Biofilm disruption potential of a glycolipid biosurfactant from marine Brevibacterium casei


  • Editor: Roger Bayston

Correspondence: Joseph Selvin, Department of Microbiology, School of Life Sciences, Bharathidasan University, Tiruchirappalli 620024, India. Tel.: +91 431 240 7082; fax: +91 431 240 7084; e-mail: or


The antibiofilm activity of a glycolipid biosurfactant isolated from the marine actinobacterium Brevibacterium casei MSA19 was evaluated against pathogenic biofilms in vitro. The isolate B. casei MSA19 was a potential biosurfactant producer among the 57 stable strains isolated from the marine sponge Dendrilla nigra. The biosurfactant production was optimized under submerged fermentation. The purified glycolipid showed a broad spectrum of antimicrobial activity. Based on the minimum inhibitory concentration/minimum bactericidal concentration ratio, the glycolipid was determined as bacteriostatic. The glycolipid biosurfactant disrupted the biofilm formation under dynamic conditions. The disruption of the biofilm by the MSA19 glycolipid was consistent against mixed pathogenic biofilm bacteria. Therefore, the glycolipid biosurfactant can be used as a lead compound for the development of novel antibiofilm agents.


Biofilms are complex aggregation of microorganisms growing on a solid substrate (Davis et al., 1989). They are mainly formed on biotic or abiotic surfaces as single- or multiple-species communities. Biofilm formation is an important mechanism for microbial survival in the environment. Biofilm-forming microorganisms are less susceptible to many antimicrobial agents as well as other biocides. Biofilm formation on medical devices plays an important role in the problem of many nosocomial and health-associated diseases. Based on the above reasons, novel antibiofilm compounds are required for the prevention/control of pathogenic microbial biofilms on the surfaces and hosts. It was established that most of the marine organisms have evolved efficient strategies to combat epibiosis. Especially, marine sponges produce specific deterrents to ward off biofilm-forming microorganisms (Selvin et al., 2010). However, it has been hypothesized that these toxic deterrents might be produced by the associated microorganisms instead of the host sponge.

Biosurfactants are a heterogeneous group of bioactive amphiphilic molecules produced mostly on microbial cell surfaces or extracellularly (Karanth et al., 1999; Soumen et al., 2006). The most significant advantages of microbial surfactants are biodegradability and nontoxicity to natural environments (Banat, 1993). The biomedical importance of biosurfactants was established due to their antibacterial, antifungal and antiviral properties; inhibition of fibrin clot formation; and their antiadhesive action against several pathogenic microorganisms (Meylheuc et al., 2001, 2006; Singh & Cameotra, 2004; Rodrigues et al., 2006). Sponge-associated marine bacteria are emerging as a potential source of novel biosurfactants (Gandhimathi et al., 2009; Kiran et al., 2009, 2010). It has been hypothesized that the antimicrobial fouling process represents a chemical defense of host sponges mediated by the associated bacteria. Therefore, a glycolipid biosurfactant produced by the sponge-associated marine actinobacterium was evaluated for the control of pathogenic biofilms. In the present study, we report the biofilm disruption potential of Brevibacterium casei MSA19, a glycolipid biosurfactant producer, isolated from the marine sponge Dendrilla nigra.

