Journal of Applied Microbiology ISSN 1364-5072

ORIGINAL ARTICLE

Antimicrobial potential of a lipopeptide biosurfactant derived from a marine Bacillus circulans

P. Das, S. Mukherjee and R. Sen
Department of Biotechnology, Indian Institute of Technology, Kharagpur, West Bengal, India

Keywords antimicrobial activity, HPLC purication, marine bacteria, MBC, MDR strains, MIC, microbial surfactant. Correspondence Ramkrishna Sen, Department of Biotechnology, Indian Institute of Technology, Kharagpur, West Bengal, India. E-mail: rksen@yahoo.com

Abstract Aims: To isolate the biologically active fraction of the lipopeptide biosurfactant produced by a marine Bacillus circulans and study its antimicrobial potentials. Methods and Results: The marine isolate B. circulans was cultivated in glucose mineral salts medium and the crude biosurfactant was isolated by chemical isolation method. The crude biosurfactants were solvent extracted with methanol and the methanol extract was subjected to reverse phase high-performance liquid chromatography (HPLC). The crude biosurfactants resolved into six major fractions in HPLC. The sixth HPLC fraction eluting at a retention time of 273 min showed the maximum surface tension-reducing property and reduced the surface tension of water from 72 mNm)1 to 28 mNm)1. Only this fraction was found to posses bioactivity and showed a pronounced antimicrobial action against a panel of Gram-positive and Gram-negative pathogenic and semi-pathogenic micro-organisms including a few multidrug-resistant (MDR) pathogenic clinical isolates. The minimum inhibitory concentration (MIC) and minimum bactericidal concentration (MBC) of this antimicrobial fraction of the biosurfactant were determined for these test organisms. The biosurfactant was found to be active against Gram-negative bacteria such as Proteus vulgaris and Alcaligens faecalis at a concentration as low as 10 lg ml)1. The biosurfactant was also active against methicillin-resistant Staphylococcus aureus (MRSA) and other MDR pathogenic strains. The chemical identity of this bioactive biosurfactant fraction was determined by post chromatographic detection using thin layer chromatography (TLC) and also by Fourier transform infrared (FTIR) spectroscopy. The antimicrobial HPLC fraction resolved as a single spot on TLC and showed positive reaction with ninhydrin, iodine and rhodamine-B reagents, indicating its lipopeptide nature. IR absorption by this fraction also showed similar and overlapping patterns with that of other lipopeptide biosurfactants such as surfactin and lichenysin, proving this biosurfactant fraction to be a lipopeptide. The biosurfactant did not show any haemolytic activity when tested on blood agar plates, unlike the lipopeptide biosurfactant surfactin produced by Bacillus subtilis. Conclusions: The biosurfactant produced by marine B. circulans had a potent antimicrobial activity against Gram-positive and Gram-negative pathogenic and semi-pathogenic microbial strains including MDR strains. Only one of the HPLC fractions of the crude biosurfactants was responsible for its antimicrobial action. The antimicrobial lipopeptide biosurfactant fraction was also found to be nonhaemolytic in nature. Signicance and impact of the study: This work presents a nonhaemolytic lipopeptide biosurfactant produced by a marine micro-organism possessing a pronounced antimicrobial action against a wide range of bacteria. There is a high demand for new antimicrobial agents because of the increased resistance
1675

2007 1144: received 18 July 2007, revised 7 November 2007 and accepted 13 November 2007
doi:10.1111/j.1365-2672.2007.03701.x

2008 The Authors Journal compilation 2008 The Society for Applied Microbiology, Journal of Applied Microbiology 104 (2008) 16751684

Biosurfactant with antimicrobial activity

P. Das et al.

shown by pathogenic micro-organisms against the existing antimicrobial drugs. This study provides an insight into the search of new bioactive molecules from marine micro-organisms.

Introduction Microbial surfactants or biosurfactants are surface-active amphipathic molecules produced by a number of microorganisms. They occur in nature as a diverse group of molecules comprising of glycolipids, lipopeptides and lipoproteins, fatty acids, neutral lipids, phospholipids, polymeric and particulate biosurfactants (Desai and Banat 1997; Mukherjee et al. 2006). Beside their regular roles in enhanced oil recovery, bioremediation and industrial emulsication, in recent years, microbial surfactants have been found to possess several properties of therapeutic and biomedical importance, e.g. antibacterial, antifungal and antiviral properties. They also have antiadhesive action against several pathogenic micro-organisms (Cameotra and Makkar 2004; Singh and Cameotra 2004; Rodrigues et al. 2006). Lipopeptides form the most widely reported class of biosurfactants having antimicrobial action. Among the lipopeptides, surfactin, produced by Bacillus subtilis is the rst and the most well-known member (Arima et al. 1968). Other antimicrobial lipopeptides include fengycin, iturin, bacillomycins and mycosubtilins produced by B. subtilis (Vater et al. 2002). Lichenysin (Yakimov et al. 1995) and pumilacidin (Naruse et al. 1990) produced by Bacillus licheniformis and Bacillus pumilus, respectively are the other antimicrobial lipopeptides. The production of these antimicrobials by Bacillus probiotics is one of major mechanisms by which they inhibit the growth of pathogenic micro-organisms in the gastrointestinal tract (Hong et al. 2005). Other biosurfactants that are reported of having antimicrobial actions are rhamnolipids produced by Pseudomonas aeruginosa (Abalos et al. 2001; Benincasa et al. 2004) and sophorolipids produced by Candida bombicola (Kim et al. 2002). Mannosylerythritol lipids (Kitamoto et al. 1993) (MEL-A and MEL-B) produced by Candida antarctica strain have also been reported to exhibit antimicrobial action against Gram-positive bacteria. Most of the biosurfactants having antimicrobial properties till date have been obtained from organisms isolated from terrestrial and contaminated sites. Marine environment forms a vast majority of the Earths surface; however there are only a few reports (Poremba et al. 1991; Schulz et al. 1991; Passeri et al. 1992; Abraham et al. 1998; Maneerat et al. 2006) of biosurfactant producers of marine origin. Their biosurfactants have not been evaluated for their antimicrobial potentials. In this paper, we
1676

