APPLIED AND ENVIRONMENTAL MICROBIOLOGY,
`0099-2240/97/$04.00⫹0
`Copyright 䉷 1997, American Society for Microbiology
`
`Jan. 1997, p. 44–49
`
`Vol. 63, No. 1
`
`Antimycoplasma Properties and Application in Cell Culture of
`Surfactin, a Lipopeptide Antibiotic from Bacillus subtilis
`DIRK VOLLENBROICH,1 GEORG PAULI,2 MUHSIN O¨ ZEL,2 AND JOACHIM VATER1*
`Max-Volmer-Institut fu¨r Biophysikalische Chemie und Biochemie, Fachgebiet Biochemie und Molekulare
`Biologie, Technische Universita¨t Berlin, 10587 Berlin,1 and Robert Koch-Institut,
`Fachbereich Virologie, 13353 Berlin,2 Germany
`
`Received 5 July 1996/Accepted 10 September 1996
`
`Surfactin, a cyclic lipopeptide antibiotic and biosurfactant produced by Bacillus subtilis, is well-known for its
`interactions with artificial and biomembrane systems (e.g., bacterial protoplasts or enveloped viruses). To
`assess the applicability of this antiviral and antibacterial drug, we determined the cytotoxicity of surfactin with
`a 50% cytotoxic concentration of 30 to 64 ␮M for a variety of human and animal cell lines in vitro. Concom-
`itantly, we observed an improvement in proliferation rates and changes in the morphology of mycoplasma-
`contaminated mammalian cells after treatment with this drug. A single treatment over one passage led to
`complete removal of viable Mycoplasma hyorhinis cells from various adherent cell lines, and Mycoplasma orale
`was removed from nonadherent human T-lymphoid cell lines by double treatment. This effect was monitored
`by a DNA fluorescence test, an enzyme-linked immunosorbent assay, and two different PCR methods. Disin-
`tegration of the mycoplasma membranes as observed by electron microscopy indicated the mode of action of
`surfactin. Disintegration is obviously due to a physicochemical interaction of the membrane-active surfactant
`with the outer part of the lipid membrane bilayer, which causes permeability changes and at higher concen-
`trations leads finally to disintegration of the mycoplasma membrane system by a detergent effect. The low
`cytotoxicity of surfactin for mammalian cells permits specific inactivation of mycoplasmas without significant
`deleterious effects on cell metabolism and the proliferation rate in cell culture. These results were used to
`develop a fast and simple method for complete and permanent inactivation of mycoplasmas in mammalian
`monolayer and suspension cell cultures.
`
`Mycoplasmas are causative agents of serious diseases of
`humans and animals, such as acute respiratory inflammation
`(including pneumonia) and diseases of the urogenital tract,
`and seem to be cofactors in the pathogenesis of AIDS (4, 30).
`These smallest free-living organisms are parasites of eukary-
`otic cells and are one of the major contaminants that affect
`tissue culture cells. The most prevalent agents that do this are
`the mollicute species Mycoplasma orale (a human species),
`Mycoplasma hyorhinis (a porcine species), Mycoplasma arginini
`(a bovine species), and Acholeplasma laidlawii (a bovine spe-
`cies) (16). Contaminating mycoplasmas affect a variety of cel-
`lular processes and cell morphology, deplete the nutrients in
`the growth medium, and interfere with virus replication (5, 24).
`For both biological and ecological reasons, it is important to
`eliminate these agents from cell cultures used for basic re-
`search, diagnosis, and biotechnological production. The most
`effective procedure for eliminating, inactivating, or suppressing
`mycoplasmas in cell cultures is treatment with antibiotics (6,
`22, 27). In general, antibiotic therapies do not result in long-
`lasting successful decontamination, and undesirable side ef-
`fects on eukaryotic cells due to cytotoxic effects and the devel-
`opment of resistant mycoplasma strains have been observed
`(22, 27). Mycoplasmas lack a cell wall but are encircled by a
`three-layer cytoplasmic membrane, so that antibiotics such as
`the penicillins, which are common additives in cell culture
`media and interfere with murein formation in cell walls, are
`not effective against them.
`
`Screening for new antimycoplasma agents with novel modes
`of action is necessary. The secondary metabolites of various
`bacteria, fungi, and yeasts are some of the most profitable
`sources of new antibiotics. The soil bacterium Bacillus subtilis
`produces an abundance of substances with antibiotic proper-
`ties and with diverse structures (31). One of these, surfactin, is
`a cyclic lipopeptide antibiotic with a molecular weight of 1,036.