Materials and methods

Isolation and screening of a biosurfactant producer

The host sponge D. nigra was collected from the Vizhinjam coast (southwest coast of India) by SCUBA diving at 12 m depth. Only unbroken samples were used for microbiological analysis to avoid cross contamination. The collected specimens were kept for 2 h in sterilized aged seawater (SAS) to remove loosely associated microorganisms from the inner and outer sponge surfaces. One cubic centimeter of sponge tissue was excised from the middle of the entire sponge using sterile scissors. The excised portion was washed three times with SAS to remove any bacteria within the current canals and then the tissue was homogenated using a tissue homogenizer (Omni). The resultant homogenate was serially diluted with SAS and preincubated at 40 °C for 1 h for the activation of dormant and inactive cells. The resultant aliquot was plated on various isolation media and incubated at 25 °C under dark aerobic conditions until visible colonies appeared. Standard bacteriological media (Himedia) such as Zobell marine agar, seawater agar, nutrient agar supplemented with 2% NaCl, actinomycetes agar supplemented with 2% NaCl and marine sponge agar (Selvin et al., 2004) were used for the isolation of sponge-associated bacteria. Only consistent morphotypes in subsequent subcultures were selected for screening. The biosurfactant activity was determined based on the hemolytic activity (Carillo et al., 1996), drop-collapsing test (Youssef et al., 2004), oil displacement test (Morikawa et al., 1993), emulsification index (Paraszkiewicz et al., 1992) and lipase activity (Kiran et al., 2009). The isolate MSA19 was optimized under submerged fermentation conditions (Gandhimathi et al., 2009) for enhanced biosurfactant production.

Chemical characterization of glycolipid

The strain MSA19 was grown in a 250-mL Erlenmeyer flask containing 100 mL of actinomycetes isolation broth supplemented with 2% NaCl and incubated at 26 °C for 7 days at 200 r.p.m. The culture was centrifuged (Eppendorf) at 12 000 g for 10 min at 4 °C and the supernatant was filtered through a 0.2-μm filter. The cell-free supernatant and the pellet were acidified with concentrated HCl to obtain pH 2.0 and extracted with an equal volume of solvents such as ethyl acetate, diethyl ether and dichloromethane. The resultant aliquot was evaporated in a rotary vacuum evaporator (Yamato) and tested for emulsification activity as described above. The orcinol assay (Chandrasekaran & Bemiller, 1980) was used to determine the amount of glycolipids in the sample. Glycolipid production was confirmed by thin-layer chromatography (TLC) using chloroform–methanol–water (65 : 25 : 4) as a solvent system (Zhang & Miller, 1994). The TLC-purified fraction was analyzed in preparative HPLC, GC-MS (Shimadzu 2014) and Fourier transform infrared spectroscopy (FT-IR) to confirm the purity and structural moieties of the surface-active molecule. The antimicrobial activity of the TLC-purified compound was evaluated using the double-layer cylinder plate method (Gandhimathi et al., 2008). The list of pathogenic and nonpathogenic bacteria used in the antimicrobial assay is presented in Table 1. The minimum inhibitory concentration (MIC) and the minimum bactericidal concentration (MBC) were determined using the microdilution method (Selvin et al., 2009).

Table 1.   Panel of pathogens used for antimicrobial activity, biofilm formation and biofilm disruption assay
Pathogens usedAccession numberNature of the strains
Candida albicansFC1Clinical isolate
Escherichia coliMTCC 2939Human pathogen
Proteus mirabilisPC1Clinical isolate
Hemolytic StreptococcusPC2Clinical isolate
Pseudomonas aeruginosaMTCC 2453Human pathogen
Klebsiella pneumoniaePC3Clinical isolate
Vibrio parahaemolyticusMTCC 451Fish pathogen
Vibrio harveyiMTCC 3438Fish pathogen
Vibrio alginolyticusMTCC 4439Fish pathogen
Vibrio alcaligenesMTCC 4442Fish pathogen
Vibrio vulnificusMTCC 1145Fish pathogen
Thalassomonas sp.MMD12Marine biofilm bacterium
Alteromonas sp.MMD16Marine biofilm bacterium
Pseudoalteromonas sp.MMD18Marine biofilm bacterium
Pseudoalteromonas sp.MMD19Marine biofilm bacterium
Ruegeria sp.MMD27Marine biofilm bacterium

