Appl Microbiol Biotechnol (1989) 31:486-489
`
`Enhanced biosurfactant production
`by a mutant Bacillus subtilis strain*
`
`Applied
`and MkroMotogy
`Biotechnology
`
`© Springer-Verlag 1989
`
`Catherine N. Mulligan1, Terry Y.-K. Chow2, and Bernard F. Gibbs3
`1 Biochemical Engineering Section, National Research Council of Canada, Biotechnology Research Institute, 6100 Royalmount
`Avenue, Montreal, Quebec H4P 2R2, Canada
`2 Genetic Engineering Section, National Research Council of Canada, Biotechnology Research Institute, 6100 Royalmount
`Avenue, Montreal, Quebec H4P 2R2, Canada
`3 Protein Engineering Section, National Research Council of Canada, Biotechnology Research Institute, 6100 Royalmount
`Avenue, Montreal, Quebec H4P 2R2, Canada
`
`Summary. Ultraviolet mutation of Bacillus subtilis
`ATCC 21332 yielded a stable mutant that pro­
`duced over three times more of the biosurfactant,
`surfactin, than the parent strain. By protoplast
`fusing the mutant (Suf-1) with the marker strain,
`B. subtilis BGSC strain IA28, the mutation was lo­
`cated between argCA and hisAi on the genetic
`map.
`
`Introduction
`
`Biosurfactants are produced as metabolic prod­
`ucts or membrane components. These compounds
`are lipopeptides, glycolipids, lipopolysaccharides,
`neutral lipids and fatty acids or phospholipids
`(Cooper 1986; Cooper and Zajic 1980; Margaritis
`et al. 1979; Rosenberg 1982; Zajic and Steffens
`1984). Surfactants are important as they are used
`in many multiphase processes (Layman 1985).
`Biosurfactants are biodegradable and potentially
`less toxic than the synthetic compounds currently
`used (Cooper and Zajic 1980). They can also be
`produced from a variety of substrates.
`Surfactin, produced by Bacillus subtilis ATCC
`21332 is one of the most effective biosurfactants
`known. It reduces the surface tension of water
`from 72 to 27 mN/m at a concentration as low as
`0.005% (Arima et al. 1968). This lipopeptide con­
`tains seven amino acids bonded to the carboxyl
`and hydroxyl groups of the 14-carbon acid (Ka-
`kinuma et al. 1969). It inhibits clot formation,
`lyses erythrocytes (Arima et al. 1968; Bernheimer
`and Avigad 1970), lyses bacterial spheroplasts
`and protoplasts and inhibits cyclic 3',5'-mono-
`
`* NRCC 30549
`Offprint requests to: B. F. Gibbs
`
`phosphate diesterase (Hosono and Suzuki 1983).
`It is produced from glucose instead of hydrocar­
`bons as is the case of most biosurfactants (Cooper
`et al. 1979).
`To increase commercial interest, yields of sur­
`factin must be improved. Until now the only
`methods which have been utilized to enhance pro­
`duction are strain selection of the manipulation of
`environmental or nutritional factors (Cooper et al.
`1981; Guerra-Santos et al. 1986). These methods,
`however, have their limitations as the levels of in­
`crease are marginal. Other approaches are there­
`fore necessary. Our research has concentrated on
`the isolation of mutants which overproduce sur­
`factin.
`
`Materials and methods
`
`Ultraviolet mutagenesis. The B. subtilis prototroph strain
`ATCC 21332 was grown to logarithmic phase and then ap­
`proximately 3000 cells were plated on nutrient agar plates. The
`cells were then UV radiated for 35 s with short wave in a Chro-
`mato-Vue Cabinet Model CC-60 (UVP, San Gabriel, Calif,
`USA). This dosage of UV light had been previously deter­
`mined to give 10%-20% survival of the colonies. The UV-irra-
`diated cells were incubated at 37° C in the dark until colonies
`were visible.
`
`Screening method for enhanced biosurfactant production. The B.