`It contains a mixture of several ␤-hydroxy fatty acids with chain
`lengths of 13 to 15 carbon atoms as its lipid portion. The main
`component is 3-hydroxy-13-methylmyristic acid, which forms a
`lactone ring system with an anionic heptapeptide (Fig. 1). As a
`consequence of this amphiphilic structure, surfactin is a pow-
`erful biosurfactant with high surface activity (1, 12, 14) and
`various interesting biological properties. It exhibits antifungal
`properties, moderate antibacterial properties (3, 26), and he-
`molysis; inhibits fibrin clot formation (1, 3); induces the for-
`mation of ion channels in lipid bilayer membranes (23); inhib-
`its enzymes such as cyclic AMP phosphodiesterase (10);
`exhibits antiviral and antitumor activities (13, 29); and inhibits
`starfish oocyte maturation (25).
`In this paper we show that improvements in the proliferation
`rates of mycoplasma-contaminated mammalian cells after
`treatment with surfactin were due to the antimycoplasma ac-
`tivity of this drug. We investigated the mode of surfactin action
`and developed an efficient method for eliminating mycoplas-
`mas from adherent and nonadherent mammalian cells.
`
`* Corresponding author. Mailing address: Max-Volmer-Institut fu¨r
`Biophysikalische Chemie und Biochemie, Fachgebiet Biochemie und
`Molekulare Biologie, Technische Universita¨t Berlin, Franklinstraße
`29, 10587 Berlin, Germany. Phone: 49 - 30 - 31425609. Fax: 49 - 30 -
`31424783. E-mail: jovajceh@mailszrz.zrz.tu-berlin.de.
`
`Source of surfactin. Surfactin was purchased from Biomol (Hamburg, Ger-
`many) and Sigma (Deisenhofen, Germany) or was purified from culture super-
`natants of B. subtilis OKB105 by acid precipitation, extraction with methanol,
`charcoal treatment, and gel filtration with Pharmacia Sephadex LH-20 as de-
`scribed previously (2). For the experiments described below, a 1 mM surfactin
`
`MATERIALS AND METHODS
`
`44
`
`
`1 of 6
`
`FRESENIUS-KABI, Exh. 1025
`
`

`
`VOL. 63, 1997
`
`ANTIMYCOPLASMA AGENT SURFACTIN
`
`45
`
`100-␮l portion of a cell culture supernatant was transferred into a sterile Ep-
`pendorf tube and boiled for 5 min. The tube was centrifuged briefly, and the
`supernatant was diluted 1:10 with sterile double-distilled water. A mycoplasma
`PCR primer set obtained from Stratagene (La Jolla, Calif.) was used to differ-
`entiate mycoplasma types, and the primer set of van Kuppeveld et al. (28) was
`used to identify groups, as described elsewhere (18). Each final PCR mixture
`(total volume, 50 ␮l) contained 10 ␮l of template and 40 ␮l of amplification
`mixture containing 16 mM (NH4)2SO4, 67 mM Tris-HCl (pH 8.8 at 25⬚C), 0.01%
`Tween 20, 1.5 mM MgCl2, each deoxynucleoside triphosphate at a concentration
`of 50 ␮M, each primer at a concentration of 0.2 ␮M, and 1.25 U of Thermus
`aquaticus DNA polymerase (InViTek, Berlin, Germany). Each reaction mixture
`was overlaid with 2 drops of mineral oil. PCR amplification was performed with
`a Perkin-Elmer thermal cycler. The initial 5-min denaturation step at 94⬚C was
`followed by a 1.75-min annealing step at 55⬚C. Next there was a 3-min primer
`extension step at 72⬚C, followed by a 45-s denaturation step at 94⬚C and a
`1.75-min annealing step at 55⬚C. The remaining 40 cycles each consisted of
`extension for 3 min at 72⬚C, denaturation for 45 s at 94⬚C, and annealing for 45 s
`at 55⬚C. The posttreatment steps consisted of 10 min at 72⬚C and then 10 min at
`27⬚C. The amplified PCR products were separated by standard agarose gel
`electrophoresis in 2 and 4% (wt/vol) agarose gels for the Stratagene system and
`the system of van Kuppeveld et al., respectively, and were visualized by ethidium
`bromide staining.