report the isolation and purication of a lipopeptide biosurfactant from a marine Bacillus circulans and its antimicrobial potentials against pathogenic and semi-pathogenic bacterial strains along with a few multidrug-resistant (MDR) clinical isolates. Materials and Methods Micro-organism, production medium and culture conditions The biosurfactant producer was isolated from a marine water sample collected from the Andaman and Nicobar Islands, India. It was later characterized as Bacillus circulans. A glucose mineral salts (GMS) medium was used for the production of the biosurfactant. The GMS production medium which was utilized for biosurfactant production had the following composition: glucose (2%), NH4NO3 (03%), K2HPO4 (022%), KH2PO4 (0014%), NaCl (0001%), MgSO4 (006%), CaCl2 (0004%), FeSO4 (0002%) and 05 ml l)1 of trace elements solution containing (l)1): 232 g of ZnSO47H2O, 178 g of MnSO44H2O, 056 g of H3BO3, 10 g of CuSO45H2O, 039 g of Na2MoO42H2O, 042 g of CoCl26H2O, 10 g of EDTA, 0.004 g of NiCl26H2O and 066 g of KI. Zobell marine broth (Hi-Media, Mumbai, India) was used for the preparation of the primary inoculum. After overnight growth, the inoculum was transferred to 100 ml of sterile GMS production medium in 500-ml Erlenmeyer asks and incubated at 37C with shaking at 180 rev min)1 for 26 h. Isolation of surface-active molecules The crude biosurfactant molecules produced by this strain were obtained primarily by the chemical isolation method (Sen and Swaminathan 1997). Briey, after about 26 h of growth the culture broth was centrifuged at 10 000 g to pellet cells and obtain the supernatant. Concentrated HCl was added to the supernatant till its pH was lowered to 2. The acidied supernatant was then kept at 4C overnight for complete precipitation of the biosurfactant. The precipitate was centrifuged at 10 000 g for 20 min to get the crude biosurfactant as a pellet. This pellet was then resuspended in water; pH was raised to 75, lyophilized and solvent-extracted with methanol. The methanol-soluble fraction was transferred into another broad-mouthed con-

2008 The Authors Journal compilation 2008 The Society for Applied Microbiology, Journal of Applied Microbiology 104 (2008) 16751684

P. Das et al.

Biosurfactant with antimicrobial activity

tainer and allowed to dry at 60C. After the removal of methanol, minimum quantity of water was added to dissolve the biosurfactant, which was again lyophilized and stored at 20C for further studies. At this stage the biosurfactant was partially puried. Purication of biosurfactant by high-performance liquid chromatography High-performance liquid chromatography (HPLC) was used to resolve the different biosurfactant fractions in order to check their individual surface and antimicrobial activities. Reverse phase (RP)-HPLC was performed with a Zorbax Eclipse plusTM reverse phase column [C18, 5 lm, 46 (i.d.) 250 mm] on an Agilent 1100 series HPLC instrument with a diode array detector (DAD) system. The lyophilized methanol extract of the biosurfactant was re-dissolved in methanol to make a biosurfactant solution of 1 mg ml)1 which was passed through 022-lm lter (Millipore, Billerica, MA) and 50 ll of this was injected into the injector port of the instrument using a Hamilton HPLC syringe. The elution was done using a linear gradient of 595% of acetonitrile containing 01% (v v) triuoroacetic acid (TFA) and water for 40 min at a ow rate of 04 ml min)1. The absorbance of the eluent was measured at 210 nm. The resolved biosurfactant fractions were collected in a fraction collector for the activity studies. All the collected fractions were lyophilized, re-dissolved in water and surface tension of the individual fractions was measured using a digital surface tensiometer (DCAT, Data Physics Instrumnets GmbH, Filderstadt, Germany) using the principles of Wilhelmy plate technique. Thin layer chromatography The identity of the active biosurfactant fraction was determined by resolving the HPLC-puried active fraction on thin layer chromatography (TLC) followed by postchromatographic detection. The active biosurfactant fraction obtained from HPLC was dissolved in methanol and spotted on various lanes on a 10 10 cm precoated silica gel GF 254 plate (Merck, Darmstadt, Germany) with the help of a Linomat-5 sample applicator (CAMAG, Switzerland). The plate was then developed with a solvent system comprising chloroform, methanol and water in the ratio 65 : 25 : 4. The plates were visualized under a short-wave (254 nm) ultraviolet (UV)-emitting mercury lamp and the spots were marked to view in white light. The developed plates were then treated with 02% ninhydrin solution in absolute alcohol followed by heating at 110C to detect peptides and free amino groups and iodine and rhodamine-B reagents to detect lipids.

