrinolysis both in vitro and in vivo (Kikuchi and Hasumi, 2002). The plasminogen-plasmin system is involved in blood clot dissolution as well as in a variety of physiological and pathological processes requiring localized proteolysis. In a rat pulmonary embolism model, surfactin C increased plasma clot lysis when injected in combination with pro-urokinase (Kikuchi and Hasumi, 2003). More recently, Lim et al.

(2005) have shown that some iso forms of surfactin (produced by B. subtilis) have an anti- thrombotic activity and are able to prevent the platelet aggregation leading to an inhibition of additional fibrin clot formation and to enhance fibrinolysis, thus facilitating the diffusion of fibrinolytic agents.

Applications of high molecular weight SACs (bioemulsifiers) The class of high molecular weight SACs, also called bioemulsifϊers coat the droplets of oil, thereby forming very stable emulsions or dispersions that never coalesce. Since the emulsifier adheres to the oil, it is concentrated in the oil/water interphase even when the solvent is replaced. As a result, it stays with the oil when the water is removed. Consequently the emulsions are stable even in very dilute solutions and the concentration of the product that has to be added is relatively low (1 :50-1 :1000). In addition, the emulsifier is present in only very low concentrations in the water, allowing the water to be recycled (Rosenberg and Ron, 1997 and 1999). The potential commercial applications of bioemulsifϊers include (Rosenberg and Ron, 1997): bioremediation of oil polluted soil and water, enhanced oil recovery, formulation of pesticides and herbicides, replacements of chlorinated solvents use in cleaning up oil contaminated pipes, vessels and machinery used in the detergent industry (Rosenberg and Ron, 1998; JP 62.2970397) and the formation of stable oil- in- water emulsions for the food and cosmetics industries (Klekner and Kosaric, 1993; Patents: EP 242.296, DE 3.610.384, JP 62.286914, JP62289508).

1) Oil industry

Oil transportation and pipelining

Surfactants have been studied for use in reducing the viscosity of heavy oils, thereby facilitating transportation and pipelining (Banat et al, 2000). Hayes et al. (1986) have demonstrated the ability of Emulsan, produced by Acinetobacter, to reduce the viscosity of Boscan (Venezuelan heavy oil) from 200,000 to 100 cP, making it feasible to pump heavy oil in 26,000 miles of commercial pipe line.

2) Remediation (hydrocarbon removal)

Bioemulsifϊers found potential commercial applications in bioremediation of oil polluted soil and water (Volkering et al., 1997). Stable emulsions of crude oil, produced by bioemulsifϊers, can easily be diluted in large volumes of water reducing the ratio of crude oil to mineral nutrients and thus speeding up further biodegradation.

High molecular weight Emulsan^® has been commercialised for the removal of water from the emulsions prior to processing, and in the release of bitumen from tar sands (Anon, 1984; Mulligan, 2005).

Alasan increases the apparent aqueous solubility of polycyclic aromatic hydrocarbons (PAHs) and enhances their biodegradation rate. Thanks to their activities, future applications of Alasan as enhancer in bioremediation can be now envisaged (Barkey et al, 1999).

Bioemulsifiers stabilise emulsions of immiscible organophosphorous pesticides (Patel and Gopinathan, 1986) and organo-chlorides (Appaiah et al, 1991). Thanks to their activities, future applications of bioemulsifiers in formulation and treating of pesticides and herbicides can be now envisaged.

2') Remediation (heavy metal removal)

Many bacterial polysaccharides have been shown to bind heavy metals. Emulsan, produced by Acinetobacter Iwoffii RAG-I, forms stable oil- in- water emulsions. In this system, metal ions bind primarily at the oil/water interphase, enabling their recovery and concentration from relatively dilute solutions. Cations bound to the emulsion can be completely removed to the water phase when pH was lowered. Thanks to their activities, future applications of bioemulsifiers in remediation of heavy metal and radionuclides and in heavy metal recovery from dilute solutions can be now envisaged (Gutnick and Bach, 2000). 3) Mining

Biodispersan, produced by Acinetobacter calcoaceticus A2, is an anionic polysaccharide which preventes flocculation, and disperses limestone in water mixture. Biodispersan catalyzes the fracturing of limestone into smaller particles. This property make it potentially useful in the paper, ceramics, paint and textile industries (Rosenberg and Ron, 1997).

4) Food industry

Chemically synthesized emulsifϊers are gradually losing favour due to increased pressure from consumers to reduce the use of artificially synthesized additives in foods. Thus, an increasing consciousness among consumers is driving a steady increase in demand for more natural food ingredients and additives. Yeasts belonging to Candida utilis, a food- grade organism, have been characerised for their ability to produce bioemulsifiers. Preliminary trials showed that the carbohydrate-based bioemulsifϊer from C. utilis had potential for use as food ingredient (e.g. salad cream) (Shepherd et al., 1995). In dairy products, such as soft cheeses and ice creams, the addition of polymeric emulsifϊers improves the texture and creaminess. This quality is of special value for low-fat products (Rosenberg and Ron, 1999).

A novel polysaccharide emulsifier produced by Klebsiella shows potent inhibition of the autooxidation of soybean oil (Kawaguche et al, 1996). Structures and properties of low molecular weight SACs (biosurfactants)

RHAMNOLIPIDS

Certain species of Pseudomonas are known to produce large quantities of a glyco lipid consisting of 2 mol rhamnose and 2 mol β-hydroxydecanoic acid. The hydroxyl group of one of the acids is involved in glycosidic linkage with the reducing end of the rhamnose disaccharide, whereas the hydroxyl group of the second acid is involved in ester formation. Since one of the carboxylic groups is free, the rhamno lipids are anions above pH 4. The pure rhamno lipid lowered the interfacial tension against n-hexadecane to about 1 mN/m and had a CMC of 10-30 mg/1, depending on the pH and salt conditions (Rosenberg and Ron, 1999).

SOPHOROSE LIPIDS

Different species of the yeast Candida (synonymous Torulopsis) produce extracellular sophorose lipids, which consist of two glucose units linked β-1,2. The 6- and 6'-hydroxyl groups are generally acetylated. The lipid portion is connected to the reducing end through a glycosidic linkage. The terminal carboxyl group of the fatty acid can be in the lactonic form or hydrolyzed to generate an anionic surfactant. The sophorose lipids lower surface and interfacial tensions, although they are not effective emulsifying agents. The pure lactonic sophorose lipid (10 mg/1) lowered the interfacial tension between n-hexadecane and water from 40 mN/m to about 5 mN/m, relatively independently of pH (6-9), salt concentration and temperature (20-90 ⁰C) (Rosenberg and Ron, 1999). LIPOPEPTIDES

Bacillus subtilis produces a cyclic lipopeptide called surfactin or subtilisin, which has been reported to be the most active biosurfactant that has been discovered to date. Surfactin has a CMC in water of 25 mg/1 and lowers the surface tension to 27 mN/m. The minimum interfacial tension against hexadecane is 1 mN/m. The synthesis of one or more peptide antibiotic during the early stages of sporulation is common to most, if not all, members of the genus Bacillus (Rosenberg and Ron, 1999).

Structures and properties of high molecular weight SACs (bioemulsifiers) Bioemulsifiers are produced by a wide diversity of Bacteria (Gram-positive and Gram- negative) and Archaea. Most of the emulsifϊers are composted by mixtures of hydrophobic and hydrophilic polymers. The most extensively studied are the bioemulsifiers produced by different species of the genus Acinetobacter (Rosenberg and Ron 1998 and 1999). Emulsan, produced by the strain Acinetobacter Iwqffii RAG-I (previously Acinetobacter calcoaceticus), is a complex of an anionic heteropolysaccharide and protein. Emulsan is an effective emulsifier at low concentrations (0.01%-0.001%), representing emulsifϊer-to- hydrocarbon ratios of 1 :100-1 :1000, and exhibits considerable substrate specificity. Emulsan does not emulsify pure aliphatic, aromatic, or cyclic hydrocarbons; however, all mixtures that contain an appropriate mixture of an aliphatic and an aromatic (or cyclic) alkanes are emulsified efficiently. Maximum emulsifying activity is obtained in the presence of divalent cations (2-10 mM Mg²⁺) at pH values of 5.0-7.5. Alasan, produced by a strain of Acinetobacter radioresistens, is a complex of an anionic polysaccharide and proteins. However, the polysaccharide component of alasan is unusual in that it contains covalently bound alanine. The protein component of alasan appears to play an important role in both the structure and activity of the complex. One of the Alasan proteins, with an apparent molecular mass of 45 KDa, was purified and shown to constitute most of the emulsifying activity (Toren et al, 2002). SAC in genus Rhodococcus

Rhodococcus and Gordonia are phylogenetically closely related genera belonging to the Corynebacterineae suborder, the my colic acid-containing group within the Actinomycetales order (Stackebrandt et al., 1997). In Corynebacterineae, the SAC production has been investigated thoroughly in the Rhodococcus genus. Very little is known about the SACs produced by the Gordonia members (Lang and Philip, 1998).