`Electron microscopy. ML cells were propagated in DMEM supplemented with
`5% FCS. After the culture was confluent, the surfactin concentration in the
`medium was adjusted to 12.5, 25, or 50 ␮M. After 1 h of incubation at 37⬚C, the
`cells were fixed with 2.5% glutardialdehyde in PBS and concentrated by gentle
`centrifugation after successive washes with PBS. The fixed pellets were enclosed
`in an agar block, dehydrated in ethanol, and embedded in Epon by standard
`techniques (8). Ultrathin sections were examined at 80 kV with a Zeiss model
`EM 902 electron microscope.
`Mycoplasma elimination procedures. Reproducible decontamination of ad-
`herent and nonadherent cell lines was performed as follows. The surfactin used
`in the procedures described below was diluted in PBS to a concentration of 1
`mM, autoclaved at 122⬚C for 30 min, and directly added to the medium.
`(i) Adherent cell lines. Approximately 106 freshly trypsinized cells of an ad-
`herent cell line were transferred into a petri dish (diameter, 10 cm) containing 10
`ml of DMEM supplemented with 5% (vol/vol) FCS and 40 ␮M surfactin. The
`cells were maintained in this medium for one passage (approximately 3 to 8 days)
`under normal growth conditions. After this, the cells were subcultured in sur-
`factin-free standard medium.
`(ii) Suspension cell lines. Depending on the proliferation rate, 1 ⫻ 105 to 5 ⫻
`105 cells of a suspension cell line were transferred into a centrifuge tube con-
`taining the elimination mixture (5 ml of RPMI 1640 medium, 60 ␮M surfactin,
`5% FCS, and 50% 0.125% [wt/vol] trypsin–5 mM EDTA in PBS). The mixture
`was vortexed and shaken gently for 30 min at room temperature. After centrif-
`ugation at 600 ⫻ g for 10 min, the supernatant was discarded. The cells were
`resuspended in 5 ml of RPMI 1640 medium supplemented with 30 ␮M surfactin
`and 5% (vol/vol) FCS. The cells were incubated in this medium for 3 days in a
`25-ml culture flask under normal growth conditions and pelleted by centrifuga-
`tion for 10 min at 600 ⫻ g, the supernatant was discarded, and the cells were
`resuspended in 5 ml of the elimination mixture. Then the steps from vortexing to
`incubation for 3 days in a 25-ml culture flask under normal growth conditions
`were repeated. After 3 days of cultivation in surfactin-containing medium, the
`cells were transferred into surfactin-free growth medium.
`After surfactin treatment, the cultures were grown for at least four passages
`before samples were taken for mycoplasma detection.
`
`RESULTS
`
`Effect of surfactin on cell proliferation. Recently, we inves-
`tigated the antiviral properties of surfactin (29) by using sur-
`factin-containing media and ML cells for virus titration. In the
`virus-free control cultures we observed a significant increase in
`cell proliferation, and the confluent culture was vital and
`healthy and had less contrast than an untreated culture. When
`mycoplasma-free cell cultures were treated with the drug, no
`changes in morphology or proliferation rate could be detected.
`In order to investigate the effect of surfactin in more detail,
`several cell lines were collected from different sources and
`tested for mycoplasma infection. Positive cell lines were ex-
`posed to surfactin at various concentrations. The cytotoxic
`effect of this drug was measured by the crystal violet technique
`for adherent cell lines or by the MTT assay for suspension cell
`lines (Fig. 2). The 50% cytotoxic concentrations of surfactin for
`the adherent and nonadherent cell lines were as follows: ML,
`40 ␮M; 293, 30 ␮M; Hep2, 42 ␮M; CV1, 50 ␮M; Molt 4/8, 35
`␮M; MT-4, 30 ␮M; and H9, 43 ␮M. At concentrations greater
`
`FIG. 1. Structure of surfactin.
`
`solution in phosphate-buffered saline (PBS) was sterilized by heat treatment
`(121⬚C, 30 min).