Fourier transform infrared spectroscopy Fourier transform infrared spectroscopy (FTIR) was used to determine the functional groups and the chemical bonds present in the biologically active fraction of the biosurfactant and thus determine its chemical nature. The FTIR analysis was performed by using a Nexus-870 FT-IR spectrometer (Thermo Electron Co., Yokohama, Japan) with sample dispersed in the pellets of KBr. The lyophilized active fraction of biosurfactant samples (0305 mg) were ground in about 80 mg of spectral-grade KBr (Merck, Germany) and pressed into pellets under about 56 tons cm)2 pressure with the help of a hydraulic press (Specac, Orpington, Kent, UK). The spectrometer was purged with dry nitrogen for about an hour to remove traces of carbon dioxide prior to spectral measurements. The spectral measurements were carried out in the absorbance mode. The spectra were normally acquired with the use of 4 cm)1 resolution yielding IR traces over the range of 400 4000 cm)1. The IR spectra of each sample were the average of 32 data scanning over the entire range of wave numbers. All data were corrected for background spectrum. The spectrum of a standard lipopeptide biosurfactant, surfactin (Sigma, Salt Lake City, UT) was also obtained in parallel. The two spectra were compared in order to conrm the chemical nature of the active biosurfactant fraction. Bacterial strains Antimicrobial tests were performed using the following test microbial strains procured from the National Collection of Industrial Microorganisms (NCIM, National Chemical Laboratory, Pune, India) and the Microbial Type Culture Collection (MTCC, IMTECH, Chandigarh, India). The test organisms included Gram-positive organisms like Micrococcus avus (NCIM 2376), Micrococcus luteus (NCIM 2673), Mycobacterium smegmatis (NCIM 5138), Leuconostoc mesenteroides (NCIM 2073), Streptococcus agalactii (NCIM 2401), Bacillus subtilis (NCIM 2063), Bacillus pumilis (MTCC 2466), B. pumilis (MTCC 2296), Lactobacillus brevis (NCIM 2090) and Lactobacillus lactis (NCIM 2369), B. subtilis (ATCC 21332) and Gram-negative bacteria like Escherichia coli (NCIM 2931), Serratia marcescens (NCIM 2397), Salmonella typhimurium (NCIM 2501), Proteus vulgaris (NCIM 2857), Citrobacter freundii (NCIM 2488), Proteus mirabilis (NCIM 2300), Alcaligenes faecalis (NCIM 2105), Acetobacter calcoaceticus (NCIM 2886), Bordetella bronchiseptica (NCIM 2267), Klebsiella aerogenes (NCIM 2098) and Enterobacter cloacae (NCIM 2164). The MDR (multi-drug resistant) human pathogenic bacterial strains like methicillin-resistant Staphylococcus aureus (MRSA), MDR E. coli and Klebsiella sp. were collected from the Peerless Hospital and B.K. Roy Research Centre, Kolkata, India (Table 1).
1677

2008 The Authors Journal compilation 2008 The Society for Applied Microbiology, Journal of Applied Microbiology 104 (2008) 16751684

Biosurfactant with antimicrobial activity

P. Das et al.

Table 1 Drug resistance prole of the multidrug-resistant pathogenic bacterial strains

Multidrug-resistant bacterial strains* Methicillin-resistant Staphylococcus aureus (MRSA) Escherichia coli

separate sterile glass vials. These stocks were used for the antimicrobial tests. Agar disc susceptibility test The susceptibility of the test strains against biosurfactant was tested with the help of the agar disc diffusion test performed on Mueller-Hinton agar (MHA; HiMedia) plates. Bacterial test strains including the MDR strains were cultured overnight in nutrient broth at 37C. The OD of these broth cultures were adjusted to 01 equivalents to an inoculum of c. 108 CFU ml)1 (according to McFarland turbidity standards). Exactly 500 ll of these was used to inoculate MHA plates by ooding their surface and spreading them uniformly on plates. The excess liquid broth was allowed to air dry for 10 min under the sterile environment of a laminar air ow hood (Klenzaids). Sterile Whatman lter paper (no. 1) discs of diameter 6 mm impregnated with 10 ll of the different stock solutions of the biosurfactant samples in methanol were kept aseptically on the surface of these preinoculated MHA plates. A disc soaked in methanol was kept as negative control. The plates were incubated at 37C for 24 h and after growth, the microbial growth inhibition zones around the wells and discs were measured using an antibiotic zone scale (HiMedia). All tests were performed in triplicate and the inhibition zone diameter values (mm) in Table 2 represent the mean value SD. Studentss t-test was performed to determine the statistically signicant differences in the zone of inhibition
Table 2 In vitro antimicrobial activity of the solvent extracted and HPLC-puried active biosurfactant fraction against a panel of Gram-positive and Gram-negative bacterium by agar disc diffusion method

Isolation site Nasal swab

Resistance pattern Resistant to methicillin and streptomycin Resistant to ciprooxacin, ooxacin, levooxacin, streptomycin and penicillin Resistant to ceftazidine, ciprooxacin, levooxacin, noroxacin, ooxacin, streptomycin and penicillin Resistant to ceftriaxone, ciprooxacin, ooxacin, levooxacin, noroxacin, piperacillin + tazobactam, streptomycin and penicillin

Ascitic uid

E. coli

Urine

Klebsiella pneumoniae

Stool

*Strains collected from B.K. Roy Research Centre, Peerless Hospital, Kolkata, India. As indicated by the hospital records.

Antimicrobial activity Preparation of biosurfactant stock solutions The stock solutions of crude (solvent extract) and HPLCpuried active fraction were prepared in methanol. For this 1 mg ml)1 of the solutions were prepared in methanol. These solutions were passed through 022-lm membrane lter (Whatman, Maidstone, Kent, UK) and kept in

Antimicrobial zone diameter (mm) Solvent-extracted biosurfactant (1000 lg ml)1)

Organism Gram-positive bacteria Micrococcus avus (NCIM 2376) Bacillus pumilis (MTCC 2296) Mycobacterium smegmatis (NCIM 5138) Gram-negative bacteria Escherichia coli (NCIM 2931) Serratia marcescens (NCIM 2397) Proteus vulgaris (NCIM 2857) Citrobacter freundii (NCIM 2488) Proteus mirabilis (NCIM 2300) Alcaligenes faecalis (NCIM 2105) Acetobacter calcoaceticus (NCIM 2886) Bordetella bronchiseptica (NCIM 2267) Klebsiella aerogenes (NCIM 2098) Enterobacter cloacae (NCIM 2164)

HPLC-puried bioactive fraction (1000 lg ml)1)

