Glycoprotein Excreted by Pseudoalteromonas antarctica NF3 as a

Jan 6, 1998 - Abstract. The ability of an exopolymer of glycoproteic character (GP) excreted by a new Gram-negative species Pseudoalteromonas antarcti...
0 downloads 9 Views 197KB Size
Download PDF
42

Langmuir 1998, 14, 42-48

Glycoprotein Excreted by Pseudoalteromonas antarctica NF3 as a Coating and Protective Agent of Liposomes against Sodium Dodecyl Sulfate A. de la Maza,*,† J. L. Parra,† M. Sabe´s,‡ F. Congregado,§ N. Bozal,§ and J. Guinea§ Departamento de Tensioactivos, Centro de Investigacio´ n y Desarrollo (C.I.D.), Consejo Superior de Investigaciones Cientı´ficas (C.S.I.C.), C/. Jorge Girona, 18-26, 08034 Barcelona, Spain, Unitat de Biofisica, Department de Bioquı´mica i Biologia Molecular, Facultat de Medicina, Universitat Auto` noma de Barcelona, Edifici M, 08193 Bellaterra (Barcelona), Spain, and Departamento de Microbiologı´a, Facultad de Farmacia, Universidad de Barcelona, Av. Joan XXIII, s/n 08028 Barcelona, Spain Received April 18, 1997X The ability of an exopolymer of glycoproteic character (GP) excreted by a new Gram-negative species Pseudoalteromonas antarctica NF3, to coat phosphatidylcholine (PC) liposomes and to protect these bilayers against the action of the sodium dodecyl sulfate (SDS) surfactant was investigated. Transmission electron microscopy micrographs of freeze fractured liposome/GP aggregates reveal that the addition of GP to liposomes led to the formation of a film (polymer adsorbed onto the bilayers) that tightly coated PC bilayers. The complete coating was achieved at a PC:GP weight ratio of about 9:1. Higher GP amounts resulted in a growth of this film, which exhibited at the highest GP proportion (50% of GP in weight) a multilayered structure. An increasing resistance of PC liposomes to be affected by SDS at both subsolubilizing and solubilizing levels occurred as the proportion of GP in the system rose, although this protective effect was more effective at low GP proportions (PC:GP weight ratios from 9:1 to 8:2). Although a direct dependence was found between the growth of the enveloping structure and the resistance of the coated liposomes to be affected by SDS, the best protection occurred when this structure was a thin film (thickness of about 20-25 nm for a PC:GP weight ratio ranging from 9:1 to 8:2).

Introduction Cell surface structures on prokaryotes are a twodimensional array of proteinaceous subunits termed S-layers.1-3 Isolated S-layer subunits from many organisms reassemble into regularly structured lattices in the presence or absence of supporting layers or interfaces.4,5 Recently, it was demonstrated that a solubilized S-layer protein form of Bacillus coagulans E38-66 can recrystallize onto positively charged phosphatidylcholine (PC) unilamellar liposomes. This characteristic feature is being exploited at present for coating and stabilizing these bilayer structures in spite of the complex process of isolation of these crystalline cell surface layers.6,7 These recrystallized S-layers could subsequently be used as a regularly structured matrix for immobilizing functional molecules. Interest in the ecology of microbial populations has also focused attention on the importance of extracellular †

C.I.D., C.S.I.C. Universitat Auto`noma de Barcelona. § Universidad de Barcelona. X Abstract published in Advance ACS Abstracts, December 15, 1997. ‡

(1) Sleytr, U. B. Int. Rev. Cytol. 1978, 53, 1-64. (2) Sleytr, U. B.; Messner, P. Crystalline Bacterial Cell Surface Layer; Sleytr, U. B., Messner, P., Pum, D., Sa´ra, M., Eds.; Springer-Verlag: Berlin, 1988; pp 160-186. (3) Austin, J. W.; Stewart, M.; Murray, R. G. J. Bacteriol. 1990, 172, 808-817. (4) Sleytr, U. B.; Messner, P. Electron Microscopy of Subcellular Dynamics; Plattner, H., Ed.; CRC Press: Boca Raton, FL, 1989; pp 13-31. (5) Pum, D.; Weinhandel, M.; Ho¨del, C.; Sleytr, U. B. J. Bacteriol. 1993, 175, 2762-2766. (6) Sleytr, U. B. USA patents 4,849,109 and 4,886,604, 1989. (7) Ku¨pcu¨, S.; Sa´ra, M.; Sleytr, U. B. Biochim. Biophys. Acta 1995, 1235, 263-269.