`Cells and culture conditions. All cell lines investigated for mycoplasma elim-
`ination were continuous cell lines cultured in medium without any antibiotic at
`37⬚C in a humidified atmosphere containing 5% CO2 in air. Adherent cell lines
`ML (mink lung), 293 (human embryonal kidney), CV1 (African green monkey
`kidney), and Hep2 (human larynx carcinoma) were cultivated in Dulbecco’s
`modified Eagle’s medium (DMEM) (Gibco, Uxbridge, Great Britain) supple-
`mented with 5% heat-inactivated fetal calf serum (FCS) (ICN Pharmaceuticals
`Inc., Irvine, Calif.) in petri dishes (Nunc, Roskilde, Denmark). Suspension cell
`lines MT-4 (human T-cell leukemia virus type 1 transformed), Molt 4 clone 8
`(Molt 4/8), and H9, all of which are human T-lymphoid cell lines, were grown in
`RPMI 1640 medium (ICN) supplemented with 10% (vol/vol) FCS and 2 mM
`L-glutamine in tissue culture flasks (Nunc). All suspension cell lines and adherent
`cell line CV1 were subcultured once a week, and all other adherent cell lines
`were subcultured twice a week.
`Cytotoxicity assay. The 50% cytotoxic concentrations of surfactin for the
`adherent cell lines were determined by the crystal violet dye uptake assay by
`using the method described by Flick and Gifford (7). Adherent cells (approxi-
`mately 105 cells/ml) were grown on microtiter plates (200 ␮l of cell suspension/
`well) with serial dilutions of surfactin (concentration range, 10 to 70 ␮M) mixed
`with DMEM supplemented with 5% (vol/vol) FCS. After the control culture was
`confluent, the cells were fixed with 1% glutardialdehyde, stained with crystal
`violet, washed with H2O, and dried. The dye was dissolved in 100 ␮l of ethanol-
`water-acetic acid (50:49.9:0.1). A550 values were determined with a microplate
`reader. The proliferation rates of the nonadherent cell lines after treatment with
`surfactin were determined by the colorimetric tetrazolium dye reduction assay
`and the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) as-
`say (17). The yellow compound MTT (Sigma) is reduced by mitochondrial
`dehydrogenases to the water-insoluble blue compound formazan, depending on
`the viability of the cells. A 20-␮l portion of a solution of MTT (5 mg/ml in PBS)
`was added to every well. The plate was incubated for 4 h at 37⬚C in a CO2
`incubator. After incubation 150 ␮l of medium was removed from every well
`without disturbing the cell clusters. A 100-␮l portion of acidified isopropanol (2
`ml of concentrated HCl added to 500 ml of isopropanol) was added to each
`sample, and the preparations were mixed thoroughly on a plate shaker with the
`cells containing formazan crystals. After all of the crystals were dissolved, the
`A550 values were determined with a microplate reader.
`Mycoplasma detection. (i) Cytochemical staining of DNA with DAPI. The cell
`cultures used for cytochemical staining of DNA with 4⬘,6-diamidino-2-phenylin-
`dole (DAPI) (Boehringer, Mannheim, Germany) (21) were grown on coverslips
`in petri dishes to 70% confluence. The culture medium was removed, and the
`cells were fixed with methanol at room temperature. The cells were incubated for
`5 min at 37⬚C with a DAPI staining solution (1 ␮g of DAPI per ml of methanol)
`and rinsed with methanol. Using PBS as the mounting medium, we examined the
`cells with a fluorescence microscope fitted with a Zeiss filter combination con-
`sisting of type BP 365 and 520-560 filters. Mycoplasmas appeared as bright
`yellow-green spots against a dark cytoplasmic background next to the fluores-
`cence signal of the stained nuclear DNA of the mammalian cells.
`(ii) Immunological detection: enzyme-linked immunosorbent assay (ELISA).
`Immunological detection and identification of M. orale, M. hyorhinis, A. laidlawii,
`and M. arginini were performed with a commercially available mycoplasma de-
`tection kit (Boehringer). Biotinylated polyclonal antibodies directed against spe-
`cific mycoplasma antigens were reacted with the cells. Binding of the antibodies
`was visualized by the streptavidin-alkaline phosphatase assay. After enzymatic
`hydrolysis of 4-nitrophenyl phosphate, the yellow nitrophenol product was quan-
`tified with a microplate reader (model EAR 400 AT; SLT-Labinstruments,
`Gro¨dig, Austria) at a wavelength of 405 nm.
`(iii) PCR amplification of mycoplasma rRNA. The templates used for myco-
`plasma PCR were extracts of cell-free culture media prepared by boiling. A
`
`
`2 of 6
`
`

`
`46
`
`VOLLENBROICH ET AL.