2666 057 2033 057 2200 000* 1966 240 1466 1766 1266 1766 1566 1533 1900 1366 152* 10 057 152 152 057 057 057 000 057*

1700 100 1533 057 1600 100 1466 1400 1066 1100 1033 1200 1166 1200 1200 1100 057 100 057 100 057 100 057 000 100 100

Zone diameters (mm) represent the mean SD of three independent readings. Signicant difference with respect to solvent-extracted biosurfactant: *P < 005, P < 001, P < 0001. HPLC, high-performance liquid chromatography.
2008 The Authors Journal compilation 2008 The Society for Applied Microbiology, Journal of Applied Microbiology 104 (2008) 16751684

1678

P. Das et al.

Biosurfactant with antimicrobial activity

produced by HPLC-puried active biosurfactant fraction with respect to the crude biosurfactant. Minimum inhibitory concentrations and minimum bactericidal concentrations Minimum inhibitory concentration (MIC) is the minimum concentration of an antimicrobial compound at which it inhibits the growth of bacteria. Thus at the MIC the antimicrobial is bacteriostatic, i.e. it prevents the spread of bacterial infection. Minimum bactericidal concentration (MBC) is the minimum concentration or dose of an antimicrobial agent at which it becomes lethal. The microbial culture when treated with an antimicrobial agent at or above its MBC cannot be revived even if transferred to an antimicrobial-free growth-supporting medium (Celiktas et al. 2007; Kuete et al. 2007). The MBC value of a particular antimicrobial agent is either equal or greater than its MIC value for a particular microbial strain. Two standard methods namely, the agar disc diffusion assay and broth microdilution susceptibility assay were used for the determination of MIC and MBC, respectively for the different test organisms including the MDR pathogenic strains. MIC: agar disc diffusion test The agar disc diffusion test was carried out to determine the MIC of the biosurfactants against different test organisms. The MHA plates inoculated with the test organisms were prepared as discussed earlier. Sterile Whatman (no. 1) lter paper discs were impregnated with 10 ll of different dilutions of the puried active biosurfactant fraction. These were then kept aseptically on the surface of the preinoculated MHA plates. The plates were incubated under normal atmospheric conditions at 37C for 24 h. Parallel sets of experiments were conducted with standard antibiotics such as penicillin (b-lactam antibiotic) and streptomycin (aminoglycoside antibiotic) (Hi-Media). The minimum concentration of the biosurfactant or standard antibiotics which produced a clear zone of inhibition around the application disc was considered as their MIC against that particular organism. All the experiments for determination of MIC were performed in triplicate. MBC: broth microdilution susceptibility assay A broth microdilution assay was also performed, as per the recommendations of NCCLS (1999) for the determination of the MBC (NCCLS 2000). Overnight grown cultures of the bacterial test strains, adjusted to a nal density of 108 CFU ml)1 were used to inoculate 96-well microtitre plates containing serially diluted puried active biosurfactant fraction (10 001 mg ml)1) in MHB (Mueller Hinton Broth). The same tests were performed simultaneously for negative control

(only MHB), growth control (MHB + test organism) and sterility control (MHB + biosurfactant). Plates were incubated under normal atmospheric conditions at 37C for 24 h. Bacterial growth was monitored by absorbance at 600 nm in a microtitre plate reader (Multiskan Spectrum, Technocorp, Multiskan, Waltham, MA). The wells that showed >90% inhibition of growth were identied and 5 ll of contents of each well was transferred to nutrient agar (NA) plates and incubated at 37C for 24 h. The lowest concentration showing no revival of the test culture on NA plates was considered as the MBC. The susceptibility of the micro-organisms against penicillin and streptomycin (HiMedia) was also checked with the same experimental set up, with only the biosurfactant being replaced by either penicillin or streptomycin. All the tests for determination of MBC were performed in triplicate. Analysis of haemolytic activity The haemolytic potential of the biosurfactant was evaluated on blood agar plates. Blood agar plates were prepared by adding blood from a voluntary human donor into sterile blood agar base medium and by allowing the plates to solidify under aseptic environment of a laminar air ow. After solidication of agar the plates were divided into two parts. In one part of the blood agar plate, B. subtilis ATCC 21332, a well-known surfactin-producing strain was streaked, while the marine B. circulans was streaked on the other part of the plate. The plates were incubated under normal atmospheric conditions at 37C overnight. After overnight incubation, the plates were checked for the zone of haemolysis. Results Isolation and partial purication of the biosurfactant The crude biosurfactants obtained by chemical isolation method (acidication of the cell-free broth) were extracted with methanol to obtain the partially puried biosurfactants. These solvent-extracted partially puried biosurfactants were then subjected to reverse phase HPLC on a C18 reverse phase column. With a linear gradient of acetonitrile and water, the crude biosurfactant was resolved into six fractions. The major six fractions were eluted at retention times 33, 55, 202, 225, 235 and 273 min, respectively (Fig. 1). Among all the HPLC fractions the surface tension of the sixth major fraction (retention time: 273 min) was found to be lowest, i.e. 28 mNm)1. Chemical nature of the bioactive fraction The antimicrobial biosurfactant fraction was seen as a single spot on TLC. This fraction showed positive reaction
1679

2008 The Authors Journal compilation 2008 The Society for Applied Microbiology, Journal of Applied Microbiology 104 (2008) 16751684

Biosurfactant with antimicrobial activity

P. Das et al.

mAU 300 250 200 150 100 50 0 0

Absorbance at 210 nm

1 3 4

1 095 Transmittance 09 085 08 075 07 065 Bioactive fraction Surfactin

6 (Bioactive fraction)

10

15 20 25 30 35 Retention time (min)

40

45 min

Figure 1 Separation of the biosurfactant fractions on reverse phase high-performance liquid chromatography (HPLC). The solventextracted biosurfactants were resolved in a C-18 column using a linear gradient of 595% acetonitrile (01% triuoroacetic acid) and water. The absorbance was measured at 210 nm for detection of lipopeptides.

