1
`
`1
`
`1
`
`1
`
`1
`
`1
`
`Biotechnology Techniques, Vol 11, No 6, June 1997, pp. 413–416
`
`Recovery and purification of the
`lipopeptide biosurfactant of
`Bacillus subtilis by ultrafiltration
`Sung-Chyr Lin* and Horng-Jyh Jiang
`Department of Chemical Engineering, National Chung Hsing University, Taichung, 402, Taiwan, R.O.C.
`
`Surfactin, a lipopeptide biosurfactant produced as micelles by Bacillus subtilis, was recovered from the fermentation
`broth by ultrafiltration with a 30 kDa MWCO membrane. The retained surfactin micelles were then ruptured and
`collected in the permeate by adding methanol to 50% (v/v). The final yield of surfactin was 95%.
`
`24 pts min base to base from Key words to line 1 of text
`
`Introduction
`The recovery and purification of biosurfactants from
`complex fermentation broth is a major problem in the
`commercialization of biosurfactants. The most widely
`used approaches for the recovery and purification of
`biosurfactants involve precipitation at extreme pHs and
`extraction with organic solvents. For example, the
`current practice for the recovery of surfactin, a lipopep-
`tide biosurfactant produced by Bacillus subtilis, involves
`precipitation at pH 2.0 followed by organic extraction,
`adsorption chromatography, or thin layer chromatog-
`raphy (Arima et al., 1968), which usually lead to the
`generation and release of hazardous wastes. Therefore,
`it is necessary to develop a more economic and envi-
`ronmentally-friendly approach for the recovery and
`purification of biosurfactants.
`
`At concentrations above the critical micelle concentra-
`tion (CMC), surfactant molecules associate to form
`supramolecular structures, such as micelles or vesicles,
`with nominal molecular diameters up to two to three
`orders of magnitude larger than that of the single unas-
`sociated molecules. It is, therefore, possible to retain
`surfactant molecules in the form of micelles in the
`retentate by ultrafiltration. This approach has been
`successfully employed for the recovery of surfactin, a
`lipopeptide surfactant produced by Bacillus subtilis, from
`complex fermentation medium (Mulligan and Gibbs,
`1990). In this study, a modified ultrafiltration process
`for the recovery and purification of surfactin from
`fermentation medium will be discussed.
`
`Materials and methods
`Microorganism and growth conditions
`Bacillus subtilis ATCC 21332 was grown in 2 liters
`of mineral salt medium containing 4% (w/v) glucose
`
`(Cooper et al., 1981) at 30ºC in a 3-liter fermenter for
`48 h. Cells were removed from the culture by centrifu-
`gation at 12,000 g for 10 minutes.
`
`Ultrafiltration
`Cell-free culture was concentrated by an Amicon
`magnetically stirred ultrafiltration cell (Beverly, MA,
`USA) with cellulose membranes of varying molecular
`weight cut-offs (MWCO) at pressure in the range of 6.9
`· 104 and 2.1 · 105 Pa. Hollow fiber ultrafiltration
`cartridges (9 · 337 mm, A/G Technology, Needham,
`MA, USA) were used for continuous operation at a pres-
`sure of 1.7 · 105 Pa.
`
`HPLC
`The concentration of surfactin in the cell-free culture
`was determined by HPLC with with a Techsphere 5
`␮m ODS C18 reverse phase column. For each assay 100
`␮l cell-free culture was injected and eluted with a linear
`gradient of 70–79% methanol in 10 mM KH2PO4
`buffer at pH 6.0 at 0.5 ml/min within 60 min. The
`absorbancy of the eluent was monitored at 210 nm.
`Surfactin purchased from Sigma was used as standard.