`
`APPL. ENVIRON. MICROBIOL.
`
`FIG. 2. Cytotoxic effects of surfactin on different cell lines. The inhibitory
`effects of different doses of surfactin on cell growth were determined by the
`crystal violet assay. Cell lines ML, 293, Hep2, and CV1 were treated with
`surfactin at concentrations ranging from 10 to 70 ␮M. No surfactin was added to
`the control culture. The percent growth reduction was calculated from the
`extinction difference between a surfactin-treated cell culture and the control. The
`50% cytotoxic surfactin concentrations are indicated by the dotted lines.
`
`than 70 ␮M (100% cytotoxic concentration), no cells survived
`after one passage. Surfactin concentrations less than 10 to 25
`␮M had no toxic effects on the cells.
`Mycoplasma detection and differentiation. Stock cultures
`and surfactin-treated cultures were tested for mycoplasma in-
`fection by the DAPI staining method, which showed that all of
`the stock cultures were mycoplasma positive (Fig. 3a), while
`the cultures treated with the antibiotic were free of mycoplas-
`mas (Fig. 3b). The fluorescence DAPI test is only useful for
`screening, as only massive mycoplasma contaminations can be
`detected. In order to enhance the detection limit for mycoplas-
`mas, the cultures were tested by the highly sensitive PCR and
`ELISA techniques two passages after treatment at the earliest.
`The species of contaminating mycoplasmas were identified by
`the ELISA for all of the cell lines investigated. The myco-
`plasma species residing in adherent cell lines ML, 293, Hep2,
`and CV1 was identified as M. hyorhinis. The mycoplasmas in
`suspension cell lines Molt 4/8, H9, and MT-4 belonged to M.
`orale. The Stratagene PCR method permits differentiation be-
`tween PCR products derived from different mycoplasma spe-
`cies. Figure 4 shows typical fingerprint results obtained with
`two strains of mycoplasmas. For M. orale in MT-4, Molt 4/8,
`and H9 cells, one band at 650 bp was observed, and for M.
`hyorhinis in ML, 293, Hep2, and CV1 cells, four bands at 700,
`600, 250, and 150 bp were detected. No double infections were
`observed with either method.
`Biological activity of surfactin against mycoplasmas. To un-
`derstand the mode of action of the drug, we treated a confluent
`ML cell culture that was highly contaminated with M. hyorhinis
`with surfactin at several concentrations below the 80% cyto-
`
`FIG. 3. DNA fluorescence staining of surfactin-treated cells with DAPI: ML
`cells heavily contaminated with M. hyorhinis (a) and mycoplasma-free cultures
`(b) after treatment with surfactin. Mycoplasmas were detected after the DNA in
`the culture was stained with the fluorochrome dye DAPI; they appear as small
`fluorescent spots against a dark background in the cytoplasm and intercellular
`spaces.
`
`toxic concentration (12.5, 25, and 50 ␮M) and investigated the
`effects of the drug by transmission electron microscopy (Fig.
`5). Mycoplasmas incubated in a cell culture without surfactin
`were visible as intact particles at the surfaces of the ML cells.
`After incubation of mycoplasmas with 12.5 ␮M surfactin at
`37⬚C for 1 h, we observed formation of small holes in the
`mycoplasma membrane and swelling of the particles. Espe-
`cially mycoplasmas attached to the cell surface were directly
`affected by the drug. At a surfactin concentration of 25 ␮M
`bursting of the particles was induced. Also, mycoplasmas which
`
`
`3 of 6
`
`

`
`VOL. 63, 1997
`
`ANTIMYCOPLASMA AGENT SURFACTIN
`
`47
`
`FIG. 4. Gel electrophoretic analysis of PCR amplification products of cell
`culture supernatants before and after surfactin treatment. Only mycoplasma
`rRNA was amplified by PCR when the type-specific Stratagene mycoplasma
`primer set was used (a) or the species-specific primer pair of van Kuppeveld et
`al. was used (b). Templates obtained from the culture medium containing sur-
`factin-treated cells (post) produced no PCR signals. The Stratagene primer set
`revealed the M. hyorhinis-specific pattern (700, 600, 250, and 150 bp) for the
`untreated ML culture (pre) and an M. orale-specific pattern (650 bp) for the
`untreated Molt 4/8 culture (pre). The PCR of van Kuppeveld et al. (28) was
`considered mycoplasma positive if a single 280-bp product was amplified.