06 4000 3500 3000 2500 2000 1500 1000 Wavenumbers [1/cm]

500

Figure 3 Fourier transform infrared spectra (FTIR) spectra of the high-performance liquid chromatography (HPLC)-puried bioactive fraction of the biosurfactant produced by marine Bacillus circulans overlapped with FTIR spectra of the standard surfactin (Sigma Chemicals).

Figure 2 Thin layer chromatography of crude and reverse phase (RP)high-performance liquid chromatography (HPLC)-derived antimicrobial biosurfactant fraction on silica gel GF254 precoated plates. The plates were developed with a solvent system comprising of chloroform : methanol : water (65 : 25 : 4). Lanes A and B, crude and active biosurfactant fractions, respectively under 254-nm ultraviolet light; lanes C and D, crude and active biosurfactant fractions, respectively developed with iodine vapours; lanes E and F, crude and active biosurfactant fractions, respectively developed with 02% ninhydrin solution.

ing the presence of CCH3 bonding or long alkyl chains. The peak with highest absorbance in the spectrum was observed at 1656 cm)1. Absorbance in this region signies the presence of peptide group in the molecule. Adjacent to this peak was another high intensity peak at 1545 cm)1. Absorbance in these regions are because of the presence of C=O bonds and is caused due to C=O stretching vibrations. Other signicant peaks observed at 1459 and 1403 cm)1 correspond to CH bending vibrations and is common in compounds with alkyl chains. Peaks at 1238 and 1118 cm)1 are probably because of COC vibrations in esters. The lipopeptide biosurfactant, surfactin (Sigma) also yielded a similar IR absorption pattern and absorbed approximately at the same wave number positions. Antimicrobial action of biosurfactant The solvent-extracted biosurfactant showed antimicrobial activity against majority of the organisms tested. The sixth major HPLC fraction (retention time: 273 min) was primarily found to be responsible for antimicrobial activity of the biosurfactant evident from the large inhibition zones (Fig. 4) against the test organisms as shown in Table 2. The solvent extracted as well as the HPLC-puried active biosurfactant fraction showed activity against Gram-positive bacteria such as B. pumilis (MTCC 2296), M. avus (NCIM 2376), M. luteus (NCIM 2673) and Myco. smegmatis (NCIM 5138) and Gram negatives like E. coli (NCIM 2931), Ser. marcescens (NCIM 2397), Pr. vulgaris (NCIM 2857), Cit. freundii (NCIM 2488), Pr. mirabilis (NCIM 2300), Alc. faecalis (NCIM 2105), Acet. calcoaceticus (NCIM 2886), Bord. bronchiseptica (NCIM

with ninhydrin reagent and iodine vapour indicating the presence of peptide and lipid moieties in the molecule (Fig. 2). The lipopeptide nature of the biosurfactant was further conrmed by the IR spectra of the compound (Fig. 3). A broad absorbance with wave numbers ranging approximately from 3500 cm)1 to 3200 cm)1 having its maxima at 3305 cm)1 was seen. Absorbance in this region is caused as a result of CH stretching vibrations and NH stretching vibrations and is a characteristic of carbon-containing compounds with amino groups. An absorbance in this region also signies the presence of intramolecular hydrogen bonding. Two other sharp absorbance peaks are seen at 2960 and 2927 cm)1 signify1680

2008 The Authors Journal compilation 2008 The Society for Applied Microbiology, Journal of Applied Microbiology 104 (2008) 16751684

P. Das et al.

Biosurfactant with antimicrobial activity

Minimum inhibitory concentration and minimum bactericidal concentration

6 5 C 4 3 1 2

Figure 4 Antimicrobial activity of the high-performance liquid chromatography (HPLC)-puried biosurfactant fractions against Micrococcus avus. The numbers (16) denote the HPLC-eluted fractions dissolved in methanol. Methanol (C) was taken as control in one of the wells.

2267), Kl. aerogenes (NCIM 2098) and Ent. cloacae (NCIM 2164). Largest inhibition zones by the biosurfactant were those produced against M. avus NCIM 2376, E. coli NCIM 2931, Myco. smegmatis NCIM 5138 and B. pumilis MTCC 2296. The zones produced by HPLCpuried active fraction of the biosurfactant were signicantly larger than those produced by the solvent-extracted crude biosurfactant (Table 2). The biosurfactant was also found to have inhibitory action against the MDR strains.

The MIC and MBC of the biosurfactant were found using agar disc diffusion method and broth microdilution susceptibility assay (Table 3). The puried biosurfactant was able to inhibit bacterial growth in concentrations as low as 10 lg ml)1. Proteus vulgaris (NCIM 2857) and Alc. faecalis (NCIM 2105) were affected at such a low biosurfactant concentration. However, the MIC values for most of the tested bacteria ranged from 30 to 80 lg ml)1. Bordetella bronchiseptica (NCIM 2267) and M. avus (NCIM 2376) were seen to be inhibited at relatively higher biosurfactant concentrations of 100 and 200 lg ml)1, respectively. The parallel experiments with penicillin and streptomycin helped to reveal the relative efcacy of the biosurfactant product as an antimicrobial agent. As a general observation, the MIC values for the biosurfactant against the test organisms were found to be lesser than both the antibiotics, particularly penicillin. A resistance to penicillin was seen in case of Cit. freundii (NCIM 2488), Myco. smegmatis (NCIM 5138) and Ent. cloacae (NCIM 2164), which were not inhibited up to a concentration of 1000 lg ml)1 of penicillin in the growth medium. Apart from these test strains, the puried biosurfactant was also found to be effective against the MDR clinical isolates. The biosurfactant inhibited the MDR Klebsiella sp. at concentrations as low as 60 lg ml)1. However, the two MDR strains of E. coli were inhibited at relatively higher