polymers produced by cells, either as adhesive polymers or as structural polymer biofilm matrices.8-10 In aquatic habitats microbial polymers occur as discrete, firmly cellbound material or as slime fibers loosely associated or nonassociated with the cell walls.11 Since bacteria are unicellular life forms, the different supramolecular architectures of the envelope function as an important interface between the environment and the cell. A number of investigations have been devoted to the understanding of the principles governing the interaction of sodium dodecyl sulfate (SDS) with simplified membrane models.12-18 This interaction in excess water leads to the breakdown of lamellar structures and to the formation of lipid-surfactant mixed micelles.19 In this work we studied the ability of the glycoprotein (8) Beveridge, T. J. Bacteria in Nature, Structure, Physiology, and Genetic Adaptability; Poindexter, J. S., Leadbetter, E. R., Eds.; Plenum: New York, 1989; Vol. 3, pp 1-65. (9) Beveridge, T. J.; Graham, L. L. Microbiol. Rev. 1991, 55, 684705. (10) Sleytr, U. B.; Messner, P. Encyclopedia of Microbiology; Lederberg, J., Ed.; Academic Press: San Diego, CA, 1992; Vol. 1, pp 605614. (11) Read, R.; Costerton, J. W. Can. J. Microbiol. 1987, 33, 10801090. (12) Urbaneja, M. A.; Alonso, A.; Gonza´lez-Man˜as, J. M.; Gon˜i, F. M.; Partearroyo, M. A.; Tribout, M.; Paredes, S. Biochem. J. 1990, 270, 305-308. (13) Inuoe, T.; Yamahata, T.; Shimozawa, R. J. Colloid Interface Sci. 1992, 149, 345-358. (14) Downing, D. T.; Abraham, W.; Wegner, B. K.; Willman, K. W.; Marshall, J. L. Arch. Dermatol. Res. 1993, 285, 151-157. (15) de la Maza, A.; Parra, J. L. Langmuir 1993, 9, 870-873. (16) de la Maza, A.; Parra, J. L. Langmuir 1995, 11, 2435-2441. (17) de la Maza, A.; Parra, J. L. Langmuir 1996, 12, 6218-6223. (18) Silvander, M.; Karlsson, G.; Edwards, K. J. Colloid Interface Sci. 1996, 179, 104-113. (19) Lichtenberg, D. Biochim. Biophys. Acta 1985, 821, 470-478.

S0743-7463(97)00403-4 CCC: $15.00 © 1998 American Chemical Society Published on Web 01/06/1998