Rhodococci produce low molecular weight SACs. Most of the biosurfactants produced by rhodococci are nonionic trehalose lipids which consist of a trehalose residue linked by an ester bond to different long chain fatty acids. Among the latter, α-branched β-hydroxy acids are the most frequent esterified groups. Few strains have been characterised which are able to produce anionic trehalose lipids (Table 4; Lang and Philip, 1998).

Biosurfactants produced by rhodococci lower the surface and interfacial tensions; they also stabilise oil- water emulsions. Various types and quantities of trehalose lipids are produced by the same bacteria strain depending on growth conditions, e.g. cell growth associated production, production under growth limiting conditions and production by resting cells. Members of the genus Rhodococcus produce biosurfactants in response to the presence of water-insoluble compounds, such as hydrocarbons, in the liquid culture medium. Trehalose lipids have been proposed to have a role in the uptake of water insoluble hydrocarbons. Essentially, all of the trehalose lipids produced by rhodococci are cell-bound (extractable with n-hexane) (Lang and Philip, 1998; Rosenberg and Ron, 1999). This is the most important limitation to an industrial exploitation of bio surfactants produced by rhodococci as the majority of the product are not be found in the culture medium. The extraction of trehalose lipids from the bacterial cells causes an increase in the downstream process costs (Lang and Philip, 1998).

Table 4. Surfactant properties of SACs produced by Rhodococcus strains (Lang and Philip,

water; water/n-hexadecane; water/kerosene; water/n-decane; n.d. = not determined Production of surface-active exopolysaccharides by Rhodococcus strains have been reported. The surface-active properties have been attributed to the presence of deoxy sugars and O-acetyl groups (Neu 1996). SAC in genus Gordonia

Gordonia was recently recognised as an emerging genus in industrial and environmental biotechnologies (Arenkotter et al. 2004). Most species were isolated due to their abilities to degrade xenobiotics, environmental pollutants, or otherwise slowly biodegradable natural polymers (e.g. natural and synthetic rubbers) as well as to perform biodesulfurization of fuels and to synthesize useful compounds (e.g. carotenoids, imidazol-2-yl amino acids, polysaccharide). The variety of chemical compounds being transformed, biodegraded, and synthesized by gordoniae makes these bacteria potentially useful for environmental and industrial biotechnology. Few Gordonia strains have been found to produce SACs and very little information is currently available about the SACs produced by the Gordonia members (Arenkotter et al. 2004). Most of the studies have been carried out on Gordonia amarae (Arenkotter et al. 2004). This species is commonly found in foaming-activated sludge wastewater treatment plants, where both cells and their extracellular biosurfactants participate in the formation of stable foams (Iwahori et al., 2001; Pagilla et al., 2002).

Biosurfactants produced by G. amarae (previously Nocardia amarae) have found potential application in the removal and recovery of non- ionic organics from aqueous solutions (Sutton, 1992). Iwahori et al. (2001) have demonstrated that the G. amarae SCl is able to produce bioemulsifϊers which form stable oil-in-water emulsions. However, the emulsification activity (determined as E24% parameter) has not been measured. Pagilla et al. (2002) have characterised the surface properties of SACs produced by G. amarae. G. amarae produces biosurfactants which lower the surface tension up to 40 mN/m. Furthermore, the biosurfactant yield is 5 x CMC when the bacterial strain is grown on both n-hexadecane and acetate, as carbon sources.

To the best of our knowledge, the Gordonia sp. 321 strain is the first SAC-producer belonging to the genus Gordonia isolated from a hydrocarbon contaminated ecosystem (Nazina et al., 2003). When Gordonia sp. 321 is grown on liquid paraffins, culture broths significantly decrease the surface tension (up to 35 mN/m) and the interfacial tension. Furthermore, the culture broth also produces stable emulsions (E24% equal to 20%). There is no information about the SAC localisation in this strain (e.g. extracellular, cell-bound). At present, the chemical structures of SACs produced by Gordonia members have not been determined.

Catabolic abilities in genus Gordonia

Several Gordonia strains degrade and/or transform hydrocarbons, xenobiotics, environmental pollutants, or otherwise slowly biodegradable natural polymers (e.g. natural and synthetic rubbers) (Table 5). Thus, Gordonia are powerful candidates for bioremediation, biocatalysis and bioconversion processes (Arenkotter et al., 2004). Table 5. Overview of described Gordonia species and their relevant characteristics (Arenkotter et al. 2004).

Degraded and/or transformed

Strain/species Class compound

Gordonia sp. MTCC Butyl Benzyl Phthalate and other

Phthalic acid esters

4818 phthalic acid esters

G. rubripertincta DSM Atrazin, diethyl simazin, melanin and s-Triazine

10347 other similar compounds

G. nitida (sinonimo G.

Alchyl pyridine 3 -methyl pyridine, 3 -ethyl pyridine alkanivorans) organic sulphurated Benzothiophene, dibenzothiophene,

Several strains/species compounds diesel t-butyl ether, t-butyl Methyl ether ,t- Several strains/species Ethers amyl Methyl ether linear Alkanes, cycloalkanes,

Several strains/species Hydrocarbons aromatics

G. polyisoprenivorans e Synthetic and natural cis-l,4-polyisoprene, nytril G. westfalica rubbers

Recently, Medina-Moreno et al. (2005) have identified a microbial consortium able to degrade up to 47.6% of a weathered hydrocarbon mixture extract from a chronically contaminated soil. Any Gordonia strains have been isolated from the consortium.

SUMMARY OF THE INVENTION

The Gordonia strains of the invention, denominated BS25, BS29 and M22 produce extracellular bioemulsifeires. In addition, SACs produced by the Gordonia strains BS25, BS29 and M22 lower surface tension up to 28 mN/m. Thus, they are more effective than SACs produced by Gordonia strains previously described in literature (Pagilla et al., 2002; Nazina et al., 2003).

Moreover, Gordonia strains BS25, BS29 and M22 show an efficient biosurfactant production (129 x CMD), among the highest when compared to literature values.

SACs produced by Gordonia strains of the invention (BS25, BS29 and M22) generate stable oil/water emulsions showing an emulsification up to 73%. Thus, they are more effective than SACs produced by Gordonia strains previously described in literature (Nazina et al., 2003). The bioemulsifϊers produced by Gordonia strains of the invention (BS25, BS29 and M22) emulsify both pure hydrocarbons (aliphatic, aromatic and cyclic compounds) and hydrocarbon mixtures. Well-studied bioemulsifϊers (e.g. Emulsan RAG-I; Rosenberg and Ron, 1999) do not show this property. Gordonia bioemulsifϊers have never been studied for this property.

Gordonia BS29 strain is able to grow and produced bioemulsifϊers and biosurfactants on renewable substrates. Gordonia strains have never been studied for this property. Gordonia M22 strain is capable of degrading the major components of a weathered diesel. Furthermore, Gordonia M22 also shows very appreciable capability of degrading branched hydrcarbons. These compounds are considered extremely recalcitrant to biodegradation and often remain in the environment as residual contaminants after bioremediation (Nocentini et al, 2000). Presently, it is impossible to compare the Gordonia M22 strain with the consortium (Medina-Moreno et al., 2005) for the catabolic abilities because the initial hydrocarbon mixtures and the residual hydrocarbons at the end of the experiment have not been chemically characterised by Medina-Moreno et al. (2005). Thus, it is unknown if the consortium can degrade recalcitrant hydrocarbons, such as branched hydrocarbons. In summary, Gordonia strains of the invention (BS25, BS29 and M22) have the following properties:

1. Production of biosurfactants that lower surface and interfacial tensions.

2. Production of extracellular bioemulsifϊers that stabilize oil/water bioemulsions.

3. Efficient biosurfactant production. 4. Production of bioemulsifϊers and biosurfactants on cheap renewable substrates.

5. Degrade linear and branched alkanes, and major components of diesel and weathered diesel.

It is therefore an object of the present invention a micro-organism belonging to the genus

Gordonia able to produce and release extracellularly surface active compounds (SAC) and having a DNA sequence transcribed to 16 S RNA comprised in the group of: Seq ID No. 1,

Seq ID No. 2 or Seq ID No. 3. Preferably the micro-organism is of the strain named M22,

BS25 or BS29.

It is another object of the invention the use of the micro-organism of the invention for the production of surface active compounds, for the degradation of hydrocarbons or derivatives thereof, for environment decontamination process, for biocatalytic and/or bioconversion process.

It is a further object of the invention a surface active compound or a mixture thereof obtainable from the micro-organism of the invention. It is another object of the invention the use of the surface active compound of the invention or a mixture thereof for the preparation of a medicament. Preferably, the medicament has a disinfectant, anti-adhesive, anti-thrombosis, anti-microbial and/or anti- viral property. It is a further object of the invention the use of the surface active compound of the invention or a mixture thereof for the preparation of a cosmetic product, a food product and/or an emulsion.