Table 3 Minimum inhibitory concentration (MIC) and minimum bactericidal concentration (MBC) of the active biosurfactant fraction and standard antibiotics penicillin and streptomycin against different Gram-positive and Gram-negative bacterial strains
Antimicrobial agent Biosurfactant MIC* (lg ml)1) 40 200 30 30 10 30 20 50 10 40 100 80 20 MBC (lg ml)1) 80 400 60 60 20 60 200 100 20 80 200 300 60 Penicillin MIC* (lg ml)1) 10 30 300 300 800 >1000 600 >1000 400 800 80 500 >1000 MBC (lg ml)1) 30 60 400 400 900 NT 1000 NT 600 1000 200 800 NT Streptomycin MIC* (lg ml)1) 10 400 40 30 20 30 20 30 800 20 200 >1000 80 MBC (lg ml)1) 50 600 100 50 40 60 50 50 1000 100 1000 NT 100

Bacterial strains Escherichia coli NCIM 2931 Micrococcus avus NCIM 2376 Serratia marcescens NCIM 2397 Bacillus pumilis MTCC 2296 Proteus vulgaris NCIM 2857 Citrobacter freundii NCIM 2488 Proteus mirabilis NCIM 2300 Mycobacterium smegmatis NCIM 5138 Alcaligens faecalis NCIM 2105 Acetobacter calcoaceticus NCIM 2886 Bordetella bronchiseptica NCIM 2267 Klebsiella aerogenes NCIM 2098 Enterobacter cloacae NCIM 2164

NT, not tested as MIC was not determined. *MIC determined by agar disc diffusion method. MBC determined by broth microdilution assay.
2008 The Authors Journal compilation 2008 The Society for Applied Microbiology, Journal of Applied Microbiology 104 (2008) 16751684

1681

Biosurfactant with antimicrobial activity

P. Das et al.

Table 4 Minimum inhibitory concentration (MIC) and minimum bactericidal concentration (MBC) of the active biosurfactant fraction against multidrug-resistant pathogenic bacterial strains
Biosurfactant Multidrug-resistant bacterial strains* Methicillin-resistant Staphylococcus aureus (MRSA) Escherichia coli E. coli Klebsiella pneumoniae MIC (lg ml)1) 200 800 200 60 MBC (lg ml)1) 500 1000 600 200

*Strains collected from B.K. Roy Research Centre, Peerless Hospital, Kolkata, India. MIC determined by agar disc diffusion method. MBC determined by broth microdilution assay.

concentrations of 200 and 800 lg ml)1. The MRSA was inhibited at 200 lg ml)1 concentration of the biosurfactant. These MDR strains were found to be insensitive to up to 1000 lg ml)1 concentration of penicillin and streptomycin along with other drugs listed in the table. These MDR pathogenic strains however were found to be inhibited by much lower concentrations of the puried biosurfactant. The MIC and MBC values of the biosurfactant against these strains are listed in Table 4. Haemolytic properties of the biosurfactant After overnight incubation, a clear zone of haemolysis was observed on that half of the blood agar plate which was streaked with B. subtilis ATCC 21332. The haemolysis is caused by this strain because of the production of the lipopeptide biosurfactant, surfactin. On the other half of the plate on which the marine B. circulans was streaked, no haemolysis was observed. The crude as well as the puried biosurfactant from our strain did not show any haemolysis in contrast to the biosurfactant isolated from B. subtilis ATCC 21332 and standard surfactin (Sigma) which showed a prominent haemolytic activity (data not shown). Discussions The antimicrobial properties of the biosurfactants have been widely reported. However, the biosurfactants with antimicrobial properties reported till date is produced mostly by the micro-organisms of terrestrial origin. The oceans cover more than 70% of the Earths total surface and supports very rich and diverse microora. However, the number of reports on marine antimicrobial biosurfactant molecules is negligible. Although there have been few
1682

reports of marine biosurfactant producers, their antimicrobial potentials have not been explored in details. This problem was identied in the present work and the biosurfactant isolated from marine bacteria was tested for antimicrobial action against a battery of pathogenic and semi-pathogenic test organisms including a few MDR strains. Although the solvent-extracted biosurfactants obtained from this strain could be resolved into a few fractions by HPLC, only a major fraction eluting at a retention time of 273 min displayed pronounced antimicrobial action. The surface tension of this fraction was found to be the lowest (28 mNm)1) indicating its powerful surface tension-reducing property. The biosurfactants produced by certain strains of Bacillus sp. and Pseudomonas sp. are a mixture of different lipopeptide species or isoforms (Naruse et al. 1990; Abalos et al. 2001; Vater et al. 2002; Benincasa et al. 2004; Mukherjee and Das 2005). They differ in the length and the type of aliphatic carbon chain or in the amino acid residues in the peptide part in case of lipopeptides and sugars in case of glycolipids, in their molecules. The potential of a particular lipopeptide species to reduce surface tension or exhibit antimicrobial property depends largely on its molecular structure. These different species of the biosurfactants are separated by chromatography because of slight differences in their properties arising out of minute structural dissimilarities. The antimicrobial fraction of the biosurfactant was found to be a lipopeptide as it showed positive reaction with ninhydrin reagent showing the presence of peptide. It also reacted separately with both iodine vapours and rhodamine-B reagent indicating the presence of a lipid part in the molecule. The lipopeptide nature of the biosurfactant was further conrmed by comparing the IR spectra of this fraction with pure surfactin from Sigma. The biosurfactant and surfactin both absorbed at approximately the same wave number positions and showed an overlapping pattern. This type of FTIR spectra is characteristic of lipopeptides, e.g. other lipopeptide biosurfactants like lichenysin reported in literature have also yielded similar IR absorption spectra (Lin et al. 1994; Yakimov et al. 1995).The degree of the purication obtained for the biosurfactant was reected by inhibition zone diameters in the sense that the zone produced by equal concentrations of puried compound were signicantly larger and well dened than those of the solvent-extracted biosurfactants. The isolated biosurfactant nonselectively showed activity against both Gram-positive and Gramnegative bacterial strains. This is quite contrasting to earlier reports on antimicrobial actions of the biosurfactants where the lipopeptide biosurfactants have been reported to be active mostly against Gram-positive bacteria (Kitamoto et al. 1993; Singh and Cameotra 2004). Furthermore, the puried biosurfactant inhibited Gram-negative bacte-