`
`were hidden in pockets and clefts of the cell membrane began
`to disintegrate. Finally, at a surfactin concentration of 50 ␮M
`disruption of the mycoplasma lipid bilayer included total dis-
`integration of the membrane systems, which led to bursting of
`all microorganisms.
`Mycoplasma elimination procedure. On the basis of the
`observations described above, we developed a procedure for
`removing mycoplasmas from mammalian suspension and
`monolayer cells by using the lytic effect of surfactin on these
`organisms. In order to determine the optimal dose and dura-
`tion of exposure, adherent cell lines Hep2, ML, 293, and CV1
`were treated with 10, 20, 30, 40, 50, and 60 ␮M surfactin over
`a period of one or two cell passages. After lipopeptide treat-
`ment, cells were cultivated in the absence of antibiotics so that
`a low level of infection would be detectable after two passages
`by each of the methods. Contamination with mycoplasmas was
`assessed by using at least three different procedures. For all
`cell lines treatment with 10 or 20 ␮M surfactin for two passages
`did not completely eliminate mycoplasma contamination, but
`after incubation for one passage with 30 ␮M surfactin for ML
`cells and with 40 ␮M surfactin for CV1 and Hep2 cells, the
`cultures were free of mycoplasmas. For cell line 293 two con-
`secutive treatments with 40 ␮M surfactin were necessary for
`effective elimination of all viable M. hyorhinis. The viability of
`these cells was significantly lower than the viability of CV1 and
`Hep2 cells after one passage with 40 ␮M surfactin and the
`viability of ML cells after one passage with 30 ␮M surfactin. A
`surfactin concentration of 50 or 60 ␮M led to complete re-
`moval of mycoplasmas, but the cells were either dead or seri-
`
`FIG. 5. Thin-section electron micrographs of mycoplasma-contaminated ML
`cells before and after addition of surfactin. When an ML cell culture that was
`highly contaminated with M. hyorhinis was confluent, surfactin was added to the
`culture at final concentrations of 12.5, 25, and 50 ␮M, and the preparations were
`incubated for 60 min at 37⬚C. No surfactin was added to the control culture.
`Interaction of the membrane-active surfactant with the outer part of the lipid
`membrane bilayer induced permeability changes. At the higher concentrations
`the drug finally caused the mycoplasma membrane system to burst by a detergent
`effect.
`
`ously damaged, which led to unacceptably low proliferation
`rates. Therefore, we recommend using a surfactin concentra-
`tion of 40 ␮M for two passages as a standard elimination
`procedure for adherent cell lines in order to be sure that all cell
`lines are devoid of mycoplasma infections. The efficiency of the
`treatment is shown in Table 1 and Fig. 4.
`Using the results obtained with the adherent cell lines, we
`determined the duration of surfactin exposure and the concen-
`trations required for treatment of suspension cell lines Molt
`4/8, H9, and MT-4, as described above. By double treatment
`with surfactin at concentrations of 30 to 50 ␮M we obtained
`complete elimination of mycoplasmas in all cell lines. Unfor-
`tunately, these results were not reproducible when several cell
`cultures were tested in different mycoplasma elimination ex-
`periments. A surfactin concentration of 50 or 60 ␮M led either
`
`
`4 of 6
`
`

`
`48
`
`VOLLENBROICH ET AL.
`
`APPL. ENVIRON. MICROBIOL.