`
`Results and discussion
`Identification of surfactin
`HPLC of the cell-free B. subtilis culture and the
`permeate through a ultrafiltration membrane with a
`MWCO of 3,000 daltons are shown in Fig. 1A and 1B,
`respectively. Compared with the chromatogram of
`surfactin standard from Sigma (data not shown), the
`peaks eluted between 31 and 45 minutes in Fig. 1A
`were identified as those of surfactin. The presence of
`multiple peaks on the chromatograms for standard
`surfactin and fermentation broth resulted from the
`existence of several surfactin structures produced by
`
`© 1997 Chapman & Hall
`
`Biotechnology Techniques · Vol 11 · No 6 · 1997
`
`413
`
`
`1 of 4
`
`FRESENIUS-KABI, Exh. 1014
`
`

`
`1
`
`1
`
`1
`
`1
`
`1
`
`1
`
`S.-C. Lin and H.-J. Jiang
`
`B. subtilis. Like most secondary metabolites, surfactin
`consists of a family of lipopeptides with similar chem-
`ical structures. So far at least nine different surfactin
`structures have been identified (Horowitz and Griffin,
`1991; Peypoux and Michel, 1992; Lin et al., 1994; Wu
`and Lin, 1996). The absence of peaks eluted during
`the same period in Fig. 1B further confirmed that
`those peaks represented surfactin, because only a small
`number of unassociated surfactin molecules could pene-
`trate the membrane with a MWCO of 3,000 daltons.
`The surfactin concentration in the cell-free broth was
`approximately 250 mg/l by HPLC analysis.
`
`Recovery of surfactin
`Surfactin was concentrated from 120 ml to 6 ml by
`ultrafiltration with membranes of MWCOs ranging
`from 1,000 to 100,000 daltons. In ultrafiltration sys-
`tems with high MWCO membranes, small molecules
`in the fermentation medium, such as salts, amino acids,
`organic acids and alcohols, and other small metabo-
`lites are allowed to pass the membrane freely and are
`thus released into the permeate, while macromolecules,
`such as extracellular proteins, with nominal molecular
`diameters higher than MWCO of the membarne are
`
`concentrated in the retentate. At concentrations above
`its CMC, the excess surfactin molecules associated into
`micelles leaving only a small number of surfactin mole-
`cules, at concentration close to its CMC, unassociated.
`While the unassociated surfactin molecules can easily
`penetrate ultrafiltration membrane and be collected
`in the permeate, surfactin micelles are restricted from
`permeation and are concentrated in the retentate. The
`concentrations of surfactin in the retentate and the
`permeate were determined by HPLC. The retention
`factors of surfactin by ultrafiltration, defined as
`[surfactin
`in the retentate]/[total surfactin], with
`different MWCO membranes is shown in Fig. 2.
`Percentages of surfactin retained in the form of associ-
`ated supramolecules with 10,000 and 30,000 daltons
`MWCO membranes were about 98.8 and 97.9%,
`respectively. The percentages of surfactin retained
`decreased significantly to 86 and 53% when 50,000 and
`100,000 daltons MWCO membranes were used.
`
`The results indicated that a majority of the surfactin
`micelles behaved as macromolecules with molecular
`weights in the range of 30,000 to 50,000 daltons, which
`could be effectively retained with membranes with a
`
`Figure 1 HPLC of cell-free B. subtilis fermentation broth (A) and permeate through a 3,000 MWCO ultrafiltration mem-
`brane (B).
`
`414
`
`Biotechnology Techniques · Vol 11 · No 6 · 1997
`
`
`2 of 4
`
`

`
`1
`
`1
`
`1
`
`1
`
`1
`
`1
`
`Recovery and purification of the lipopeptide biosurfactant of Bacillus subtilis by ultrafiltration
`
`MWCO of up to at least 30,000 daltons. Although the
`loss of surfactin was minimal with 3,000 daltons
`MWCO membrane, the filtration rate was relatively
`low. Taking filtration rate and retention factor into
`consideration, ultrafiltration membranes with MWCO
`of 10,000 and 30,000 daltons were used for further
`investigation.
`
`increase in surfactin concentration in the permeate. At
`a concentration of 50%, methanol was effective in
`breaking down almost all micelles, i.e. the complete
`leakage of surfactin molecules into the permeate. High
`molecular weight components were retained in the
`retentate because their structures were not significantly
`affected by the presence of methanol.
`
`Dissociation of surfactin micelles
`The employment of ultrafiltration for the recovery of
`surfactin from complex fermentation broth has been
`previously reported
`(Mulligan and Gibbs, 1990).
`However, in addition to surfactin micelles, other macro-
`molecules such as extracellular proteins and polysac-
`charides can also be simultaneously concentrated by
`ultrafiltration. It is, therefore, desirable to develop a
`process that can easily separate surfactin from these
`contaminating macromolecules.