It is an object of the invention a pharmaceutical composition comprising an effective amount of the surface active compound of the invention or a mixture thereof. It is a further object of the invention an emulsion comprising the surface active compound of the invention or a mixture thereof.

The surface active compound can be of low or high molecular weight or a mixture thereof. Moreover, the strains of the invention can be used to develop by means of genetic engineering plasmid or phage vectors able to code for enzymes responsible for SAC biosynthesis and/or responsible for hydrocarbons and derivatives thereof catabolism or transformation.

The surface active compound of the invention can be used in the oil industry as co- adjuvant for MEOR process, for oil transportation and pipelining, oil strorage tank and oily equipment cleaning and environment remediation.

The invention will be now described by way of non limiting examples referring to the following figures:

FIGURE 1: Multi-alignment of 16S rRNA gene sequences of M22 (SEQ ID. No.l, GenBank accession number EF064794), BS25 (SEQ ID. No.2, GenBank accession number EF064795) e BS29 (SEQ ID. No.3, GenBank accession number EF064796) strains and type strains of Gordonia amicalis (SEQ ID. No.4, GenBank accession number AF 101418) and Gordonia terrae (SEQ ID. No.5, GenBank accession number X79286). FIGURE 2: Un-rooted phylogenetic tree based on 16S rRNA gene comparison. The positions of M22, BS25, BS29 strains and the type strains of Gordonia species are shown. Bootstrap probability values that were less than 50% were omitted from the figure. The scale bar indicates substitutions per nucleotide position. The GenBank accession numbers are reported in parenthesis. FIGURE 3: Biodegradation kinetics on pure hydrocarbons. The hydrocarbon was supplied at initial concentration of 1.0 g/1. (A) Gordonia M22 (B), Gordonia BS29, (C) Gordonia BS25. n-heptadecane (■), pristane (A) and squalene (•). Optical density at 600 nm (ODβoo) (• • •)• Residual hydrocarbon (%) ( - ). Average values are presented. FIGURE 4: Biodegradation kinetics on hydrocarbon mixtures by Gordonia M22. The hydrocarbons were supplied at initial concentration of 0.9 g/1. Commercial diesel (•) and weathered diesel (■). Residual hydrocarbon (mg/1) in abiotic controls (...) and in inoculated cultures ( - ). Total DNA content (*). Average values are presented. FIGURE 5: Kinetics of SAC production by Gordonia BS29 strain on n-hexadecane. The hydrocarbon was supplied at initial concentration of 20.0 g/1. Total DNA content (0), emulsification activity (E24%) in the whole culture broth (Δ) or in the cell-free culture filtrate (A). Average values are presented.

FIGURE 6: Kinetics of SAC production by Gordonia BS29 strain on n-hexadecane. The hydrocarbon was supplied at initial concentration of 20.0 g/1. Total DNA content (0), surface tension (mN/m) (D), critical micelle dilution (CMD) (+). Average values are presented.

FIGURA 7: Determination of CMD parameter culture broth of Gordonia BS29 grown on n-hexadecane seven days at 30 ⁰C.

MATERIALS AND METHODS Weathered diesel

The non-aqueous liquid phase (NAPL) was collected by a bioslurping system from a site chronically contaminated by diesel. NAPL was separated from the groundwater in an oil/water separator. An aliquot of the NALP was sterilized by filtration. The chemical composition of the NAPL mixture was performed by GC/MS analysis and compared with commercial diesel composition. The analysis showed that, in NAPL mixture, n-alkanes, BTEX and some branched alkanes had become lower than the analytical detection limit and the major components are branched alkanes, such as pristane and phytane, and alkyl-substituted naphthalenes. The NAPL mixture is markedly different from commercial diesel, previously identified as the hydrocarbon mixture originally spilled in the soil. From now the NAPL hydrocarbon mixture is thus called "weathered diesel".

GC/MS analyses The carrier gas was helium. The temperature of the injector was 250⁰C and a column CP- Sil8-CB from Varian (0.32 mm, 0.25 mm thickness, 30 m) was used for the GC. The column temperature was first set at 60⁰C for 3 min; then it was increased at a rate of 10°C/min to 325°C and was set at 325°C for 10 min. Electron impact mass spectrometry was performed with a Thermo Electron Corporation, TRACE DSQ Single Quadrupole GC/MS. The MS was operated at 70 eV, the temperature of the ion source was 200⁰C. The tentatively identification of the components of the hydrocarbon mixture was performed using the NIST library database.

Enrichment and isolation of Gordonia strains

Hydrocarbon contaminated soil and groundwater samples were obtained from an industrial site contaminated by diesel. Bacterial strains of the present invention were obtained using an enrichment procedure. The enrichment cultures were prepared in 100 ml Erlenmeyer flasks containing 20 ml Bushnell-Haas medium (BH; Bushnell and Hass, 1941). Bacteria were first extract from soil by suspending 2.0 g of soil sample in 18 ml of NaCl solution (0.85 g/1) and vortexing for 2 min. Then, soil suspension or contaminated groundwater samples were used as inocula of the enrichment cultures. An aliquot of sterile NAPL was supplied at an initial concentration of 20 g/1 (w/v). The cultures were grown at 32 ⁰C in a rotary shaker at 250 rpm for seven days. Samples of the enrichment cultures were plated onto BH medium solidified with agar, and supplemented with 20 g/1 of sterile NAPL. The medium was vigorously shaken for 2 min to emulsify the oil prior to poring the plates. Colonies were obtained in pure culture by repeated streaks on BH agar plates. The carbon source was offered as a vapour phase by placing a filter disc with 90 mg of sterile NAPL in the lid of each plate and wrapping the plates in Parafilm. The three bacterial strains were called respectively BS25, BS29 and M22. They have been deposited according the Budapest Treaty before the BCCM/LMG, Gent, Belgium.

16S rRNA gene sequence analysis DNA purification and PCR reactions was carried out as previously described (Tamburini et al, 2003). The determination of 16S rRNA gene nucleotide sequences was performed with a Perkin-Elmer ABI 310 sequence analyser. The 16S rRNA gene sequences were compared with the prokaryotic small subunit rRNA sequence database of the Ribosomal Database Project II (Cole et al, 2005). The 16S rRNA gene sequences of the isolates and the related sequences (retrieved from Ribosomal Database Project II database) were aligned with the MULTALIN software (Corpet, 1988). The resulting alignments were checked manually and corrected if necessary. Phylogenetic trees were inferred using the neighbour-joining method and the software MEGA version 3.1 (Saitou and Nei, 1987; Kumar et al., 2004). Bootstrap analysis (1,000 replicates) was used to test the topology of the neighbour-joining method data.

Catabolic abilities of Gordonia M22, BS25 and BS29 Pure hydrocarbons

Growth on pure hydrocarbons

Liquid cultures were prepared in LD (per litre: yeast extract 5 g, tryptone 1O g, NaCl 5 g). The cells were removed by centrifugation, washed twice and suspended in Ml medium (per litre: K₂HPO₄ 1.32 g, KH₂PO₄ 1 g, NH₄Cl 0.81 g, NaNO₃ 0.84 g, FeSO₄-7H₂O 0.01 g, MgSO₄ 0.20 g, CaCl₂ 0.02 g). The cultures were prepared in 100 ml Erlenmeyer flasks containing 20 ml Ml medium and inoculated to an initial optical density at 600 nm (ODβoo) of 0.050. Each different carbon and energy source was supplied at an initial concentration of 20.0 g/1 (w/v). Cultures were grown at 30 ⁰C in a rotary shaker at 250 rpm. Determination of absorbance of hydrocarbon grown cells was difficult due to the formation of floes with low buoyant density, as previously described by White et al. (1997). Furthermore, the cells could not be recovered as a pellet after centrifugation (1 hours at 10,000 x g). Thus, the measures of bacterial growth on different carbon and energy sources were performed as follow: cell collection by filtration through 0.2 μm filters after seven day growth, detachment of cells from filters by vigorously shaking for 5 min, suspension on saline solution and determination of ODβoo. All determinations were performed at least in duplicate.

Pure hydrocarbon degradation

The cultures were prepared in ten replicates with 20 ml Ml medium in 100 ml Erlenmeyer flasks. The cultures were inoculated to an initial ODβoo of 1.00. Each different carbon and energy source was supplied at an initial concentration of 1.0 g/1 (w/v). At each fixed kinetic time, two flasks were sacrificed and used for residual hydrocarbon determination. The extraction for residual hydrocarbon determination was carried out adding to the flask 20 ml o-terphenyl (Internal Standard = 500 m/1) solution in methylene chloride and 1 μl of organic fraction was analyzed without any further treatment. The analyses were performed with a HP 5890 gas chromatograph coupled to a Flame Ionization Detector with a HP 5MS column (0.25 mm i.d., 30 m length, 0.25 μm film thickness). The temperature program was 2 min at 40 ⁰C, then increasing at 40 ⁰C min ¹ up to 320 ⁰C and 15 min at 320 ⁰C. The injector and detector temperatures were respectively 280 ⁰C and 320 ⁰C. Reproducibility of the entire analytical procedure for hydrocarbon quantification was about 10%.