Liposome Protection against Surfactants

(GP) excreted by a new Gram-negative species, Pseudoalteromonas antarctica NF3, to coat liposomes and to protect these vesicles against the action of SDS. This surfactant has been selected given its activity on biological membranes.20-23 The EM freeze-fractured images of the GP/PC aggregates formed together with their interaction with SDS may enhance our understanding of the ability of this new compound to coat and to protect biological membranes against the action of this surfactant. Materials and Methods Phosphatidylcholine (PC) was purified from egg lecithin (Merck, Darmstadt, Germany) according to the method of Singleton24 and was shown to be pure by thin layer chromatography. Dipalmitoyl phosphatidylcholine (DPPC) was purchased from Sigma and used without further purification. Sodium dedecyl sulfate (SDS) was purchased from Merck and further purified by column chromatography.25 Piperazine-1,4bis(2-ethanesulfonic acid) (PIPES) buffer obtained from Merck was prepared as 20 mM PIPES adjusted to pH 7.20 with NaOH, containing 110 mM Na2SO4. 5(6)-Carboxyfluorescein (CF) was obtained from Eastman Kodak (Rochester, NY) and was purified by column chromatography.26 Glycoprotein Analysis. Obtaining and purifying the exopolymer has been previously reported.27,28 Polymer protein was determined using bovine serum albumin as standard. Total carbohydrate content was calculated using glucose as standard.27,28 Uronic acids were determined by the carbazole reaction described by Bitter and Muir.29 Total lipids were determined with the lipid test-combination.30 For amino acid analysis, dried polymer samples were hydrolyzed in 6 N HCl for 24 h at 110 °C in sealed tubes. Hydrolyzates were evaporated to dryness and redissolved in water. Amino acid concentration was determined using a Beckman 119-CL amino acid analyzer (Beckman Instrument, Inc., Palo Alto, CA); the identification of each amino acid was established with standards supplied by Sigma. To determine the protein molecular weight, a precision column PC 3.2/30 (2.4 mL) prepacked with Superdex TM 75 (Pharmacia LKB) was used to perform a high-resolving gel filtration separation of the protein fraction of the exopolymer obtained in mineral medium. The buffer used to perform the gel filtration was 0.05 M sodium phosphate pH 7 + 0.15 M NaCl.27 Preparation and Characterization of Liposomes/GP Aggregates. Unilamellar PC liposomes of a defined size (about 200 nm, PC concentration 5.0 mM) were prepared by extrusion of large unilamellar vesicles previously obtained by reverse phase evaporation in PIPES buffer.16 PC concentration was determined by thin-layer chromatography coupled to an automated flame ionization detection system (TLC-FID, Iatroscan MK-5, Iatron Lab. Inc. Tokyo, Japan).31 DPPC liposomes were prepared by mechanical dispersion at 50 °C. (20) Moon, K. C.; Maibach, H. I. Exogenous Dermatoses: Environmental Dermatitis; Menne´, T., Maibach, H. I., Eds.; CRC Press: Boca Raton, FL, 1991; pp 217-226. (21) Wilhelm, K. P.; Surber, C.; Maibach, H. I. J. Invest. Dermatol. 1991, 96, 963-967. (22) Braun-Falco, O.; Korting, H. C.; Maibach, H. I. Liposome Dermatics (Griesbach Conference); Braun-Falco, O., Korting, H. C., Maibach, H. I., Eds.; Springer-Verlag: Berlin, 1992; p 301 (23) Wilhelm, K. P.; Surber, C.; Maibach, H. I. J. Invest. Dermatol. 1991, 97, 927-932. (24) Singleton, W. S.; Gray, M. S.; Brown, M. L.; White, J. L. J. Am. Oil Chem. Soc. 1965, 42, 53-57. (25) Rosen, M. J. J. Colloid Interface Sci. 1981, 79, 587-593. (26) Weinstein, J. N.; Ralston, E.; Leserman, L. D.; Klausner, R. D.; Dragsten, P.; Henkart, P.; Blumenthal, R. Self-Quenching of Carboxyfluorescein Fluorescence: Uses in Studying Liposome Stability and Liposome Cell Interaction. In Liposome Technology; Gregoriadis, G., Ed.; CRC Press, Boca Raton, FL, 1986; Vol. III, Chapter 13. (27) Bozal, N.; Manresa, A.; Castellvi, J.; Guinea, J. Polar Biol. 1994, 14, 561-567. (28) Bozal, N.; Tudela, E.; Rosello-Mora, R.; Lalucat, L.; Guinea, J. Int. Jl. Syst. Bacteriol. 1997, 42, 231-237. (29) Bitter, T.; Muir, H. M. Anal. Biochem. 1962, 4, 330-334. (30) Lipid Test-Combination, Boehringer cat. No. 124 303. (31) Ackman, R. G.; McLeod, C. A.; Banerjee, A. K. J. Planar Chromatogr. 1990, 3, 450-490.