Evaluation of biofilm disruption potential by a microtiter plate assay

The biofilm strains were cultured overnight in Zobell marine broth (Himedia) and their microtiter plate adherence was determined as per Stepanovic et al. (2000). Cells were washed and resuspended in phosphate-buffered saline (pH 7.2) to a turbidity equivalent to a 0.5 M McFarland standard. Wells of sterile 96-well U-bottomed microtiter plates were filled with 80 μL of Zobell marine broth, 10 μL of each cell suspension and 10 μL each of the test concentration/controls, in triplicate. The control was set with a newly developed biofilm, negative control wells were filled with culture broth and 10 μg mL−1 CuSO4 was used as a positive control. All the triplicate wells were set with TLC-purified glycolipid of 5, 10 and 30 μg mL−1, respectively. The control and treated wells were set in a Latin-square arrangement and the experiment was repeated six times to validate the results statistically. Plates were placed on a platform shaker to obtain dynamic culture conditions. Planktonic cells and spent media were discarded, and adherent cells were gently rinsed twice with deionized water and allowed to air dry before being stained. The biofilms were stained by 200 μL of a 0.4% crystal violet solution (w/v) for 10 min, after which the dye was discarded and the wells were rinsed twice with deionized water. The wells were allowed to air dry before solubilization of the crystal violet with 200 μL of dimethyl sulfoxide. The OD was determined at 595 nm in an enzyme-linked immunosorbent assay reader (Labnics). The plate was observed under a phase-contrast microscope with × 40 magnification (Optica) to evaluate the disruption of biofilms.

The results of the plate assay were confirmed by the in vitro biofilm-forming assay. The biofilm was developed in a cover glass immersed in a 50-mL Erlenmeyer flask containing Zobell marine broth and incubated at 28 °C for 24 h. To study the biofilm disruption, the mature biofilm was exposed to a 0.5% glycolipid biosurfactant for 30 min at 28 °C. The exposed cover slip was stained with a 0.4% crystal violet solution (w/v) and observed under a phase-contrast microscope (Optica) at × 40 magnification.

Results and discussion

Glycolipid biosurfactant from MSA19

The heterotrophic bacteria (MSA01–MSA57) isolated from the marine sponge D. nigra were screened for biosurfactant activity. Among the five potential biosurfactant producers, based on the antimicrobial activity of the biosurfactants, the isolate MSA19 was chosen for the present study. Biosurfactant production of the isolate MSA19 was confirmed using conventional screening methods including hemolytic activity (8 mm), oil displacement (8 mm), lipase activity (9 mm), positive drop-collapsing test and emulsification activity (20%). The isolate MSA19 was Gram-positive, motile and showed unique spores in Ziehl Nielsen staining. The isolate was positive for the MR-VP test and hydrolyzed starch, gelatin, cellulose and tributyrin. The isolate utilized glucose, but not mannitol. The strain MSA19 was sensitive to ciprofloxacin and chloramphenicol and resistant to ampicillin. Based on the morphological, biochemical characteristics, the results of the RDPII seqmatch program ( and phylogenetic analysis (, the isolate MSA19 was identified as Brevibacterium casei MSA19. The sequence data of the strain MSA19 were deposited in GenBank with an accession number GQ153944.

The optimum production (65% of the emulsification index) was attained with pH 7, temperature 30 °C, 2% salinity and glucose as carbon source and peptone as nitrogen source, respectively. The effective factors optimized for the production of glycolipid biosurfactant by MSA19 are presented in Table 2. Glycolipids are typically synthesized during the late-exponential or the stationary phase of growth when the cell density is high, and this production is considerably enhanced by nutrient limitation (Guerra-santos et al., 1986). However, it has been shown previously that growth in a nitrogen-limited medium increases the production of the glycolipid biosurfactant (Desai & Banat, 1997). The active compound was extracted with three solvents including ethyl acetate, diethyl ether and dichloromethane. The ethyl acetate extracts showed the highest activity compared with the other solvents. The active compound was characterized by TLC, HPLC, FT-IR and GC-MS analyses. In TLC, a single discrete spot was obtained with an Rf value of 0.81 and it was found to be glycolipid. The GC-MS (Table 3) and FT-IR (Fig. 1) analyses revealed the compound produced by MSA19 to be a glycolipid with a hydrophobic nonpolar hydrocarbon chain (9,12,15-octadecatrienoic acid, 2-[(trimethylsilyl)oxy]-1-[[(trimethylsilyl)oxy]methyl]ethyl ester, (Z,Z,Z)) and the hydrophilic sugar part as 1H-cyclopenta[c]furan-3(3aH)-one, 6,6a-dihydro-1-(1,3-dioxolan-2-yl)-, (3aR,1-trans,6a-cis). The retention times, relative intensities (%) and electron impact mass spectrometry (EIMS) of the relevant peaks are as follows: Peak A: 17.13, 60, EIMS m/z (% rel. intensity), 108 (30), 95 (40), 79 (80), 67 (75), 55 (60), 41 (100), 29 (30) and Peak B: 15.38, 40, EIMS m/z (% rel. intensity), 69 (100), 40 (100), 29 (50). Based on the FT-IR and GC-MS data, the surface-active compound was characterized as a glycolipid. The compound showed 30% higher activity over the synthetic surfactants such as sodium dodecyl sulfate and Tween 80.