Weathered diesel

Growth on weathered diesel

The bacterial strains were isolated on agar plates of BH medium. The carbon source was offered as a vapour phase by placing a filter disc with 90 mg of sterile NAPL in the lid of each plate and wrapping the plates in Parafϊlm. Bacterial growth was visually checked after fifteen day at 30 ⁰C.

The ability of the bacterial strains to break the NAPL layer was tested growing the microorganisms with NAPL at concentration equal to 20.0 g/1 (w/v) and inoculating the cultures to an initial optical density at 600 nm (OD₆Oo) equal to 0.05.

Weathered diesel degradation by Gordonia M22

Liquid cultures were prepared in TSB and grown at 30⁰C in a rotary shaker at 250 rpm for seven days. The cells were removed by centrifugation, washed twice and suspended in Ml medium. The cultures were prepared in 100 ml Erlenmeyer flasks containing 20 ml Ml medium and inoculated to an initial OD₆Oo of 0.500. Sterile NAPL was supplied as carbon source (950 mg/1 w/v). Uninoculated control flaks were incubated in parallel to monitor the abiotic loss of the hydrocarbons. At each fixed kinetic time, the residual hydrocarbons were extracted from inoculated cultures and uninoculated controls (two replicates) and determined according to the EPA 8015B. Quantitative and qualitative analyses of the hydrocarbons were carried out by gas chromatography with flame ionization and MS detector respectively (Thermo Electron Corporation, TRACE).

SAC production Liquid cultures were prepared in LD (per litre: yeast extract 5 g, tryptone 1O g, NaCl 5 g). The cells were removed by centrifugation, washed twice and suspended in Ml medium. The cultures were prepared in 100 ml Erlenmeyer flasks containing 20 ml Ml medium and inoculated to an initial ODβoo of 0.050. Each different carbon and energy source was supplied at an initial concentration of 20.0 g/1 (w/v). Cultures were grown at 30 ⁰C in a rotary shaker at 250 rpm. After seven days growth, emulsifϊcation activity, surface tension and critical micelle dilution (CMD) were determined. All determinations were performed at least in duplicate.

Kinetics of SAC production

The cultures were prepared in 500 ml Erlenmeyer flasks containing 100 ml Ml medium and inoculated to an initial ODβoo of 0.050. n-hexadecane was supplied at initial concentration of 20.0 g/1 (w/v). At each kinetic time, the following parameters were determined: total DNA content, emulsifϊcation activity, surface tension, CMD. Total DNA quantification was performed according to Burton (1956), using calf thymus DNA as standard. The cells were previously collected on glass fibre filters, as described by Bipatnath et al. (1998). All determinations were performed at least in duplicate.

Surface tension, interfacial tension and CMD Interfacial tension against diesel and surface tension were determined at room temperature. CMD was determined by dilution method as previously described by Pagilla et al. (2002). Analyses were carried out tensio meter 3 S GBX. All determinations were performed at least in duplicate.

Emulsifϊcation assay

Samples of whole culture broths (with cells) or culture filtrates (without cells) were used. The cells were removed from the culture by filtration through 0.2 μm filters. 3 mL of sample was vortexed while the same amount of hydrocarbon was added drop by drop over 30 seconds in glass graduated tubes. After this, the tube was vortexed for additional 2 minutes. The mixture was allowed to settle for 24 h. The emulsification activity (E24%) is given as percentage of middle emulsion phase normalized to the total volume (Cooper and Goldenberg, 1987). All determinations were performed at least in duplicate. Optimization of the culture conditions

Growth condition and culture preparation

Liquid cultures were prepared in LD (per litre: yeast extract 5 g, tryptone 10 g, NaCl 5 g).

Cultures were prepared with 20 ml of mineral medium in 100 ml Erlenmeyer flasks inoculating the cultures to an initial optical density at 600 nm (ODβoo) equal to 0.1 and adding the necessary amount of carbon source. A part from the investigated components, the basal composition of the mineral medium was as follow: CaCO₃ (ppm): 0.40;

ZnSO₄-7H₂O (ppm): 0.28; MnSO₄-H₂O (ppm): 0.22; CuSO₄-5H₂O (ppm): 0.05;

CoSO₄-7H₂O (ppm): 0.06; H₃BO₃ (ppm): 0.012.

Coding of the variables

In each experimental design, the independent variables were coded according to the following equation (Equation 1):

where C₁ is the coded value of i^th variable, X₁ is the i^th actual value of i^th variable, x_max and x_min are the higher and the lower limit values of the range chosen for the i^th variable.

Fractional Factorial Design (FFD)

To determine the culture factors that have significant influence on biosurfactant(s) production, on the basis of previous literature, six major inorganic components, the carbon source and the time of growth were selected to be evaluated (Table 6).

Table 6. Values of higher and lower levels for each variable in 2^^8"2-* FFD.

According to a full factorial design at two levels, 28 experiments should have been carried out. The number of experiments was reduced by using a one-quarter fractional design assuming the three-order and higher interactions between factors negligible. Therefore, the selected design was so a two level fractional factorial design with a resolution of V (2(8-2) R=V) composed by 64 different experiments to which other 4 centre point experiments were added. Each factor was present in the experimental design at two different levels and the values of the levels have been chosen on the basis of previous studies (Table 6).

Steepest Ascendant Method (SAM)

A two level Full Factorial Design (2(3)) with three centre points was then applied centred on the best conditions found in the previous design. This design was built with the significant factors NaCl and phosphates, testing also two different concentrations of n- hexadecane as carbon source. Table 7 shows the tested range for this design.

Table 7. Values of higher and lower levels for each variable in 2^ Full Factorial Design.

All the other factors were kept constant at coded value of 0 according to Eq. 1 and ranges shown in Table 6. The aim of this design was to explore the response in the region of the current optimum and to find the direction in the experimental field in which the response increases most rapidly.

Then, starting from the centre of the full factorial design, eight experiments were conducted along the linear steepest ascent calculated by the first order function that best fits the Full Factorial Design data using NaCl and n-hexadecane as independent variables.

Table 8 shows the conducted experiments.

Table 8. Experiments of the Steepest Ascendant Method.

Table 9. Optimised cultural conditions for the biosurfactant production by Gordonia BS29. The basal composition of the mineral medium was as follow: CaCO₃ (ppm): 0.40; ZnSO₄-7H₂O (ppm): 0.28; MnSO₄-H₂O (ppm): 0.22; CuSO₄-5H₂O (ppm): 0.05; CoSO₄-7H₂O m : 0.06; H₃BO₃ (ppm): 0.012.

Data Analysis

ANOVA analyses of experimental design results and regression analyses have been made using STATISTICA software.

RESULTS The three bacterial strains (called M22, BS25, BS29) are aerobic, Gram-positive, catalase- positive, nonmotile. On Tryptic Soy Agar medium, M22 strain grows forming red- pigmented, rough colonies. BS25 and BS29 strains form red-pigmented, smooth colonies with shiny surface; however, also rough colonies with irregular margins were occasionally observed.

16S rRNA gene analysis

The almost complete sequence of 16S rRNA gene of each strain was determined and compared. As determined by multi-alignment, the three sequences differ in few nucleotides (Figure 1).

The 16S rRNA gene sequences of the three strains (BS25, BS29, M22) were compared with the Ribosomal Database Project II (Cole et al., 2005). This analysis assigned the three strains (BS25, BS29, M22) to the Gordonia genus belonging to the Corynebacterineae suborder, the mycolic acid-containing group within the Actinomycetales order

(Stackebrandt et al., 1997).

The 16S rRNA gene sequences of the type strains of all Gordonia species, validly described [bacterial nomenclature up-to-date (approved lists, validation lists) compiled by

DSMZ - Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH

Braunschweig, Germany May 2006], were retrieved from Ribosomal Database Project II and GenBank databases.

The 16S rRNA gene sequences of the three strains (BS25, BS29, M22) and that of the type strains of all Gordonia species were used to construct a phylogenetic tree (Figure 2). This analysis demonstrated the BS25 and BS29 strains are phylogenetically closely related to

Gordonia terrae? (99.8% similarity), whereas the M22 strain is closely related (100.0% similarity) to Gordonia amicalis .

Catabolic abilities of Gordonia M22, BS25 and BS29

The characterisation of the catabolic abilities of the three Gordonia strains (M22, BS25, BS29) is mainly related to the use of the strains themselves and/or their metabolic products as enhancer in remediation of hydrocarbons. A) Pure hydrocarbons Growth on pure hydrocarbons

The ability of the three Gordonia strains (M22, BS25, BS29) to grow using different hydrocarbons was tested in batch liquid cultures in Ml medium. The different hydrocarbons (20 g/1) were supplied as the only carbon and energy source. The three strains (M22, BS25, BS29) are able to grow using a wide range of straight hydrocarbons (Xi-C_10-28), branched aliphatic hydrocarbons (pristane) and diesel and showed a fast biomass increase on these substrates.