`
`This can be accomplished by dissociating surfactin
`micelles into unassociated molecules, which can then be
`easily forced through ultrafiltration membranes. It is
`well known that organic solvents such as alcohol and
`acetone can destabilize surfactant micelles. The effect of
`methanol concentration on the stability of surfactin
`micelles was studied, Fig. 3. The concentration of
`surfactin in the permeate in the form of unassociated
`molecules was again monitored by HPLC. The
`percentage of micelles ruptured was defined as [surfactin
`in the permeate]/[initial surfactin in the retentate]. The
`increase
`in methanol
`concentration
`resulted
`in
`the decrease in micelle stability, indicated by the
`
`The surfactin preparation collected in the permeate
`was further concentrated by drying under vacuum. TLC
`analysis on silica gel 60 in the solvent chloroform:
`methanol: water (65:25:4; by vol.) indicated that the
`purity of the surfactin preparation obtained by the
`proposed process was at least as high as that of the stan-
`dard surfactin obtained by conventional process. The
`final yield of surfactin was 95%.
`
`Conclusions
`The recovery and purification of biosurfactants from the
`complex fermentation broth have been one of the major
`hurdles to the successful commercialization of biosur-
`factants. The current practices for biosurfactant recovery
`and purification usually leads to the generation of large
`volume of hazardous wastes. It is, therefore, necessary
`to develop alternative approaches
`
`Surfactant micelles can be effectively concentrated by
`ultrafiltration. A modified ultrafiltraiion process, char-
`acterized by the subsequent elution of surfactin from
`the filtration system by the addition of methanol
`
`Figure 2 Retention factors of surfactin by ultrafiltration with
`different MWCO membranes.
`
`Figure 3 The effect of methanol concentration on the
`stability of surfactin micelles, indicated by the percentage of
`surfactin recovery in the permeate with ultrafiltration mem-
`branes with MWCO of 10,000 (䊊) and 30,000 (䊉) daltons.
`
`Biotechnology Techniques · Vol 11 · No 6 · 1997
`
`415
`
`
`3 of 4
`
`

`
`S.-C. Lin and H.-J. Jiang
`
`following preliminary concentration, allows the recovery
`and purification of surfactin from fermentation broth in
`a single ultrafiltration unit. Furthermore, arrays of
`hollow-fiber cartridges can be developed for continuous
`recovery and purification of surfactin from fermentation
`broth.
`
`Compared to conventional concentration processes,
`ultrafiltration processes have the advantages of minimal
`use and release of hazardous compounds and ease of
`scaling up. Furthermore, the elution of surfactin from
`the filtration system can not only effectively separate
`surfactin from other contaminating macromolecules but
`also minimize surfactin loss due to adhesion to ultrafil-
`tration membranes, which can be very significant when
`more hydrophobic membranes are employed. This
`process can be further modified and employed for the
`recovery and purification of most surfactants from
`aqueous solutions at concentrations above the critical
`micelle concentration. Operational systems involving
`the employment of arrays of hollow fiber ultrafiltration
`
`cartridges can also be developed for continuous opera-
`tion.
`
`Acknowledgement
`This work is supported by a grant NSC 86–2214–
`E–005–005 from the National Science Council, Taiwan,
`R.O.C.
`
`References
`Arima, K., Kakinuma, A. and Tamura, A. (1968). Biochem.
`Biophys. Res. Commun. 31, 488–494.
`Cooper, D.G., MacDonald, C.R., Duff, S.J.B. and Kosaric, N.
`(1981). Appl. Environ. Microbiol. 42, 408–412.
`Horowitz, S. and Griffin, W.M. (1991). J. Ind. Microbiol. 7,
`45–52.
`Lin, S.C., Minton, M.A., Sharma, M.M. and Georgiou, G. (1994).
`Appl. Environ. Microbiol. 60, 31–38.
`Mulligan, C.N. and Gibbs, B.F. (1990). J. Chem. Tech. Biotechnol.
`47, 23–29.
`Peypoux, P. and Michel, G. (1992). Appl. Microbial. Biotechnol.
`36, 515–517.
`Wu, C.H. and Lin, S.C. (1996). J. Chin. Colloid Interface Soc. 19,
`125–136.
`
`Received 5 March 1997;
`Revisions requested 27 March 1997;
`Revisions received 14 April 1997;
`Accepted 14 April 1997
`
`1
`
`1
`
`1
`
`1
`
`1
`
`1
`
`416
`
`Biotechnology Techniques · Vol 11 · No 6 · 1997
`
`
`4 of 4