Langmuir, Vol. 14, No. 1, 1998 43 To determine the liposome permeability alterations, PIPES buffer was supplemented with 100 mM CF. Liposomes were extruded through 800-200 nm polycarbonate membranes to achieve a uniform size distribution. PC liposomes were combined with the same volume of GP aqueous dispersions in PIPES buffer to obtain PC/GP mixtures (weight ratios ranging from 9:1 to 5:5). These mixtures were incubated at 25 °C in a test tube rotator, type 3025 (GFL Burgwedel, Germany) with a rotation speed of 10 min-1 for 30 min. The resulting liposome/GP aggregates were freed of the GP compound nonassembled with liposomes. To this end, these aggregates were sedimented for 30 min at 40000g at 10 °C and then resuspended in PIPES buffer. No PC was detected by TLC-FID31 in any supernatant in spite of its opalescent aspect due to the presence of free GP compound. These suspensions were used to determine the size distribution of these aggregates and to study their solubilization by SDS. To determine permeability changes, CF containing liposome/GP aggregates were freed of unencapsulated CF on a Sephadex G-50 medium gel bed (Pharmacia Biotech, Uppsala, Sweden).32 The integrity of the aggregates formed was examined by freezefracture transmission electron microscopy (TEM), and their size distribution and polydispersity index (PI) were determined with a photon correlator spectrometer (Malvern Autosizer 4700c PS/ MV). The studies were made by particle number measurement at 25 °C with a reading angle of 90°. Differential Scanning Calorimetry (DSC). DSC measurements were carried out on a MicroCal MC 2 microcalorimeter (MicroCal, Inc., Northampton, MA). Data were processed with Origin software. DPPC and DPPC/GP samples (GP weight ratio 9:1) were introduced into the chamber cell at a final DPPC concentration of 1.64 mg/mL in water. The same medium was placed in the reference cell. The heating rate was 69 °C/h. Data were collected from 20 to 100 °C. After each scan, samples were cooled to 15 °C and reheated at the same conditions in order to compare calorimetric behaviors. Thermograms of water were taken as instrumental baseline and substracted from sample scans. Interaction of Coated Liposomes with SDS. The permability changes of uncoated and coated PC vesicles resulting in the action of SDS were determined quantitatively by monitoring the increase in the fluorescence intensity due to the release of the CF trapped into these structures.16 The measurements were made at 25 °C with a spectrofluorophotometer, Shimadzu RF540 (Kyoto, Japan). On excitation at 495 nm, a fluorescence maximum emission of CF was obtained at 515.4 nm. Equal volumes of SDS (2.0 mM) were added to the liposomes/GP aggregates (freed of the unencapsulated CF and of the nonassembled exopolymer), and the resulting systems were left to equilibrate for 4 h. This interval was chosen as the minimum period of time needed to achieve a constant level of CF release in the PG:GP weight ratio range investigated. The experimental determination of this interval is indicated in the results and discussion section. The percentage of CF released was calculated by means of the equation

% CF release ) (IT - I0)/(I∞ - I0) × 100

(1)

where I0 is the initial fluorescence intensity of CF-loaded liposome/ GP aggregates in the absence of surfactant and IT is the final fluorescence intensity measured after 4 h. I∞ is the fluorescence intensity remaining after the destruction of liposomes by the addition of 150 µL of 10% (v/v) aqueous solution of Triton X-100 (final PC concentration 5.0 mM). As for the liposome solubilization, it has been demonstrated that turbidity determinations constituted a very convenient technique for the quantitative study of this process.12,33,34 Accordingly, the solubilizing perturbations produced by SDS in pure or coated PC vesicles were monitored using this technique. The overall solubilization was characterized by two parameters termed SSAT and SSOL corresponding to the surfactant concentra(32) New, R. R. C. Liposomes: a Practical Approach; I.R.L. Press: Oxford, 1990; Chapter 3. (33) Partearroyo, M. A.; Urbaneja, M. A.; Gon˜i, F. M. FEBS Lett. 1992, 302, 138-140. (34) Ruiz, M. B.; Prado, A.; Gon˜i, F. M.; Alonso, A. Biochim. Biophys. Acta 1994, 1193, 301-306.