However, the three strains (M22, BS25, BS29) did not show a significant biomass increase on n-hexane, cyclohexane, naphthalene, 1 -methyl- and 2-methyl-naphthalenes, and phenanthrene after seven days of growth (Table 10).

Table 10. Growth of the three Gordonia strains (M22, BS25 e BS29) in batch liquid cultures on pure hydrocarbons (20 g/1) as the only carbon and energy source.

+: ODβoo greater than 1.0 after seven days at 30 ⁰C. Pure hydrocarbon degradation

The hydrocarbon degradation potential of the three Gordonia strains (M22, BS25, BS29) was evaluated, n-heptadecane, pristane and squalene were chosen as middle-range hydrocarbons to test the catabolic abilities of selected bacteria. These compounds cover a wide range of molecular weight and different degrees of branching. The degradation was tested in batch liquid cultures in Ml medium. The different hydrocarbons (1.0 g/1) were supplied as the only carbon and energy source. At each fixed kinetic time, the residual hydrocarbon concentration was determined by gas chromatography with flame ionization (Figure 3). The three Gordonia strains (M22, BS25, BS29) were able to completely degrade n- heptadecane and pristane within one week. They also degraded squalene, but around 30% of the initial hydrocarbon, still remained after seven days. During the first two days, bacterial growth and hydrocarbon degradation were fast. The biodegradation data can be well modelled by a first order kinetics. Thus, the three Gordonia strains (M22, BS25, BS29) efficiently degrade a wide spectrum of pure hydrocarbons.

B) Weathered diesel Diesel is a complex mixture composed of four main structural classes of hydrocarbons: linear, branched and cyclo alkanes, and aromatics (Marchal et al, 2003). The biodegradation of commercial diesels has been studied thoroughly by several authors.

Several microbial strains (Gram-positive and Gram-negative bacteria, fungi) have been characterised for the ability to degrade the diesel hydrocarbons (Jonathan et al., 2003).

However, many studies reported the incomplete degradation of diesel oil in contaminated sites.

In chronically contaminated environments, complete biodegradation of hydrocarbon contaminant mixtures does not often occur even in absence of other environmental limiting factor (such as nutrients or oxygen), due to the presence of recalcitrant compounds.

Branched hydrocarbons are considered extremely recalcitrant to biodegradation and often remain in the environment as residual contaminants after bioremediation (Nocentini et al,

2000).

The susceptibility to biodegradation of the diesel components depends on their chemical structure. The branched structures have been found to be the least biodegradable class among diesel components (Whyte et al., 1998; Marchal et al., 2003).

The degradation of weathered hydrocarbon mixtures has been poorly studied in literature.

However, chronically contaminations are widespread and represent a serous environmental and health risk. The non-aqueous liquid phase (NAPL) was collected from a site chronically contaminated by diesel. The chemical composition of the NAPL mixture was performed by GC/MS analysis and compared with that of the commercial diesel. The major components of

NAPL mixture are branched alkanes and alkyl-substituted naphthalenes. The NAPL mixture, called "weathered diesel", is markedly different from the commercial diesel and is enriched with respect to less biodegradable hydrocarbons, such as branched compounds.

Growth on weathered diesel

The three Gordonia strains (M22, BS25, BS29) were fast growing on solid media with weathered diesel as the carbon and energy source, and formed well visible colonies after seven days of growth. The ability of the three Gordonia strains (M22, BS25, BS29) to grow using weathered diesel was tested in batch liquid cultures in Ml medium. The hydrocarbon mixture (20 g/1) was supplied as the only carbon and energy source. Among the three Gordonia strains, only Gordonia M22 broke the NAPL layer on the aqueous surface rendering the culture turbid after seven days of incubation. This preliminary result suggests that the Gordonia M22 can degrade, at least partially, the weathered diesel. Degradation of commercial and weathered diesels by Gordonia M22

The catabolic capabilities of the Gordonia M22 strain were evaluated in batch liquid cultures in Ml medium supplemented with weathered diesel or heating diesel as carbon and energy source. At each fixed kinetic time, the residual hydrocarbon concentration was determined by gas chromatography with flame ionization. Since absorbance of cells grown on hydrocarbons was difficult to determine, due to their low buoyant density, and since the cells could not be recovered as a pellet after centrifugation, bacterial growth was monitored by total DNA quantification (Figure 4).

After three days growth, the overall extents of degradation of heating and weathered diesels were 49% and 23%, respectively. Further degradation was not detected until the end of the experiment (14 days).

The Gordonia M22 grew exponentially during the first 48 hours after inoculation with commercial diesel as substrate. This phase was followed by the stationary phase. On the contrary, no bacterial growth was observed with weathered diesel as substrate. After 7 days growth, the residual hydrocarbons in the abiotic controls and in Gordonia M22 cultures were analyzed by GC/MS. In order to determine the extent of degradation of different classes of weathered diesel hydrocarbons, the areas of individual peaks (from duplicate samples) were quantified by integration and expressed as the percentage relative to the amount of the corresponding peak in the abiotic controls. The analysis was performed using chromatograms generated from selected characteristic ions of branched alkanes (selected ions, 71) or alkyl-substituted monoaromatics, alkyl- substituted naphtahlenes, and naphthenes (selected ions, 119+134+141+155+169). In weathered diesel experiments, linear and branched alkanes were extensively degraded by M22 (45%); while alkyl-substituted monoaromatics, alkyl-substituted naphthlenes and naphthenes were partially degraded, 34%, 23% and 26% respectively. Thus, Gordonia M22 strain is capable of degrading the major components of the weathered diesel hydrocarbon mixture. Furthermore, in diesel experiments, the degradation of linear and branched alkanes was 88%. Alkyl-substituted monoaromatics and alkyl-substituted naphthenes were degraded at the same extend than in weathered diesel degradation (33%, and 23% respectively), while the degradation of alkyl substituted naphthalenes was 35%. Thus, the Gordonia M22 strain degrades the major components of commercial and weathered diesels. SAC production and properties

The properties of the SACs produced by the three Gordonia strains (M22, BS25, BS29) were evaluated by determining the following parameters: (i) emulsification activity (E24%). The parameter measures the ability of bioemulsifϊers to stabilise oil/water emulsions (Cooper and Goldenberg, 1987); (ii) surface and interfacial tensions. The parameters measure the biosurfactant effectivity; (iii) critical micelle dilution (CMD, determined as the reciprocal of the dilution factor to achieve the critical micelle concentration CMC) (Pagilla et al., 2002). The parameter is proportional to the amount of biosurfactant produced and measures the efficiency of microbial production. A) Bioemulsifier properties The ability of the three Gordonia stains (M22, BS25, BS29) to produce SACs which stabilise oil/water emulsions was evaluated.

The spectrum of carbon and energy sources that the three Gordonia strains (M22, BS25, BS29) utilise as substrate to produce bioemulsifϊers was evaluated in batch liquid cultures with different hydrocarbons (n-Cio-π, branched alkanes and diesel) or water-soluble substrates (alcohols, organic acids, sugars) as carbon and energy source.

The ability of whole culture broths (with cells) and culture filtrates (without cells) to produce a stable kerosene-water emulsion was tested (Table 11). High emulsification activities were detected in whole culture broths and in cell- free filtrates when the three Gordonia strains (M22, BS25, BS29) were grown on linear and branched hydrocarbons (up to 70.0%) and on water-soluble substrates, such as alcohols, organic acids, sugars (up to 73.3%). The bioemulsions remained stable for more than one year.

Table 11. Emulsification activity of whole culture broths and cell- free culture filtrates of Gordonia M22, BS25 and BS29 grown on different carbon and energy sources for 7 days at 30 ⁰C. Values are means based on two separate experiments with two independent measurements each.

E24% (%) Carbon sources

Gordonia M22 Gordonia BS25 Gordonia BS29 (20 g/1)

n-C10 9.8 35.5 48.3 70.0 43.1 64.9 n-C12 3.2 45.8 53.8 65.6 50.4 59.8 n-C16 56.3 52.5 49.5 41.5 39 .6 44 .0 n-C17 37.3 57.5 39.6 38.2 37 .2 28 .3

Pristane 44.1 68.3 57.9 61.8 57 .5 63 .5

Diesel 2.2 42.4 46.7 44.0 40 .4 63 .4

Fructose 28.3 26.5 43.3 38.4 46 .3 50 .4

Glucose 0.0 0.0 ND^a ND^a 14 .6 1. 7

Sucrose 25.9 20.9 25.0 47.6 30 .2 10 .6

Ethanol 16.7 22.5 23.3 48.4 12 .0 39 .0

Sodium acetate 0.0 0.0 ND^a ND^a 40 .9 45 .0

Potassium citrate ND^a ND^a 53.8 31.1 73 .3 35 .0

Sodium citrate ND^a ND^a 56.7 60.5 62 .6 51 .1

Palmitic acid 8.3 43.2 53.4 50.8 45 .0 38 .7 a Biomass increase was not detected.