`subtilis mutants from UV mutagenesis were replica plated or
`individually spotted on to rich nutrient agar medium plates
`containing 5% sheep blood cells, 4% glucose, 0.1% nutrient
`broth, 0.1% yeast extract and mineral salts (Cooper et al.
`1981). The cells were incubated at 37° C and the haemolytic
`zone surrounding the colonies was scored visually. The degree
`of lysis of red blood cells has been shown to be related to the
`level of surfactin production by B. subtilis (Mulligan et al.
`1984).
`
`Surfactin production in liquid media. From a sheep blood agar
`plate, B. subtilis was inoculated into 4% glucose and mineral
`salts medium, supplemented with 3.2 x 10~4 M FeS04 (100/
`500 ml flasks) (Cooper et al. 1981). After 3 days growth, 10 ml
`
`PETITIONERS
`
`EXHIBIT NO. 1028 Page 1 of 4
`
`

`
`C. N. Mulligan et al.: Enhanced surfactin production by a B. subtilis mutant
`
`487
`
`culture was transferred to another flask. After 6 h growth,
`100 ml of this medium was used as an inoculum for a 3.7-1
`CHEMAP fermentor (Volketswil, Switzerland).
`The continuously stirred tank fermentor was operated un­
`der the following conditions: a 2.0-1 working volume, a tem­
`perature of 37° C, a 5.0 l/min aeration rate and pH control at
`6.7. The surfactin was removed continuously in the foam into
`a flask on the air exhaust line (Cooper et al. 1981).
`Growth was monitored by measuring the optical density
`(OD) of the fermentor broth at 600 nm; samples above OD 1.0
`AUFS were diluted appropriately. For both strains, on OD of
`1.0 AUFS represents 0.21 g/1 biomass. Insignificant amounts
`of cells were found in the foam. Surface tension of the broth
`was measured by a Fisher Surface Tensiomat Model 21 (Mon­
`treal, Canada), a du Nouy tensiometer (Cooper et al. 1979).
`
`Chemical isolation of surfactin. Using concentrated hydro­
`chloric acid, the pH of the collapsed foam after cell removal
`was adjusted to 2±0.5 (Cooper et al. 1981). This precipitated
`the proteins and lipids. After decanting the supernatant, the
`residue was resuspended in dichloromethane in a separatory
`funnel and shaken vigorously. Surfactin was recovered in the
`organic (top) layer. Two more extractions were performed and
`the organic layers were pooled and evaporated. The residue
`was redissolved in slightly basic water (pH 8.0) and filtered
`through Whatman no. 1 paper to remove undissolved impuri­
`ties. The filtrate was again adjusted to pH 2 ±0.5 and ex­
`tracted with dichloromethane three times and evaporated as
`described above.
`
`Surfactin assay. The biosurfactant yield of the collapsed foam
`was determined by both critical micelle concentration (CMC)
`and amino acid analysis. The CMC values were determined by
`measuring the surface tension at various dilutions (Cooper et
`al. 1979). The logarithm of the dilution was plotted as a func-
`
`tion of surface tension. The CMC is the point of abrupt sur­
`face tension increase. The surfactant concentration is thus a
`function of the CMC.
`An aliquot of the collapsed foam was trichloracetic acid-
`treated to precipitate peptides and proteins. The supernatant
`was assayed to determine the free amino acids in the medium.
`The residue (pellet) was redissolved in 25% trifluoroacetic acid
`and an aliquot was removed, dried and acid hydrolysed for
`2.5 h at at 150oC in a Waters PICO-TAG Amino Acid Analy­
`sis System (Mass, USA). This enabled us to determine the am­
`ino acids in the peptide portion of surfactin. According to the
`molecular formula (Kakinuma et al. 1969), the peptidyl por­
`tion of surfactin is 76.5% by weight of the intact molecule. All
`amino acid analyses were done on a Beckman System 6300
`High Performance Analyser equipped with a Beckman Model
`7000 Data Station (Palo Alto, Calif, USA).