2008 The Authors Journal compilation 2008 The Society for Applied Microbiology, Journal of Applied Microbiology 104 (2008) 16751684

P. Das et al.

Biosurfactant with antimicrobial activity

ria like Alc. faecalis and Pr. vulgaris at concentrations less than 10 lg ml)1. On the other hand, rhamnolipids (a type of glycolipid biosurfactant) produced by Ps. aeruginosa have been reported to be active against Gram-negative bacteria. For example, rhamnolipid mixture M7 produced by Ps. aeruginosa AT10 have been reported to be active against both Gram-positive and Gram-negative organisms (Abalos et al. 2001). A comparative analysis of the MIC values for rhamnolipid M7 (Abalos et al. 2001) and our biosurfactant products against Gram-negative bacteria shows that the puried biosurfactant from this marine micro-organism is comparatively more efcient in inhibiting Gram negatives such as Pr. mirabilis (60 lg ml)1), Cit. freundii (30 lg ml)1) and Alc. faecalis (10 lg ml)1). This biosurfactant was also found to be more effective than rhamnolipids produced by Ps. aeruginosa LBI (Benincasa et al. 2004) in inhibiting Gram-negative pathogens like E. coli (40 lg ml)1) and Bord. bronchiseptica (60 lg ml)1). This shows the effectiveness of this lipopeptide biosurfactant in inhibiting Gram-negative microbial strains along with Gram-positive strains. In general, the lesser MIC and MBC values obtained for the puried biosurfactants in case of Gram negatives like Cit. freundii (30 lg ml)1), Pr. vulgaris (<10 lg ml)1), Alc. faecalis (<10 lg ml)1), Acet. calcoaceticus (40 lg ml)1), Kl. aerogenes (80 lg ml)1) and Ent. cloacae (20 lg ml)1) and Gram positives such as Myco. smegmatis (50 lg ml)1) compared with that of penicillin and streptomycin indicates that a much lower concentration of this compound can prove to be effective in treating infections caused by these pathogens. The lipopeptide biosurfactants had fairly good antimicrobial action against the MDR strains used in this study. Klebsiella is a Gram-negative nonmotile genus of pathogens causing a wide variety of diseased states in humans including pneumonia, urinary tract infections and septicemia and soft tissue infections. The lipopeptide was found to be effective against a MDR strain of this human pathogen at a dosage as low as 60 lg ml)1, showing its effectiveness and potential use in treating such infections. MRSA is a notorious hospital pathogen insensitive to practically all b-lactam antibiotics (Hiratmatsu et al. 2001). The biosurfactant was effective against this pathogen, although at a relatively higher concentration of 200 lg ml)1. In the haemolysis experiment, part of the blood agar plate, which was inoculated and incubated with our microbial culture did not show any sign of haemolytic activity in contrast to the prominent haemolysis that occurred on the other part of the agar medium inoculated with B. subtilis ATCC 21332 culture. The puried biosurfactant was also detected to be nonhaemolytic in contrast to standard surfactin (Sigma) as well as surfactin produced by B. subtilis strain. This in turn advocates for the safe use of the biosurfactant

product as a potential antimicrobial drug candidate in humans and animals, when most potent biosurfactants with antimicrobial properties disqualify as a drug for humans for the simple reason that they cause haemolysis of the human red blood cells (RBC; Symmank et al. 2002). The current scenario is such that the drug resistance among the pathogenic organisms for many lifethreatening diseases is on the rise because of indiscriminate use of the presently employed antibiotics. Thus these new biomolecules from the marine micro-organisms can provide an alternative to the existing antimicrobial agents. Acknowledgements P. Das acknowledges IIT, Kharagpur and S. Mukherjee acknowledges CSIR, New Delhi for the nancial assistances. R. Sen acknowledges the Department of Biotechnology (DBT), Government of India for the project grant (BT PR-6827 AAQ 03 263 2005) in marine biotechnology. The authors also gratefully acknowledge Dr Satadal Das, B.K. Roy Research Centre, Peerless Hospital, Kolkata, India for supplying the MDR strains. References
Abalos, A., Pinazo, A., Infante, M.R., Casals, M., Garcia, F. and Manresa, A. (2001) Physicochemical and antimicrobial properties of new rhamnolipids produced by Pseudomonas aeruginosa AT10 from soybean oil renery wastes. Langmuir 17, 13671371. Abraham, W.R., Meyer, H. and Yakimov, M. (1998) Novel glycine containing glucose lipids from the alkane using bacterium Alcanivorax borkumensis. Biochim Biophys Acta 1393, 5762. Arima, K., Kakinuma, A. and Tamura, G. (1968) Surfactin, a crystalline peptide lipid surfactant produced by Bacillus subtilis: isolation, characterization and its inhibition of brin clot formation. Biochem Biophys Res Commun 31, 488494. Benincasa, M., Abalos, A., Oliveira, I. and Manresa, A. (2004) Chemical structure, surface properties and biological activities of biosurfactant produced by Pseudomonas aeruginosa LBI from soapstock. Antonie van Leeuwenhoek 85, 18. Cameotra, S.S. and Makkar, R.S. (2004) Recent applications of biosurfactants as biological and immunological molecules. Curr Opin Microbiol 7, 262266. Celiktas, O.Y., Kocabas, E.E.H., Bedir, E., Sukan, F.V., Ozek, T. and Baser, K.H.C. (2007) Antimicrobial activities of methanol extracts and essential oils of Rosmarinus ofcinalis, depending on location and seasonal variations. Food Chem 100, 553559. Desai, J.D. and Banat, I.M. (1997) Microbial production of surfactants and their commercial potential. Microbiol Mol Biol Rev 61, 4764.