`
`TABLE 1. Mycoplasma elimination by surfactina
`
`DISCUSSION
`
`Cell line
`
`Detection assay
`
`Passage(s) tested for
`mycoplasma
`contamination
`
`Adherent
`cell lines
`ML
`
`CV1
`
`293
`
`Hep2
`
`Suspension
`cell lines
`Molt 4/8
`
`MT-4
`
`H9
`
`DAPI
`ELISA
`PCR (Stratagene method)b
`PCR (van Kuppeveld method)c
`DAPI
`ELISA
`PCR (Stratagene method)
`PCR (van Kuppeveld method)
`DAPI
`PCR (Stratagene method)
`PCR (van Kuppeveld method)
`DAPI
`PCR (Stratagene method)
`PCR (van Kuppeveld method)
`
`2, 5, 15, 20
`6
`2, 5, 15, 20
`2, 5, 15, 20
`2, 3, 4
`3
`3, 5
`2, 5
`8, 17, 28
`4, 12, 28
`4, 12, 14, 18, 25, 28
`8, 20
`2, 5
`4, 9, 16, 20
`
`ELISA
`PCR (Stratagene method)
`PCR (van Kuppeveld method)
`ELISA
`PCR (Stratagene method)
`PCR (van Kuppeveld method)
`ELISA
`PCR (Stratagene method)
`PCR (van Kuppeveld method)
`
`10
`4, 9, 15, 20
`3, 4, 6, 9, 15, 20
`10
`4, 6, 20
`3, 4, 6, 10, 16, 20
`10
`3, 5, 9, 20
`3, 5, 7, 9, 13, 16, 20
`
`a The effectiveness of the elimination process was controlled during several
`cell passages by using different mycoplasma detection methods. All of the cul-
`tures tested after surfactin treatment were free of contaminants. In the cell
`passages tested no mycoplasma recontamination occurred during cultivation.
`b PCR performed with primers obtained from Stratagene.
`c PCR performed by the method described by van Kuppeveld et al. (28).
`
`to cell death or to the reduction of proliferation rates below
`acceptable levels after one to two passages.
`To eliminate the possibility that mycoplasmas hidden in in-
`tercellular spaces, as well as in pockets and clefts of the cell
`membrane, would escape contact with the drug, we used tryp-
`sin to detach the cells from each other and to smooth the cell
`surfaces. By also using a pretreatment with 60 ␮M surfactin for
`30 min, we were able to effectively eliminate all mycoplasmas
`from suspension cells (Fig. 4 and Table 1). A pronounced
`decrease in the viability of the culture cells tested of approxi-
`mately 50 to 85% was observed after the second treatment, but
`enough viable suspension cells could be recovered for further
`subcultivation. Under these conditions our results were nicely
`reproducible.
`We tested the reemergence of residual mycoplasmas by cul-
`tivating the cells for up to 28 passages. As Table 1 shows, in the
`periods investigated all of the cell lines could be cultivated so
`that they were free of contamination. In no case was any
`permanent growth inhibition of mammalian cells detected.
`The surviving cells regained the normal rate of growth and
`compensated for the loss in cell numbers after one to two
`passages. The initial cell density and the FCS concentration
`were important factors for the success of this procedure.
`Higher cell densities and FCS concentrations in the reaction
`mixture decreased the efficiency of elimination of mycoplas-
`mas, but lower concentrations decreased the level of viability.
`
`Our investigation of the mycoplasmacidal effect of surfactin
`was initiated by the observation that mycoplasma-contami-
`nated adherent and nonadherent cells exhibited improved pro-
`liferation rates and changes in morphology after treatment
`with the drug. These effects of surfactin were described previ-
`ously by Hosono and Suzuki (11) with Chinese hamster ovary
`(CHO-K1) cells and were similar to the effects of dibutyryl
`cyclic AMP. This activity was interpreted as being due to in-
`hibition of cyclic AMP phosphodiesterase by surfactin (10).
`Our electron microscopic studies provided evidence that sur-
`factin affects the envelopes of contaminating mycoplasmas.
`Obviously, surfactin disrupts the plasma membrane, which is
`its primary site of activity. It causes leakage, at higher concen-
`trations it leads to complete disintegration of the membrane
`systems, and finally it causes the mycoplasmas to burst. Erad-
`ication of the contaminants resulted in native morphology and
`a native proliferation rate of the mammalian cells.
`Previous studies with artificial membranes (15, 23), proto-
`plasts (26), and eukaryotic cells (11) demonstrated that surfac-
`tin binds readily to cell membranes with a high degree of
`selectivity, depending on the membrane lipid composition. The
`fatty acid portion of surfactin is anchored in the lipid bilayer,
`showing high affinities to cholesterol and phospholipids (11,
`16). In Mycoplasma species cholesterol was found at levels
`comparable to those in the plasma membranes of eukaryotic
`cells (25 to 30% [wt/wt] of the total membrane lipids) (19, 20).
`The higher levels of membrane phospholipids (especially phos-
`phatidylglycerol, phosphatidylcholine, and phosphatidyleth-
`anolamine) found in Mycoplasma cells compared with eukary-
`otic cells may result in the greater susceptibility of mycoplasma
`membranes to surfactin (19, 20). Surfactin is active against M.