44

Langmuir, Vol. 14, No. 1, 1998

de la Maza et al.

Table 1. Mean Size Distributions (nm) and Polydispersity Indices of Liposome/GP Aggregates versus the PC:GP Weight Ratio in the System PC:GP weight ratio

mean size distribution (nm)

polydispersity index

10:0 9:1 8:2 7:3 6:4 5:5

204 216 256 305 325 340

0.104 0.132 0.129 0.134 0.139 0.148

tion at which turbidity starts to decrease with respect to the initial value and shows no further decrease. These values corresponded to the concentrations at which SDS-saturated liposome/GP aggregates and led to a solubilization of these structures. Aliquots of liposome/GP aggregates (freed of the nonassembled exopolymer) were mixed with equal volumes of SDS solutions, and the resulting mixtures were left to equilibrate for 24 h. Turbidity measurements were made using a spectrofluorophotometer, Shimadzu RF-540, at 25 °C with both monochromators adjusted to 500 nm.16 The assays were carried out in triplicate and the results given are the average of those obtained. TEM Applied to Freeze-Fractured Liposome/GP Aggregates. Liposome/GP aggregates were mounted on gold stubs and quickly frozen by dipping them into nitrogen-cooled liquid propane. Freeze-fracturing, etching, and coating were carried out at -110 °C under a vacuum better than 5 × 10-7 mbar using a BAL-TEC instrument, type BAF-060 (Balzers AG, Balzers, Lichtenstein). The platinum/carbon-coated replicas were cleaned overnight in sodium hypochlorite. After the samples were rinsed in distilled water, they were collected on Formvar-coated copper grids. The cleaned replicas were examined with a Hitachi H-600AB transmission electron microscope operating at 75 kV.

Results and Discussion The composition of the glycoprotein produced by Pseudoalteromonas antarctica NF3 varied depending on the medium used. When glucose was supplied as the sole carbon source (mineral medium), the protein content was approximately 88% and no lipid was detected. However, the polymer recovered from cultures incubated in Trypticase Soy Broth (rich medium) was composed of 76% protein, 6.25% lipid, and 14% carbohydrate, which was higher than the polymer recovered from cells grown on mineral medium (8%). Amino acid composition of the protein fraction shows that histidine, alanine, threonine, serine, aspartic acid, and glutamic acid were in the higher proportions when NF3 was grown in the mineral medium, while aspartic and glutamic acids, glycine, isoleucine, leucine, and lysine were the main amino acid residue components in the slime produced in rich medium. The molecular weight of protein fraction determined by gel filtration showed that 49% of the compound was a protein of about 40 000 Da and 23% was a small protein of about 8900 Da.27 The glycoprotein used in the present study was only that recovered from cultures incubated in a rich medium. Interaction of SDS with Coated PC Liposomes. We previously determined the size distribution and polydispersity index (PI) of the liposome/GP aggregates (freed of the nonassembled GP) resulting in the association of PC liposomes with GP (at different PC:GP weight ratios, PC concentration 5.0 mM) (Table 1). A progressive rise in both parameters took place as the proportion of GP in the system increased, the curves showing always a monomodal distribution. The size distribution of these aggregates after preparation was reasonably homogeneous (PI always lower than 0.15). The mean size after addition of equal volumes of PIPES buffer and equilibration for 24

Figure 1. Time curves of the release of CF trapped into uncoated and coated PC liposomes (PC:GP weight ratios from 10:0 to 6:4) caused by the addition of a constant concentration of SDS (2.0 mM): PC concentration, 5.0 mM; PC:GP weight ratios, 10:0 (b), 9:1 (0), 8:2 (O), 7:3 (9), and 6:4 (2).