`
`Mass spectrometry. Mass spectra were obtained in the positive
`ion mode on a VG Analytical ZAB-HS double focussing mass
`spectrometer (Manchester, UK). The accelerating voltage was
`10 kV and the fast xenon atom beam was operated with an
`emission current of 1 mA at 8 kV. Mass spectra were recorded
`with an integrated data acquisition sytem and calibration was
`performed with Csl. Spectra for samples are an average of ten
`scans.
`
`Media for protoplast fusion. For the isolation of auxotrophs,
`transformants or transductants, Spizizen minimal medium
`(SMM) (Spizizen 1958) was used. For protoplast regeneration,
`HCP-1.5 medium containing glucose (5 g/1), casamino acids
`(5 g/1), K2HPO4 (3.5 g/1), KH2PO4 (1.5 g/1), L-tryptophan
`(0.1 g/1), polyvinylpyrrolidone (PVP) (mean molecular weight
`700000, 15 g/1), MgCla (1.9 g/1) and 25% (v/v) 2M sodium
`succinate (pH 7.3); HCP-1.5 agar contained the same compo­
`nents supplemented with agar (8 g/1). The HCP-3 medium was
`the same as HCP-1.5 with the exception of 30 g/1 PVP.
`
`.4
`
`"8
`
`Fig. 1. Haemolysis by mutant and other strains on blood agar. Parent strain B. subtilis ATCC 21332 (A), UV radiated colonies (B
`and D) and the mutant Suf-1 (C). Growth was for 2 days at 37° C
`
`PETITIONERS
`
`EXHIBIT NO. 1028 Page 2 of 4
`
`

`
`488
`
`C. N. Mulligan et al.: Enhanced surfactin production by a B. subtilis mutant
`
`Protoplast fusion. Protoplast fusion between the enhanced pro­
`duction mutant and BGSC strain 1A28 (argCA, AisAl and
`trpC2) and regeneration was according to Akamatsu and Se-
`kiguchi (1987). Cells were placed in SMM medium with
`250 ug/ml lysozyme and incubated at 42° C for 45 min. The
`protoplasts were suspended in HCP-3 medium (30° C) for 3 h
`without shaking. The protoplast suspensions from the two
`strains were mixed and then centrifuged (4000 g, 10 min). The
`pellet was vortexed in SMM medium and 40% polyethylene
`glycol 4000 (average molecular weight 3000) was then added.
`The protoplast suspension was left to stand for one min fol­
`lowed by dilution with HCP-1.5 medium. The protoplasts were
`regenerated by placing 0.1 ml suspension on the surface of
`HCP-1.5 agar medium. Colonies were streaked on nutrient
`agar and single colonies were analysed for unselected mark-
`ers.
`
`Results and discussion
`
`Selection of a mutant
`
`To enhance biosurfactant production, mutation
`was chosen since any change in the regulatory
`system of biosurfactant synthesis and secretion
`could result in an altered level of production.
`After ultraviolet mutagenesis, approximately 1000
`colonies were examined for enhanced haemolytic
`activity on sheep blood agar plates. One mutant,
`Suf-1, (C) produced a significantly larger haemo­
`lytic zone than the parent strain (Fig. 1). This mu­
`tant was not an auxotroph as it was able to grow
`in minimal medium. Without radiation, no colo­
`nies out of 1000 produced significantly higher lev­
`els of surfactant.
`
`Biosurfactant production
`
`To compare biosurfactant production by the mu­
`tant and parent strains, both were grown similarly
`
`Table 1. Enhanced production of biosurfactant by Bacillus
`subtilis Suf-1
`
`Strain
`
`Growth after 40 h
`(optical density
`at 600 nm)
`
`ATCC 21332
`Suf-1
`
`8.2
`8.3
`
`Surfactant
`yield (mg)a
`
`328
`1124
`
`a Surfactant collected in the foam was determined by amino
`acid analysis and represents the amount produced in the fer-
`mentor
`
`in a fermentor. Suf-1 produced 3.4 times more
`biosurfactant than the parent strain over the same
`time period into the foam (Table 1). In both cases
`surfactin was totally (> 99%) transferred from the
`fermentor medium into the foam. The growth of
`the two strains was approximately equal. Several
`single isolates of the same colony Suf-1 after
`many generations produced the same enhanced
`levels of surfactant in liquid medium. The Suf-1
`mutation, therefore, was stable and did not revert
`in subculturing.