2008 The Authors Journal compilation 2008 The Society for Applied Microbiology, Journal of Applied Microbiology 104 (2008) 16751684

1683

Biosurfactant with antimicrobial activity

P. Das et al.

Hiratmatsu, K., Cui, L., Kuroda, M. and Ito, T. (2001) The emergence and evolution of methicillin-resistant Staphylococcus aureus. Trends Microbiol 9, 486493. Hong, H.A., Duc, L.H. and Cutting, S.M. (2005) The use of bacterial spore formers as probiotics. FEMS Microbiol Rev 29, 813835. Kim, K.J., Yoo, D.S., Kim, Y.B., Lee, B.S., Shin, D.H. and Kim, E.K. (2002) Characteristics of sophorolipid as an antimicrobial agent. J Microbiol Biotechnol 12, 235241. Kitamoto, D., Yanagishita, H., Shinbo, T., Nakane, T., Kamisawa, C. and Nakahara, T. (1993) Surface-active properties and antimicrobial activities of mannosylerythritol lipids as biosurfactants produced by Candida antarctica. J Biotechnol 29, 9196. Kuete, V., Nguemeving, J.R., Beng, V.P., Azebaze, A.G.B., Etoa, F.X., Meyer, M., Bodo, B. and Nkengfack, A.E. (2007) Antimicrobial activity of the methanolic extracts and compounds from Vismia laurentii De Wild (Guttiferae). J Ethnopharmacol 109, 372379. Lin, S.C., Minton, M.A., Sharma, M.M. and Georgiou, G. (1994) Structural and immunological characterization of a biosurfactant produced by Bacillus licheniformis JF-2. Appl Environ Microbiol 60, 3138. Maneerat, S., Bamba, T., Harada, K., Kobayashi, A., Yamada, H. and Kawai, F. (2006) A novel crude oil emulsier secreted in the culture supernatant of a marine bacterium, Myroides sp. strain SM1. Appl Microbiol Biotechnol 70, 254259. Mukherjee, A.K. and Das, K. (2005) Correlation between diverse cyclic lipopeptides production and regulation of growth and substrate utilization by Bacillus subtilis strains in a particular habitat. FEMS Microbiol Ecol 54, 479489. Mukherjee, S., Das, P. and Sen, R. (2006) Towards commercial production of microbial surfactants. Trends Biotechnol 24, 509515. Naruse, N., Tenmyo, O., Kobaru, S., Kamei, H., Miyaki, T., Konishi, M. and Oki, T. (1990) Pumilacidin, a complex of new antiviral antibiotics production, isolation, chemical properties, structure and biological activity. J Antibiot (Tokyo) 43, 267280. NCCLS (National Committee for Clinical Laboratory Standards) (1999) Performance standards for antimicrobial susceptibility testing, 6th edn. Approved Standards, M2A6. Wayne, PA.

NCCLS (National Committee for Clinical Laboratory Standards) (2000) Methods for dilution antimicrobial susceptibility tests for bacteria that grow aerobically. Approved standard, M7A5. Wayne, PA. Passeri, A., Schmidt, M., Haffner, T., Wray, V., Lang, S. and Wagner, F. (1992) Marine biosurfactants. IV. Production, characterization and biosynthesis of an anionic glucose lipid from the marine bacterial strain MM1. Appl Microbiol Biotechnol 37, 281286. Poremba, K., Gunkel, W., Lang, S. and Wagner, F. (1991) Marine biosurfactants. III. Toxicity testing with marine microorganisms and comparison with synthetic surfactants. Z Naturforsch 46c, 210216. Rodrigues, L., Banat, I.M., Teixeira, J. and Oliveira, R. (2006) Biosurfactants: potential applications in medicine. J Antimicrob Chemother 57, 609618. Schulz, D., Passeri, A., Schmidt, M., Lang, S., Wagner, F., Wray, V. and Gunkel, W. (1991) Marine biosurfactants, I. Screening for biosurfactants among crude oil degrading marine microorganisms from the North Sea. Z Naturforsch 46c, 197203. Sen, R. and Swaminathan, T. (1997) Application of responsesurface methodology to evaluate the optimum environmental conditions for the enhanced production of surfactin. Appl Microbiol Biotechnol 47, 358363. Singh, P. and Cameotra, S.S. (2004) Potential applications of microbial surfactants in biomedical sciences. Trends Biotechnol 22, 142146. Symmank, H., Franke, P., Saenger, W. and Bernhard, F. (2002) Modication of biologically active peptides: production of a novel lipohexapeptide after engineering of Bacillus subtilis surfactin synthetase. Protein Eng 15, 913921. Vater, J., Kablitz, B., Wilde, C., Franke, P., Mehta, N. and Cameotra, S.S. (2002) Matrix-assisted laser desorption ionization-time of ight mass spectrometry of lipopeptide biosurfactants in whole cells and culture ltrates of Bacillus subtilis C-1 isolated from petroleum sludge. Appl Environ Microbiol 68, 62106219. Yakimov, M.M., Timmis, K.N., Wray, V. and Fredrickson, H.L. (1995) Characterization of a new lipopeptide surfactant produced by thermotolerant and halotolerant subsurface Bacillus licheniformis. Appl Environ Microbiol 61, 17061713.

1684

2008 The Authors Journal compilation 2008 The Society for Applied Microbiology, Journal of Applied Microbiology 104 (2008) 16751684