h at 25 °C always showed similar values to those initially obtained, although with a slight increase in PI (between 0.15 and 0.18). Hence, these aggregates were reasonably stable in the absence of SDS under the experimental conditions used. These findings suggest that the addition of GP to PC liposomes led to the formation of a GP structure that coated the liposome surface with the subsequent growth of this structure. We also investigated the capacity of SDS to solubilize GP. To this end, aqueous GP dispersions were combined with equal volumes of SDS in the concentration range needed to solubilize PC liposomes and turbidity changes of these slightly opalescent dispersions were determined vs time. After 24 h no changes in turbidity of these dispersions were detected, indicating that GP was almost unaffected by the SDS in these conditions. Hence, changes in turbidity of the liposome/GP aggregates caused by SDS were mainly attributed to the solubilization of the PC molecules building liposomes.15,19 In order to study the stability of liposome/GP aggregates against SDS at subsolubilizing level, the time needed to obtain a constant level of CF release trapped into these structures was studied. To this end, uncoated and coated liposomes (PC concentration 5.0 mM) were treated with SDS (2.0 mM) and the subsequent CF release changes were studied as a function of time. The curves obtained for different PC:GP weight ratios are shown in Figure 1. Although about 45 min was needed to achieve a CF release plateau for PC liposomes, this time clearly increased as the GP proportion in the system rose. In addition, the biphasic CF release behavior for pure PC liposomes (associated to the formation of hydrophilic pores or stable transient holes due to the surfactant incorporation into membranes18,35,36) was less pronounced as the GP proportion in the system rose. Thus, the presence of increasing GP amounts possibly hampered the formatin of the aforementioned interactions SDS/PC in coated vesicles. The spontaneous release of the CF trapped into these structures in the absence of SDS and in this period of time was negligible. The CF release curves for uncoated and coated liposomes, due to the addition of increasing SDS amounts and measured 4 h after the surfactant addition at 25 °C, are plotted in Figure 2. Increasing GP amounts in the (35) Edwards, K.; Almgren, M. Prog. Colloid Polym. Sci. 1990, 82, 190-197. (36) Edwards, K.; Almgren, M. Langmuir 1992, 8, 824-836.

Liposome Protection against Surfactants

Langmuir, Vol. 14, No. 1, 1998 45

Figure 2. Percentage changes in CF release of uncoated and coated PC liposomes (PC:GP weight ratios from 10:0 to 6:4) induced by the presence of increasing concentrations of SDS: PC concentration, 5.0 mM; PC:GP weight ratios, 10:0 (b), 9:1 (0), 8:2 (O), 7:3 (9), and 6:4 (2).

Figure 3. Percentage changes in turbidity of liposomes/GP aggregates formed in the presence of increasing amounts of GP (PC:GP weight ratios from 10:0 to 6:4) and due to the addition of SDS to the system: PC concentration, 5.0 mM; PC:GP weight ratios, 10:0 (b), 9:1 (0), 8:2 (O), 7:3 (9), and 6:4 (2).

Figure 4. (A) Variation of the SDS concentration needed to saturate (SSAT) liposome/GP aggregates versus the PC:GP weight ratio. (B) Variation of the SDS concentration needed to solubilize (SSOL) liposome/GP aggregates versus the PC:GP weight ratio.

system increased the SDS concentration needed to produce the same sublytic effect in pure PC bilayers. Thus, for the maximum GP proportion a rise of about 35-40% in the SDS concentration was needed to produce the same permeability alterations in coated structures with respect to uncoated ones. This effect could be used as a criterion for the evaluation of the stability of the coated structures against SDS at sublytic level. GP amounts higher than 50% in weight almost did not increase this surfactant concentration (results not shown). In order to study the solubilizing action of SDS on liposome/GP aggregates, turbidity changes of these systems were investigated. We first studied the time needed to achieve a complete SDS/aggregate equilibrium for each PC:GP weight ratio. The aggregates were treated with equal volumes of SDS at different concentrations, and changes in turbidity were studied versus time. About 24 h was needed to achieve a constant turbidity level, in accordance with our results reported for the interaction of SDS with PC liposomes.16 Figure 3 shows the solubilization curves of uncoated and coated liposomes (freed of the nonassembled exopolymer) arising from the addition of increasing SDS amounts. At low surfactant concentration an initial increase in the turbidity of these systems occurred. Increasing SDS amounts led to a turbidity fall up to a low constant value for liposome solubilization in all cases. A similar behavior