`
`Biosurfactant characterization
`
`The purified surfactant from the mutant and wild-
`type had similar amino acid profiles, which
`agreed with the compositon of surfactin (Kakin-
`uma et al. 1969). Impure surfactant produced dif­
`ferent peptidyl profiles. Furthermore, confirma­
`tion of the structure of the biosurfactants was ob­
`tained by FAB-MS (fast atom bombardment mass
`spectrometry). Based on the surfactin molecular
`formula (C53H93N7O13), the protonated molecular
`weight is 1036.69. The spectra of the compounds
`
`100 -1
`
`1023
`
`1036
`
`1061
`
`1047
`
`1009
`
`80 -
`
`60 -
`
`40 -
`
`20 -
`
`1037
`
`1008
`
`1023
`
`1062
`
`1046
`
`' b
`
`1077
`
`a
`
`1078
`
`1090
`
`1055
`
`1091
`
`1
`
`0
`1000
`
`1020
`
`1040
`
`1080
`
`1100 1000
`
`1020
`
`1060
`
`1080
`
`1100
`
`1060
`1040
`Mass
`Mass
`Fig. 2. Mass spectra of the purified biosurfactants isolated from the parent organism (A) and the hyperproductive mutant (B)
`
`PETITIONERS
`
`EXHIBIT NO. 1028 Page 3 of 4
`
`

`
`C. N. Mulligan et al.: Enhanced surfactin production by a B. subtilis mutant
`
`489
`
`produced by the parent strain (Fig. 2A) and the
`mutant strain (Fig. 2B) were quite similar with
`molecular ions at 1036 and 1037, respectively.
`Their fragmentation pattern was also in harmony,
`with minor differences in intensities.
`The screening method (Mulligan et al. 1984)
`for enhanced surfactin production was confirmed
`as these isolated compounds, solubilized
`in
`slightly basic water (pH 8.0), promoted haemoly­
`sis of the blood agar plates.
`
`the wild-type. The mutation for increased produc­
`tion was mapped between argCA and hisA\ on the
`B. subtilis chromosome, as determined by proto­
`plast fusion. This information provides a target
`site for future genetic manipulation.
`Acknowledgements. The authors would like to thank Dr. Orval
`A. Mamer of the McGill University Biomedical Mass Spec­
`trometry Unit (Montreal, Canada) and Dr. Marie-Renee de
`Roubin of the Biotechnology Research Institute for their ex­
`pert technical assistance.
`
`Location of the suf-1 mutation
`Protoplast fusion was performed to locate the
`area on the chromosome of the mutation respon­
`sible for enhanced surfactin production. The B.
`subtilis BGSC strain IA28 was chosen for the
`three markers (argCA, hisM and trpCl), evenly
`spaced on the genetic map (Piggot and Hoch
`1985). The marker strain produced very small
`amounts of biosurfactant on blood agar, no dif­
`ferent from strain ATCC 21332. The linkage of
`the Suf-1 phenotype and the marker gene should
`give the approximate location of the suf-l muta­
`tion with respect to the markers on the B. subtilis
`chromosome.
`Loss of enhanced surfactant production by
`Suf-1 during fusion occurred upon specific re­
`combinations of the mutated and marked areas
`(Table 2). This evidence suggested that the suf-\
`mutation corresponded with the area between
`argCA and hisM and furthest from trpCl. A sin­
`gle mutation in a specific region rather than mul­
`tiple mutations throughout the genome must be
`responsible for increased surfactin production.