E24%: Emulsification activity; C: Culture broth; F: Cell- free culture filtrate; ND: not determined.

Thus, the three Gordonia strains (M22, BS25, BS29) release extracellular bioemulsifiers in the culture medium when grown both on water-soluble and insoluble substrates. A cost-effective process is the foundation for every profit-making biotechnology industry. Microbial SACs are unlikely to be produced at low cost if extensive downstream processes are required (Makkar e Cameotra, 2002). The ability of the three Gordonia strains (M22, BS25, BS29) to produce extracellular bioemulsifiers is important for the development of an economically competitive process since the release of the product in the culture medium simplifies the following purification step.

The ability of the bioemulsifiers produced by the Gordonia BS29 strain to emulsify pure hydrocarbons (aliphatic, aromatic and cyclic compounds) and hydrocarbon mixtures was investigated (Table 12).

Table 12. Emulsification of different hydrocarbons by cell-free culture filtrates of Gordonia BS29. Values are means based on two separate experiments with two independent measurements each.

Oil phase Structural class E24% (%)

Kerosene Mixture 61.7

Cyclohexane Cycloalkane 62.1 Toluene Aromatic 59.3 n-C6 Linear alkane 46.7

Pristane branched alkane 42.5 n-C16 Linear alkane 21.7 n-C17 Linear alkane 26.2

Diesel Mixture 18.3

The Gordonia BS29 was grown in batch liquid culture in Ml medium supplemented with sodium citrate as carbon and energy source. After three days at 30 ⁰C, the bacterial cells were removed by filtration and cell- free filtrate was used to emulsify different hydrocarbons. The cell-free filtrate produced stable emulsion of pure hydrocarbons (aliphatic, aromatic and cyclic compounds) and hydrocarbon mixtures. Therefore such filtrate have a broad spectrum of applications. Well-studied bioemulsifϊers (e.g. Emulsan RAG-I; Rosenberg and Ron, 1999) do not present this property. Furthermore, Gordonia bioemulsifϊers have never been studied for this property.

B) Biosurfactant properties

Microorganisms that decrease surface tension of culture media by more than 10 mN/m are considered promising producers of surfactants (Francy et al., 1991).

The ability of the three Gordonia stains (M22, BS25, BS29) to produce SACs which reduce surface and interfacial tensions was evaluated. The parameters were measured in whole culture broths (with cells) and in culture filtrates (without cells) (Table 13).

The spectrum of carbon and energy sources that the three Gordonia strains (M22, BS25,

BS29) utilise as substrate to produce biosurfactants was determined in batch liquid cultures supplemented with different hydrocarbons (n-Ciβ and diesel) or water-soluble substrates

(alcohol, organic acids, sugars) as carbon and energy source.

Table 13. Optical density at 600 nm, surface tension and critical micelle dilution of whole culture broths of Gordonia M22, BS25 and BS29 strains grown on different carbon and energy sources for 7 days at 30 ⁰C.

Carbon Gordonia M22 Gordonia BS25 Gordonia BS29 sources ST ST ST

OD₆₀₀ CMD OD₆₀₀ CMD OD₆₀₀ CMD (20 g/1) mN/m mN/m mN/m n-C16 2.9 43.8 6 2.6 30.1 13 1.7 30.0 26 Diesel 1.5 35.8 7 2.0 32.0 7 1.4 33.5 14 Sucrose 1.1 62.3 <1 1.9 48.6 <1 1.3 53.5 <1

Fructose 0.84 49.8 <1 0. 57 51 .3 <1 1 .7 47 .1 <1

Ethanol 0.34 57.1 <1 0. 26 56 .0 <1 0. 26 61 .3 <1

Potassium

0.029 ND^a ND^a 3 .7 51 .4 <1 4 .0 43 .7 <1 Citrate a Biomass increase was not detected.

OD₆OO : Optical Density at 600 nm; ST: Surface Tension; CMD: Critical Micelle Dilution;

ND: not determined.

High biosurfactant production was observed in culture broths of the three Gordonia strains (M22, BS25, BS29) grown on hydrocarbons, which resulted in low surface tension of the cultures; high CMD values were also found. When the Gordonia BS29 was grown on n- hexadecane, the surface tension achieved in culture broth was 30.0 mN/m and the CMD values was 26, whereas the interfacial tension against diesel was 1.1 mN/m. Cell- free culture filtrates did not show significant reduction in surface tension.

The reduction of surface tension and the CMD values were considerably lower in culture broths on water-soluble compounds than the values obtained on hydrocarbons, also when comparable or higher biomass increases were observed. Thus, the three Gordonia strains (M22, BS25 e BS29) release bio surfactants when grown on insoluble substrates. The majority of the product is cell-bound.

C) Kinetics of SAC production

The kinetics of SAC production exhibit many variations among various biological systems, and only a few generalizations can be derived. The characterisation of the kinetic type is important during the optimisation of fermentative condition and the development of a cost- competitive production (Desai e Banat, 1997).

The relationships among SAC production, phase of growth and cell surface properties was addressed by a kinetic experiment carried out growing Gordonia BS29 strain on n- hexadecane as a model compound for microbial degradation of middle molecular weight hydrocarbons. Since absorbance of cells grown on hydrocarbons was difficult to determine, due to their low buoyant density, and since the cells could not be recovered as a pellet after centrifugation, bacterial growth was monitored by total DNA quantification. SAC production was quantified by measuring the following parameters: i) emulsification activity in whole culture broths and cell-free filtrates, ii) surface tension and biosurfactant concentration by dilution test (CMD) of whole culture broths.

The Gordonia BS29 grew exponentially during the first 43 hours after inoculation in minimal medium with n-hexadecane, with a generation time of about 8 hours. This phase was followed by a linear growth.

Growth phase-dependent production and release of bioemulsifiers were observed and the highest emulsifying activity was detected at the end of the exponential phase both in whole culture broths and in cell- free filtrates. The emulsification activity was constant during the linear growth in whole culture broths, whereas it slowly decreased in cell- free filtrates (Figure 5).

Growth phase-dependent production of bio surfactants was observed (Figure 6). Twenty- four hours after the inoculation, the surface tension achieved in culture broths was 31.1 mN/m and remained nearly constant during the following six days, reaching the minimum value four days after inoculation (29.7 mN/m). The biosurfactant concentration, as determined by dilution test, was lower than the critical micelle concentration during the first two days of growth. The surfactant accumulated during the linear phase and the maximum concentration was achieved after seven days (CMD = 8 x CMC) (Figure 7). The characterisation of the kinetics of SAC production will be important during the optimisation of fermentative condition and the development of a cost-competitive production of SACs by the three Gordonia strains (M22, BS25, BS29).

Development of a cost-competitive process for SAC production

The surfactant industry now exceeds US dollar 9 billion per year (Desai and Banat 1997). Most of these compounds are synthesized chemically of petroleum. Most are also toxic to the environment, not easily biodegradable, and their manufacturing processes and byproducts can be environmentally hazardous. Increased environmental awareness and strict legislation have made environmental compatibility of surfactants an important factor in their application for various uses (Makkar e Cameotra, 2002). SACs possess many commercially attractive properties and clear advantages over their synthetic counterparts. Economy is often the bottleneck of biotechno logical processes, especially in the case of SAC production. The development of a cost-competitive process for SAC production is tricky because of the low yields and the high recovery and purification costs (Mukherjee et al, 2006).

The production economy of every microbial metabolite is governed by three basic factors: i) initial raw material costs, ii) availability of suitable and economic production and recovery procedures and iii) product yield of the microbial producer.

Thus, in light of the economic constraints associated with SACs production, two strategies were adopted to make the production of SACs by Gordonia BS29 cost-competitive: -Use of cheaper and waste substrates to lower the initial raw material costs -Optimization of the culture conditions.

A) Use of cheaper and waste substrates

Often, the amount and type of a raw material can contribute considerably to the production cost. It is estimated that raw materials account for 10 to 30% of the total production costs in most biotechno logical processes (Makkar e Cameotra, 2002; Mukherjee et al., 2006). Thus, to reduce this cost it is desirable to use low-cost raw materials.

The authors tested a variety of cheap renewable raw materials, including plant- polysaccharides, agro -industrial wastes (cheese whey and sugar-beet molasses), edible and non-edible plant-derived oils, and waste frying oil, as substrates to support SAC production by Gordonia BS29 (Table 14). Table 14. Growth of the Gordonia BS29 strain in batch liquid cultures on cheap renewable raw materials and wastes as substrates.