Table 2.   Factors optimized for the production of a biosurfactant by Brevibacterium casei MSA19 under submerged fermentation
pH5–9 (with increments of 1)7
Temperature10–50°C (with increments of 10°C)30°C
Salinity1–3% (with increments of 0.5%)2%
Carbon sourcesGlucose, olive oil, kerosene, vegetable oil (1%)Glucose
Nitrogen sourcesYeast extract, beef extract, (NH4)2NO3, Peptone (1%)Peptone
MetalsFeSO4, CuSO4, MnCl2, MgCl2 (1%)FeSO4
AminoacidsAsparagine, valine, leucine, glycine, glutamic acid (1%)Asparagine
Table 3.   Predicted compounds and major fractions of GC-MS data of Brevibacteriumcasei MSA19
Sl. no.Peak nameRetention timePeak area%Peak area
1Name: cyclopropyl carbinol
Formula: C4H8O
MW: 72
3.59567 7380.2203
2Name: hydroxyurea
Formula: CH4N2O2
MW: 76
10.0467 5400.0262
3Name: propanamide
Formula: C3H7NO
MW: 73
11.1586 9070.0337
4Name: 2-aminononadecane
Formula: C19H41N
MW: 283
14.59560 9450.2177
5Name: 1H-cyclopenta[c]furan-3(3aH)-one, 6,6a-dihydro-1-(1,3-dioxolan-2-yl)-, (3aR,1-trans,6a-cis)-
Formula: C10H12O4
MW: 196
15.38102 6230.0398
6Name: benzeneethanamine, 2,5-difluoro-á,3,4-trihydroxy-N-methyl-
Formula: C9H11F2NO3
MW: 219
15.9981 1440.0315
7Name: imidazole, 2-amino-5-[(2-carboxy)vinyl]-
Formula: C6H7N3O2
MW: 153
16.17115 7370.0449
8Name: 9,12,15-octadecatrienoic acid, 2-[(trimethylsilyl)oxy]-1-[[(trimethylsilyl)oxy]methyl]ethyl ester, (Z,Z,Z)-
Formula: C27H52O4Si2
MW: 496
17.1393 0220.0361
9Name: 1,2-benzenedicarboxylic acid, butyl octyl ester
Formula: C20H30O4
MW: 334
17.57467 9750.1816
10Name: propanoic acid, 3,3′-thiobis-, didodecyl ester
Formula: C30H58O4S
MW: 514
29.87255 571 29699.1682
Figure 1.

 FT-IR spectra of the surfactive fraction separated by column chromatography.

Antibiogram of MSA19

The antibiogram of MSA19 showed that the TLC-purified glycolipid was a broad-spectrum antimicrobial agent (Fig. 2). The MIC determined against biofilm bacteria such as Escherichia coli and Vibrio alginolyticus were 68 and 72 μg glycolipid mL−1, respectively, and the concomitant MBC values were 38 and 41 glycolipid mL−1, respectively. Based on the MIC/MBC ratio, the MSA19 glycolipid was determined to be bacteriostatic. Considering the bacteriostatic potential of the MSA19 glycolipid, the efficacy in biofilm-forming capacity was determined using the lower concentrations of 5, 15 and 30 μg glycolipid mL−1, respectively.

Figure 2.