`hyorhinis and M. orale at concentrations greater than 30 ␮M. A
`critical micellar concentration of 10 ␮M was determined for
`surfactin (12) in 0.1 M NaHCO3 (pH 8.7). Obviously, surfactin
`interacts with the mycoplasma membrane in micellar form,
`inducing an osmotic influx of medium and ultimately complete
`disruption of the cells. Our observations of the lytic effect of
`surfactin on mycoplasmas were utilized to develop a procedure
`for elimination of mycoplasmas from mammalian cell cultures.
`All adherent cell lines and suspension cells tested could be
`successfully cleansed of two of the most common mycoplasmas
`associated with such cell cultures, M. hyorhinis and M. orale,
`respectively. The efficiency of the mycoplasma elimination pro-
`cedure which we developed was demonstrated by several
`highly sensitive techniques. No reemergence of the contami-
`nants was detected, which meant that the mycoplasmas were
`completely eradicated and not merely arrested in growth by a
`bacteriostatic effect of the drug. All cultures from which My-
`coplasma species were eliminated were grown under condi-
`tions under which recontamination by other contaminated cell
`cultures was not possible (i.e., in a separate incubator or lam-
`inar airflow biohazard cabinet).
`Compared with other mycoplasma elimination protocols
`performed with antibiotics, such as ciprofloxacin and enro-
`floxacin (quinolone derivatives; trade names, Ciprobay and
`Baytril, respectively), tiamulin and minocycline (pleuromutilin
`and tetracycline derivative, respectively; combined to form the
`commercially available product BM-Cyclin), or Mycoplasma
`Removal Agent (a 4-oxo-quinoline-3-carboxyl acid derivative)
`(6, 9, 22, 27), the method described here has the advantage of
`being more effective. Therefore, cells do not have to be pro-
`tected against reemergence of contaminants, and time-con-
`suming and labor-intensive replenishment of the antibiotic
`during cultivation is not necessary. Indeed, the only antibiotics
`
`
`5 of 6
`
`

`
`VOL. 63, 1997
`
`ANTIMYCOPLASMA AGENT SURFACTIN
`
`49
`
`able to eliminate mycoplasmas in a cell culture without pro-
`ducing high cytotoxicity are the tetracyclines and the fluoro-
`quinolones, both of which are known to penetrate cells and
`inhibit mycoplasma metabolism (4, 6, 27). The mycoplasmas
`blocked in growth by these antibiotics are removed from cell
`cultures mainly by dilution. In some cases removal is incom-
`plete and mycoplasmas appear again after a few passages. In
`contrast to these mycoplasmastatic drugs, surfactin does not
`inhibit growth of the mycoplasmas; instead, due to its physico-
`chemical mechanism of action, it kills the contaminating my-
`coplasmas directly. Therefore, only a short treatment is neces-
`sary for complete disintegration of these organisms. On the
`basis of the mode of action of surfactin it is expected that
`mycoplasmas will not develop resistance to this drug, which is
`a major advantage compared with the antibiotics mentioned
`above.
`The damaging effects of the agent on plasma membranes
`were mitigated in a culture medium with a high serum content,
`probably as a result of the high competitive binding capacity of
`the large amounts of proteins or lipids in the medium, as
`described previously (11, 29). Therefore, surfactin is not useful
`for eliminating mycoplasmas from samples with high protein
`contents. However, the novel mycoplasma inactivation proce-
`dure developed in this study certainly is of great medical and
`biotechnological interest. Mycoplasmas are capable of altering
`many properties of mammalian cells and parameters measured
`in cell cultures, which leads to unreliable results. Surfactin can
`be used to keep growth, morphology, and metabolism in a
`natural state. In addition, surfactin is well-suited to eliminate
`mycoplasma infections in biotechnological and pharmaceutical
`products, including vaccines, therapeutics, care products, and
`diagnostics derived from in vitro systems, as advised by most
`governmental regulatory agencies. Due to the low toxicity of
`surfactin in in vivo models (1), a small amount of this com-
`pound in the end product might be acceptable. Because of its
`antimicrobial and antiviral properties, surfactin may be used as
`an additive to microbicides for the prevention of sexually trans-
`mitted diseases caused by genital tract mycoplasmas (4, 30)
`and enveloped viruses, includin