Substrate Growth

Plant-polysaccharides

Carboxymethylcellulose 5 g/1 (w/v) -

Pectin 5 g/1 (w/v) -

Polygalacturonic acid 5 g/1 (w/v) -

Soluble starch 5 g/1 (w/v) -

Agro-industrial wastes

Cheese whey 20 g/1 (w/v) +

Sugar-beet molasses 20 g/1 (w/v) +

Edible plant-oils

Corn oil 20 g/1 (w/v) +

Palm frying oil 20 g/1 (w/v) + Peanut oil 20 g/1 (w/v) +

Rapeseed oil 20 g/1 (w/v) +

Soybean oil 20 g/1 (w/v) +

Sunflower oil 20 g/1 (w/v) +

Waste frying oil 20 g/l(w/v) +

Non-edible plant-oils +

Castor oil 20 g/1 (w/v) +

Linseed oil 20 g/1 (w/v) +

+: OD₆Oo greater than 1.0 after seven days at 30 ⁰C; -: OD₆oo< or = 1.0 after seven days at 30 ⁰C

The Gordonia BS29 strain is able to grow efficiently on molasses, cheese whey, a wide spectrum of plant-oils, and on waste frying oil from food industry. However, the strain BS29 did not show a significant biomass increase when grown on plant-polysaccharides after seven days of growth (Table 14). The emulsifϊcation activity of Gordonia BS29 is reported Table 15.

Table 15. Emulsifϊcation activity of whole culture broths and cell- free culture filtrates of Gordonia BS29 grown on different renewable substrates for 7 days at 30 ⁰C. Values are means based on two separate experiments with two independent measurements each.

E24% (%) Substrate 20 g/1 (w/v)

Non renewable substrates (control)

Potassium citrate 73.3 35 n -hexadecane 39.6 44 Agro-industrial wastes

Sugar-beet molasses 8.3 31.7 Edible plant-oils

Corn oil 30.8 31.8

Palm frying oil 16.9 35.6

Peanut oil 41.6 28.3

Rapeseed oil 21.7 28.3

Soybean oil 36.7 35.8

Sunflower oil 47.1 30.1 Non-edible plant-oils

Castor oil 0 10 .7

Linseed oil 25 31 .4

Waste oils

Waste fryin: g oil 26.7 33 .3

C: Culture broth; F: Cell- free culture filtrate.

The Gordonia BS29 strain produces bioemulsifiers when grown on soluble and insoluble renewable substrates. The same emulsifϊcation activity was found on molasses, a wide spectrum of plant-oils, and on waste frying oil from food industry (Table 15).

The surface tension and CMD of biosurfactants produced by Gordonia BS29 is reported in Table 16.

Table 16. Surface tension and critical micelle dilution of whole culture broths of Gordonia BS29 strain grown on different carbon and energy sources for 7 days at 30 ⁰C. Substrate 20 g/1 (w/v) Surface tension (niN/m) CMD

Non renewable substrates (control)

Potassium citrate 43.70 <1 n - hexadecane 30.00 26

Agro-industrial wastes

Sugar-beet molasses 53.75 <1

Edible plant-oils

Corn oil 28.90 41

Palm frying oil 28.07 41

Peanut oil 28.72 41

Rapeseed oil 30.00 41

Soybean oil 32.77 21

Sunflower oil 32.39 41

Non-edible plant-oils

Linseed oil 33.79 14

Castor oil 39.48 <11

Waste oils

Waste frying oil 29.02 41 The Gordonia BS29 strain produces biosurfactants when grown on insoluble renewable substrates. The same emulsifϊcation activity was found on a wide spectrum of edible plant- oils, and on waste frying oil from food industry (Table 16). Furthermore, the efficiency of biosurfactant production, determined as CMD, was higher on low-cost plant oil and on waste frying oil (41 x CMC) than on n-C\6 (26 x CMC), a non-renewable and expensive carbon source derived from petroleum.

B) Optimization of the culture conditions

The optimisation process involved three consecutive steps (Montgomery, 2005). In the first step a two level 2^^8"2-* Fractional Factorial Design (FFD) was used to identify culture factors that have a significant influence on biosurfactant(s) biosynthesis. Then, on the selected factors, a steepest ascent procedure was applied to obtain a second order polynomial function that fitted the experimental data in the vicinity of the optimum. The factors taken in account were inorganic nutrients such as phosphorous, ammonium and micronutrients, the carbon sources and the time of growth.

Selection of significant factors by FFD

The FFD allowed to select, among the eight tested culture conditions, three medium components, that significantly effect the amount of biosurfactant(s) produced by Gordonia BS29 strain. Among the several culture factors that can potentially affect the production of biosurfactant(s) the authors chose six inorganic nutrients and the time of growth of the cultures. For each of these factors a wide range of concentration has been selected with a 1 :10 ratio between the lower and the higher value. Furthermore, the type of carbon source was added to the design as a qualitative variable, n-hexadecane and glycerol were chosen as the best insoluble and soluble substrates according to previous tests. The authors supposed that the three and higher interaction could be neglected. This allowed reducing the number of experiments in respect to a 2® full factorial design. The assumption led to build a 2 ^^8"2-* FFD with resolution of V, with seven quantitative and one qualitative independent variables. The whole experimental design was composed by 64 runs plus 4 centre point experiments, two for each value of the qualitative variables. For each run the CMD of the crude broth was measured as a parameter proportional to biosurfactant(s) concentration. Results were then analysed by ANOVA test to determine which of the factors significantly affect the biosynthesis. The output of the ANOVA test (α = 0.05) is shown in Table 17. In the ANOVA test three and higher interactions have not been considered and only most significant factors and interactions are presented. Table 17. ANOVA test results for 2^(8~2)FFD.

The significant factors and interactions are reported Table 18.

Table 18. Estimated effects of significant factors and interactions in 2^^8"2-* FFD.

Being F0.05, 1 , 31 = 4.16, the factors and the interactions that resulted significant after the test were: -the qualitative variable of the carbon source (X7/1), -the phosphates (X3), -sodium chloride (x₆) and - the interactions between these factors (x7/ibyx3 ; X7/ibyx6 ; x₃byx6). The values of estimated effects (Table 18) permitted to evaluate the positive or negative correlation between the significant factors and the response; particularly, the positive correlation of the variable carbon source (the most important factor) means, being a qualitative variable, that the value of the variable coded by +1 (n-hexadecane) results to be better than the other (glycerol). Ignoring the insignificant factors and interactions the first- order model that best fits the experimental data was the following (Equation 2):

CMD^pred = 16.76 + 10.07x, - 1.SIx₆ + 15.2Ix₇₇₁ - 6.55X₁X₆ + 9.54x,x₇ - TASx₆X₁

Where CMD ^pred is the model predicted value of CMD. Substituting the values of the variables that gave the maximum predicted value (x₃ = 1; x₆ = -1 ; x₇/i = 1), a value of CMD^pred equal to 73.49 (90% confidence interval: 63.15 - 83.49) was obtained. Even if this was only the factor-screening step, the authors obtained a maximum CMD^pred value that can be considered among the best found in the current literature and almost 10-fold more the one that was observed for Gordonia BS29 strain cultured in BH mineral medium.

Response Surface Analysis

The SAM is an efficient tool to rapid move towards this region of the experimental field. To detect the direction in which the response increases most rapidly, a full factorial design with three quantitative variable (x₃ ; x₆ ; x₇/₂) has been applied around the best condition found in the previous design, i.e. X₁ = x₂ = x₄ = xs = xs = 0 ; x₃ = 1 ; X₆ = -1. n-hexadecane was now chosen as quantitative variable (x₇/₂) as shows in Table 7. ANOVA test for this design showed that n-hexadecane is the only significant factor in this conditions. Even if NaCl did not result significant with this level of significance, a first order model, considering only main effect and ignoring the variable x_7j was applied to find the equation (Equation 3) of the plan that best fits the experimental data:

CMD^pred = 51.59 - 12.25x₆ + 37.3Ox₇₇₂

Eight experiments were then carried out on this plan starting from the centre of the full factorial design along the direction of the highest slope (Table 8). The other variables were kept constant at the following values: X₁ = x₂ = x₄ = xs = xs = 0 ; x₃ = 1 according to Eq.1 and ranges of Table 6. Results allowed detecting a maximum along this path that coincided with the step 3. In these conditions, the concentration of n-hexadecane and NaCl were 4.25 % and 114 ppm, respectively. The authors observed a significant increase in the CMD value from the origin to the step 3 (128.3). This further confirmed the importance of the amount of n-hexadecane on the biosurfactant production.

The optimised cultural conditions for biosurfactant production by Gordonia BS29 are presented in table 9. With the optimised composition we obtained a more than 16-fold increase in biosurfactant concentration respect to the normal BH broth, reaching a CMD value (128) among the highest in literature (Banat et al, 2000 ; Mukherjee et al, 2006 ; Makkar e Cameotra, 2002).