 Antimicrobial potential of Brevibacterium casei MSA19. The activities are presented in the inhibition area of the halo produced by B. casei MSA19. The test was conducted using 100 μg TLC-purified glycolipid in 100 μL of phosphate buffer (pH 7.0).

Biofilm disruption potential of the MSA19 glycolipid

Based on their surface-active properties, we hypothesized that glycolipids could affect biofilm formation. To test the hypothesis, we examined the effects of the MSA19 glycolipid on the biofilm-forming capacity of biofilm infection-causing pathogenic bacteria. The MSA19 glycolipid potentially disrupted biofilm formation under dynamic conditions. The biofilm disruption potential of the MSA19 glycolipid was consistent against mixed biofilm bacteria (Fig. 3) and individual biofilm bacterium (Fig. 4). The biofilm-forming capacity of both mixed culture and individual strains was significantly inhibited at 30 μg glycolipid mL−1. In the cover slip assay, the biofilm disruption was evident and showed a disrupted biofilm under phase-contrast microscopic observation (Fig. 5). The microtiter plate assay also showed biofilm disruption on microtiter plates under Stereozoom microscopic observation. The present study showed that the MSA19 glycolipid could inhibit the beginning of the attachment mediated by flagella and pili (Shi & Sun, 2002). In a previous report, lauroyl glucose effectively disrupted both fungal and bacterial biofilms (Dusane et al., 2008). However, its MIC was >500 μg mL−1. Considering the lowest MIC/MBC values and biofilm disruption potential, the glycolipid biosurfactant would be an ideal lead compound for the development of a novel antibiofilm drug. This natural biosurfactant can be produced on a large scale under submerged fermentation for commercial applications.

Figure 3.

 Efficacy of MSA19 glycolipid in the biofilm-forming potential of mixed bacteria. The values are presented as the mean and SD of six trials. The inocula used for the mixed biofilm formation include Vibrio parahaemolyticus MTCC 451, Vibrio harveyi MTCC 3438, Vibrio alginolyticus MTCC 4439, Vibrio alcaligenes MTCC 4442, Vibrio vulnificus MTCC 1145, Pseudomonas aeruginosa MTCC 2453 and Escherichia coli MTCC 2339. The biofilm formation capacity was determined under dynamic conditions. Treated 1, 5 μg mL−1; Treated 2, 10 μg mL−1; Treated 3, 30 μg mL−1.

Figure 4.

 Efficacy of the MSA19 glycolipid in the biofilm-forming potential of biofilm bacteria. The values are presented as the mean and SD of six trials. The efficacy of the MSA19 glycolipid in the relative biofilm formation capacity of biofilm-forming pathogenic strains including Vibrio parahaemolyticus MTCC 451 (V.p), Vibrio harveyi MTCC 3438 (V.h), Vibrio alginolyticus MTCC 4439 (V.alg), Vibrio alcaligenes MTCC 4442 (V.alcal), Vibrio vulnificus MTCC 1145 (V.vul), Pseudomonas aeruginosa MTCC 2453 (P.a) and Escherichia coli MTCC 2339 (E.c) was determined after 24 h under dynamic conditions. Treated 1, 5 μg mL−1; Treated 2, 10 μg mL−1; Treated 3, 30 μg mL−1.

Figure 5.

 Biofilm disruption potential of the MSA19 glycolipid against biofilm-forming pathogenic bacteria under microscopic views. (a) Biofilm formation in a microtiter plate under Stereozoom microscopic view. (b) Biofilm formation in a glass coverslip under a phase-contrast microscopic view. (c) Disrupted biofilm in a microtiter plate under Stereozoom microscopic view. (d) Disrupted biofilm in a glass coverslip under a phase-contrast microscopic view.


B.S. is grateful to the Ministry of Earth Sciences, New Delhi, for the award of a Junior Research Fellowship in the MoES-funded research project. This paper is an outcome of the MoES-OASTC project. The authors are grateful to P.A. Thomas, Scientist Emeritus, Central Marine Fisheries Research Institute, Vizhinjam-692521, for the identification of marine sponge.