has been reported for the interaction of various surfactants with PC liposomes.12,13,15-18 Up to the PC:GP weight ratio 6:4, increasing GP amounts in the system increased the SDS amount needed to produce the same solubilizing effect in PC bilayers. However, the presence of higher GP proportions almost did not increase this amount (results not shown). When the SDS concentrations needed to saturate (SSAT) or solubilize (SSOL) these aggregates versus PC:GP weight ratios were plotted, the curves shown in Figure 4 were obtained. As aforementioned, up to 40% of GP the higher the GP proportion the higher the SDS amount needed to saturate and solubilize these aggregates (SDS increase of about 30% for the highest GP proportion). This finding could be also considered as a criterion for the evaluation of the stability of coated liposomes against SDS at lytic levels (solubilizing resistance). From these findings we may assume that the PC:GP weight ratios for the highest protection efficiency of liposomes against SDS do not exactly correspond to that for the maximum growth of these aggregates (see Table 1). Hence, the protection of liposomes against SDS does not seem to depend directly on the thickness of the enveloping structure given that an excessive growth of this film does not increase the protective effect. Freeze-Fractured TEM Images of Liposome/GP Aggregates. In order to clarify the aforementioned

46

Langmuir, Vol. 14, No. 1, 1998

de la Maza et al.

Figure 5. Freeze-fractured TEM micrographs of liposome/GP aggregates formed in the presence of increasing amounts of GP, the bilayer lipid concentration remaining constant (5.0 mM). Sample 1 shows a pure PC liposomes and samples 2, 3, 4, 5, and 6 show the liposome/GP aggregates for PC:GP weight ratio 9:1, 8:2, 7:3, and 5:5 and 5:5 (cross-fractured image), respectively.

controversy and to elucidate the assembly properties of these liposome/GP aggregates, freeze-fractured samples of uncoated and coated PC vesicles were examinated by

TEM, and some representative images are shown in Figure 5. Sample 1 shows a pure PC vesicle and samples 2-5 show PC vesicles coated with increasing GP amounts.

Liposome Protection against Surfactants

Increasing GP proportions in the system resulted in the spontaneous formation of a GP film that tightly coated liposomes. Sample 2 shows that already at the PC:GP weight ratio 9:1 liposomes were covered with this film. A rupture in the surface of the enveloping structure in samples 2 and 3 (see arrows) (PC:GP weight ratios 9:1 and 8:2) shows the increasing thickness of this film. These observations are consistent with the mean size of these aggregates (216 and 256 nm, respectively, Table 1) also showing the coherent structure of this covering film. Higher GP proportions (PC:GP weight ratios 7:3 and 5:5 for samples 4 and 5) led to the progressive growth of this covering surface (see arrows). Sample 6 shows a crossfractured image of coated liposomes at the highest GP proportion (corresponding to sample 5). A complex multilayered structure formed by successive GP concentric layers that coated liposomes may be observed. It is noteworthy that no ordered symmetry was found in the coating material in contrast with the aforementioned S-layers.6,7 In general terms, we may assume that although rising GP amounts led to the formation of complex multilayered structures that coated and protected liposomes against SDS, the more effective protection occurred when the enveloping structure was a thin film with a thickness not higher than 20-25 nm (PC:GP weight ratio ranging from 9:1 to 8:2). In order to study the possible conformational changes undergone in GP upon adsorption into the bilayers, calorimetric scans of dipalmitoylphosphatidylcholine (DPPD) liposomes and those coated with GP were carried out. This phospholipid was selected due to its appropriate phase transition temperature (PTT) for DSC studies. Figure 6A presents calorimetric scans for DPPC liposomes in water. It shows the pretransition at about 35 °C and the main transition centered at about 42 °C, in accordance with the results previously reported.37 When the same experiment was done with DPPC liposomes coated with GP (GP weight ratio 9:1, Figure 6B) similar thermograms were obtained in the first scans. However, changes in the pretransition and in the main transitions were observed in the second and third scans. This thermal transition with lower cooperativity than that detected for pure DPPC liposomes suggests the existence of an interaction between GP and DPPC liposomes after reheating them over the thermal transition. The extrapolation of these findings to the initial system (egg PC/GP) suggests that an interaction of low cooperativity took place between PC liposomes and the GP covering structure, given that the incubation was carried out at 25 °C, a temperature clearly higher than the PTT for PC (of about -10 °C). To determine the possible loss of protection due to the GP extraction from the bilayers, the GP covering structure was almost completely extracted from the different aggregates by extrusion through a 200 nm polycarbonate membrane. Both at sublytic and lytic levels almost 95% of protection against SDS was lost after extraction of the covering GP structure, confirming the protective efficiency of this polymer. It is known that liposomes may be sterically stabilized with polyethylene glycol molecules (PEG) covalently attached or with a polymer adsorbed onto the bilayers (adsorbed block copolymers).38 From the present findings we may assume that the adhesion of GP to PC liposomes (37) Blume, A. Applications of Calorimetry to Lipid Model Membranes. Physical Properties of Biological Membranes and Their Functional Implications; Hidalgo Plenum Press: New York, 1998; p 71-122. (38) Lasic, D. D. Liposomes: from Physics to Applications; Elsevier Science Publishers B.V.: Amsterdam, 1993; Chapter 11.