REFERENCES Anon (1984) Emulsan. Chemical Week 135, January, 58.

Appaiah et al, (1991) Biotechnol Lett 13:371 :374.

Arenskotter M., Broker D., Steinbuchel A. (2004) Appl Environ Microbiol. 70:3195-3204.

Banat I.M., et al., (1991) World J. Microb. Biot. 7: 80-88.

Banat I.M., Makkar R.S., Cameotra S.S. (2000) Appl. Microbiol. Biot. 53: 495-508. Barkay T., et al., (1999) Appl. Environ. Microbiol. 65:2697-2702.

Bertrand J.C., Bonin P., Goutex M., Mille G. (1994) Res Microbiol 145: 53-56.

Bipatnath M., Dennis P.P., Bremer H. (1998) J. Bacterid. 180: 265-273.

Burton K. (1956) Biochem. J. 62:315-323.

Bushnell D., Haas H. F. (1941) J. Bacterid. 41 : 653-673. Busscher H.J., et al., (1997) Applied Environmental Microbiology 63:3810-3817.

Christofi N., Ivshina LB. (2002) J. Appl. Microbiol. 93: 915-929.

Cole JR et al., (2005) Nucleic Acids Res Jan 33:D294-D296.

Cooper, D.G., Goldenberg, B.G. (1987) App. Environ Microbiol 53: 224-229.

Desai JD, Banat IM. (1997) Microbiol MoI Biol Rev. 61 :47-64. Francy D.S. et al., (1991) J. of Indus. Microbiol, and Biotech. 8:237- 245.

Gutnick D.L., Bach H. (2000) Appl Microbiol Biotechnol. 54:451-60.

Harvey S., et al., (1990) BioTechnology 8:228-230.

Hayes M.E., Nestaas E., Hrebenar K.R. (1986) Chemtech 4: 239-243.

Herman D.C., Artiola J.F., Miller R.M. (1995) Environmental Sci Technol 29:2280-2285 Ishigami Y., et al., (1988) Japanese Patent Kokai 63-182, 029.

Iwahori K., Tokutomi T., Miyata N., Fujita M. (2001) J. Biosci. Bioeng, 92:77-79.

Jonathan D., Van Hamme A.S., Ward. O.P. (2003) Microbiol. MoI. Biol. Rev. 67:503-49. Jukes T.H., Cantor CR. (1969) In Mammalian Protein Metabolism. Munro, H.N. (ed) Academic Press, New York, NY, vol. 3, pp. 21-132.

Kawaguchi K., et al, (1996) Nippon Suisan Gakkaishi 62:123-128.

Kikuchi T., Hasumi K. (2003) J Antibiot (Tokyo) 56:34-37. Kikuchi T., Hasumi. K. (2002) Biochim. Biophys Acta 1596:234-245.

Kim K, et al., (1998) Biochem Pharmacol. 55: 975-985.

Klekner V., Kosaric. N. (1993) In N. Kosaric (ed.), Surfactant Science Series, vol. 48. Marcel Dekker, New York, N.Y.

Kuiper L, et al., (2004) Molecular Microbiol. 51 :97-l 13. Kumar S., Tamura K., Nei M. (2004) Briefings in Bioinformatics 5:150-163.

Lang S., Philp J.C. (1998). Antonie van Leeuwenhoek 74:59-70.

Lim J., et al., (2005) J. Vet. Sci. 6:353-355.

Makkar R.S., Cameotra S.S. (2002) Appl Microbial Biotechnol 58:428-434.

Marchal R. et al., (2003) Oil and Gas Science and Technology, 58:441-448. Medina-Moreno S.A., et al., (2005) Can J Microbiol. Mar;51(3):231-9.

Meylheuc T., van Oss C.J., Bellon-Fontaine M.N. (2001) LO28. 91 :822-832.

Miller R.M. (1995) Environ Health Perspect 103:59-62.

Mireles J.R., Toguchi A., Harshey R.M. (2001) J. Bacteriol. 183:5848-5854.

Montgomery D. C. (2005) Design and analysis of experiments (6th Edition). New York: John Wiley and Sons, Inc.

Mukherjee S., Das P., Sen R. (2006) TRENDS in Biotechnology 24:509-15.

Mulligan CN. (2005) Environ. Pollut 133:183-198.

Nazina T.N., et al., (2003) Microbiology 73:173-178.

Neu T. (1996) Microbiological review. 60:151-166. Nocentini M., Pinelli D., Fava F. (2000) Chemosphere 41 : 1115-1123.

Pagilla K.R., Sood A., Kim H. (2002) Water Sci Technol.46:519-24.

Patel M.N., Gopinathan K.P. (1986) Appl Environ Microbiol. 52:1224-1226.

Reid G. (2000) Int: Dairy J. 10:415-419.

Rodrigues L.R., et al., (2004) Appl. Microbiol. Biotechnol. 66:306-311. Rodrigues L.R., et al., (2006) FEMS Immunol. Med. Microbiol. 46: 107-112.

Rodrigues L., et al., (2006) Journal of Antimicrobial Chemotherapy 57:609-618.

Rodrigues L.R., et al., (2006) Journal of Applied Microbiology. 100:470-480.

Ron E.Z., Rosenberg E. (1999) Appl Microbial Biotechnol 52:154-162. Ron E.Z., Rosenberg E. (2002) Current Opinion in Biotechnology 13:249-252.

Rosenberg, E., Ron. E. Z. (1998) In D. L. Kaplan (ed.), Biopolymers from renewable resources. Springer, New York, N. Y.

Rosenberg E., Ron E. Z. (1997) Current Opinion in Biotechnology 8:313- Stackebrandt E.,Rainey F.A., Ward-Rainey N. L. (1997) Int J Syst Bacterid, 47:479-491.

Saitou N., Nei M. (1987) MoI Biol Evol 4: 406-425.

Sandrin T.R., Chec A.M., Maier R.M. (2000) Appl Environ Microbiol 66:4585-4588.

Shepherd R., Rockey J., Shutherland I.W., Roller S. (1995) J Biotechnol 40: 207-217.

Singh P., Cameotra S. S. (2004) Trends in Biotechnology 22:142-146. Stangellini M.E., Miller R.M. (1997) Plant Disease 81 :4-12.

Stanghellini M.E., et al, (1998) U.S. Patent No. 5,767,090.

Sutton R. (1992) Water Science and Technology WSTED4, 26:2392-2396.

Tamburini E., et al., (2003) Environmental Microbiology 5:730-736.

Tan H., et al., (1994) Environ Sci Technol 28: 2402-2406. Thimon L., et al., (1995) FEMS Microbiol Lett 1995; 128: 101-6.

Toren A., Orr E., Paitan Y., Ron E.Z., Rosenberg E. (2002) J Bacteriol. 184:165-70.

Tumeo M., et al., (1994) Spill Sci Technol Bull 1: 53-59.

Velraeds M., et al., (1998) J. of Med. Microbiol. 47:1081-1085.

Volkering F., Breure A., Rulkens W. (1997) Bioremediation 8:401-417. Vollenbroich D., Ozel M., Vater J., Kamp R.M., Pauli G. (1997) Biologicals 25:289-297.

Weber W. J. Jr., Huang W., LeBoeuf E.J. (1999) Colloid Surface A. 151 :167-179.

Whyte L.G., et al., (1998) Appl Environ Microbiol. 64:2578-84.

Claims

1- A micro-organism belonging to the genus Gordonia able to produce and release extracellularly surface active compounds (SAC) and having a DNA sequence transcribed to 16 S RNA comprised in the group of: Seq ID No. 1, Seq ID No. 2 or Seq ID No. 3.

2- The micro-organism according to claim 1 being the strain named M22, BS25 or BS29.

3- Use of the micro-organism according to any one of previous claims for the production of surface active compounds. 4- Use of the micro-organism according to claim 1 or 2 for the degradation of hydrocarbons or derivatives thereof.

5- Use of the micro-organism according to claim 1 or 2 for environment decontamination process.

6- Use of the micro-organism according to claim 1 or 2 for biocatalytic and/or bioconversion process.

7- A surface active compound or a mixture thereof obtainable from the microorganism according to claim 1 or 2.

8- Use of the surface active compound or a mixture thereof according to claim 7 for the preparation of a medicament. 9-Use of the surface active compound or a mixture thereof according to claim 8 wherein the medicament has a disinfectant, anti-adhesive, anti-thrombosis, anti-microbial and/or anti- viral property. 10- Use of the surface active compound or a mixture thereof according to claim 7 for the preparation of a cosmetic product. 11- Use of the surface active compound or a mixture thereof according to claim 7 for the preparation of a food product. 12- Use of the surface active compound or a mixture thereof according to claim 7 for the preparation of an emulsion.

13-A pharmaceutical composition comprising an effective amount of the surface active compound or a mixture thereof according to claim 7.

14-An emulsion comprising the surface active compound or a mixture thereof according to claim 7.