`In summary, we increased biosurfactant pro­
`duction greater than threefold with B. subtilis Suf-
`1 by ultraviolet mutagenesis. Surfactin produced
`by this strain was identical to that produced by
`
`Table 2. Distribution of unselected marker classes among re­
`combinants obtained by protoplast fusion between B. subtilis
`IA28 (ArgCA HisM TrpCl) and Suf-1
`
`Number of colonies
`carrying indicated
`unselected markers
`
`Trp
`
`+
`+
`+
`
`His
`
`+
`+
`
`Arg
`
`+
`
`Numbers
`
`% Frequency
`of Suf-1
`
`153
`88
`32
`
`80 (121/153)"
`49 (43/88)
`0 (0/32)
`
`a In parentheses are the numbers of Suf-1 recombinants per
`number of recombinants examined in each class. Presence of
`Suf-1 was indicated by haemolysis of blood agar plates
`
`References
`
`Akamatsu T, Sekiguchi J (1987) Genetic mapping by means of
`protoplast fusion in Bacillus subtilis. Mol Gen Genet
`208:254-262
`Arima K, Kakinuma A, Tamura G (1968) A crystalline peptide
`lipid surfactant produced by Bacillus subtilis: isolation,
`characterization and its inhibition of fibrin clot formation.
`Biochem Biophys Res Commun 31:488-494
`Bernheimer AW, Avigad LS (1970) Nature and properties of a
`cytolic agent produced by Bacillus subtilis. J Gen Microbiol
`61:361-369
`Cooper DG (1986) Biosurfactants. Mirobiol Sci 3:145-149
`Cooper DG, Zajic JE (1980) Surface active compounds from
`microorganisms. Adv Appl Microbiol 26:229-253
`Cooper DG, Zajic JE, Gerson DF (1979) Production of sur­
`face active lipids by Corynebacterium lepus. Appl Environ
`Microbiol 37:4-10
`Cooper DG, MacDonald CR, Duff SIB, Kosaric N (1981) En­
`hanced production of surfactin from Bacillus subtilis by
`continuous product removal and metal cation additions.
`Appl Environ Microbiol 42:408-412
`Guerra-Santos LH, Kappeli O, Fiechter A (1986) Dependence
`of Pseudomonas aeruginosa continuous culture biosurfac­
`tant production on nutritional and environmental factors.
`Appl Microbiol Biotechnol 24:443-448
`Hosono K, Suzuki H (1983) Acylpeptides, the inhibitors of
`cyclic adenosine 3'-5' monophosphate phosphodiesterase.
`III. Inhibitors of cyclic AMP phosphodiesterase. J Antibiot
`36:679-683
`Kakinuma A, Oachida A, Shima T, Sugino H, Isono M, Ta­
`mura G, Arima K (1969) Confirmation of the structure of
`surfactin by mass spectroscopy. Agric Biol Chem 33:1669-
`1671
`Layman PL (1985) Industrial surfactants set for strong growth.
`Chem Eng News 63:23-48
`Margaritis A, Zajic JE, Gerson DF (1979) Production and sur­
`face properties of microbial surfactants. Biotechnol Bioeng
`21:1151-1161
`Mulligan CN, Cooper DG, Neufeld RJ (1984) Selection of mi­
`crobes producing biosurfactants on media without hydro­
`carbons. J Ferment Technol 62:311-314
`Piggot PJ, Hoch JA (1985) Revised genetic linkage map of Ba­
`cillus subtilis. Microbiol Rev 49:158-179
`Rosenberg E (1982) Microbial surfactants. CRC Crit Rev Bio-
`technol 1:109-32
`Spizizen J (1958) Transformation of biochemically deficient
`strains of Bacillus subtilis by deoxyribonucleate. Proc Natl
`Acad Sci USA 44:395-405
`Zajic JE, Steffens W (1984) Biosurfactants. CRC Crit Rev Bio­
`technol 1:87-107
`
`Received 15 December 1988/Accepted 9 May 1989
`
`PETITIONERS
`
`EXHIBIT NO. 1028 Page 4 of 4