Langmuir, Vol. 14, No. 1, 1998 47

Figure 6. (A) Calorimetric scans for dipalmitoylphosphatidylcholine (DPPC) liposomes in water. (B) Calorimetric scans for DPPC liposomes coated with GP (GP weight ratio 9:1): 1, first scan; 2, second and third scans.

may be mainly ruled by direct adsorption of the glycoprotein onto the bilayers (block copolymer adsorption mechanism). The decline of the protective effect at high GP amounts could be related to the reduced growth of the covering structure in these conditions (see Table 1). This fact suggests that a maximum GP amount could be adsorbed onto the liposome surface in spite of the increasing presence of GP in the system. Another reason could be that the multilayered covering structure formed could contain domains which did not protect suitably liposomes against SDS.39 Extracted protein S-layer lattices exhibit ordered symmetry that in some cases resulted in a structured matrix useful for immobilization of functional molecules in charged liposomes.6,7 The specific ability of GP to coat uncharged liposomes shows interesting differences with respect to that reported for S-layer proteins. The easy isolation of GP spontaneously excreted by this bacterial species without chemical extractions from cell wall fragments, as occurred with conventional S-layers, together with its ability to be adsorbed onto neutral PC bilayer and to protect these structures against SDS may be considered as a promising alternative in the protection and stabilization of membrane surfaces. Studies on ice-crystal growth and lectins demonstrated that the S-layer proteins act by preferentially binding to (39) Woodle, M. C.; Lasic, D. D. Biochim. Biophys. Acta 1992, 1113, 171-199.

48

Langmuir, Vol. 14, No. 1, 1998

the nonbasal planes of ice crystals, thereby modifying their growth habit to produce unique spicular faceted structures.40 Similar studies for GP showed that this glycoprotein did not have the ability to modify the conventional morphology of ice crystals in a similar way to that reported for lectins, given its very low ability to depress the freezing point noncolligatively. Thus, one possible function of this exopolymer in its Antarctica habitat would be to physically bind the microorganism to ice, assisting its confinement to its environment. The ability of GP to coat a liposome by formation of an enveloping structure adsorbed onto (40) Rubinsky, B.; Coger, R.; Ewart, K. V.; Fletcher, G. L. Nature 1992, 360, 113-114.

de la Maza et al.

the bilayers may be associated to the formation of a protective survival domain of this species in its original habitat. Acknowledgment. The TEM analysis was performed at Barcelona University. We thank Dr. Carmen Lopez for his skillful work at the microscope. We are also grateful to Mrs. Eladia Serrano, Mr. Joaquim Villaverde, and Mr. G. von Knorring for their expert technical assistance. This work was supported by funds from DGICYT (Direccio´n General de Investigacio´n Cientı´fica y Te´cnica) (Prog. PB940043), Spain. LA970403L