Spider-Web Amphiphiles as Artificial Lipid Clusters: Design, Synthesis

The obtained amphiphiles can two-dimensionally spread like a spider web. The design, synthesis, and monolayer behaviors of the “spider-web amphiphil...
0 downloads 0 Views 248KB Size
Download PDF
6762

Langmuir 2004, 20, 6762-6769

Spider-Web Amphiphiles as Artificial Lipid Clusters: Design, Synthesis, and Accommodation of Lipid Components at the Air-Water Interface Katsuhiko Ariga,*,†,‡ Toshihiro Urakawa,‡ Atsuo Michiue,‡ and Jun-ichi Kikuchi‡ Supermolecules Group, Advanced Materials Laboratory, National Institute for Materials Science (NIMS), 1-1 Namiki, Tsukuba 305-0044, and Graduate School of Materials Science, Nara Institute of Science and Technology (NAIST), 8916-5 Takayama-cho, Ikoma 630-0192, Japan Received April 19, 2004. In Final Form: May 17, 2004 As a novel category of two-dimensional lipid clusters, dendrimers having an amphiphilic structure in every unit were synthesized and labeled “spider-web amphiphiles”. Amphiphilic units based on a LysLys-Glu tripeptide with hydrophobic tails at the C-terminal and a polar head at the N-terminal are dendrically connected through stepwise peptide coupling. This structural design allowed us to separately introduce the polar head and hydrophobic tails. Accordingly, we demonstrated the synthesis of the spider-web amphiphile series in three combinations: acetyl head/C16 chain, acetyl head/C18 chain, and ammonium head/C16 chain. All the spider-web amphiphiles were synthesized in satisfactory yields, and characterized by 1H NMR, MALDI-TOFMS, GPC, and elemental analyses. Surface pressure (π)-molecular area (A) isotherms showed the formation of expanded monolayers except for the C18-chain amphiphile at 10 °C, for which the molecular area in the condensed phase is consistent with the cross-sectional area assigned for all the alkyl chains. In all the spider-web amphiphiles, the molecular areas at a given pressure in the expanded phase increased in proportion to the number of units, indicating that alkyl chains freely fill the inner space of the dendritic core. The mixing of octadecanoic acid with the spider-web amphiphiles at the air-water interface induced condensation of the molecular area. From the molecular area analysis, the inclusion of the octadecanoic acid bears a stoichiometric characteristic; i.e., the number of captured octadecanoic acids in the spider-web amphiphile roughly agrees with the number of branching points in the spider-web amphiphile.

Introduction Although cell membranes are traditionally described as simple fluidic mixtures of lipids, proteins, and other components, a recent understanding of the cell membranes has revealed their complicated aspects, i.e., nonhomogeneous distribution of the lipid components and highly restricted nature of their motion and orientation.1,2 Especially, lipid rafts, which are clusters with an abundance of cholesterol and sphingolipids, are recognized to play important roles in various biological events such as signal transduction, membrane traffic, and virus infection.3 The construction of artificial lipid rafts would have a huge contribution not only to understanding phenomena at the cell surface but also to practical applications such as the development of vaccines to viruses and membranebased drug delivery systems. The synthesis of the amphiphile linkage and controlled incorporation of external lipid components would allow us to design various raft structures. As linked amphiphiles, gemini and oligomeric surfactants are widely investigated and have been known to exhibit properties that are totally different from those observed for monomeric ones.4 However, their linkage basically extends in one dimension * To whom correspondence should be addressed. Phone: +8129-860-4597. Fax: +81-29-860-4832. E-mail: ARIGA.Katsuhiko@ nims.go.jp. † NIMS. ‡ NAIST. (1) Edidin, M Curr. Opin. Struct. Biol. 1997, 7, 528. (2) (a) Kusumi, A.; Sako, Y. Curr. Opin. Cell Biol. 1996, 8, 566. (b) Tomishige, M.; Sako, Y.; Kusumi, A. J. Cell Biol. 1998, 142, 989. (c) Simson, R.; Yang, B.; Moore, S. E.; Doherty, P.; Walsh, F. S.; Jacobson, K. A. Biophys. J. 1998, 74, 297.

and cannot cover a two-dimensional region. The use of dendrimer chemistry in this subject area appears fruitful because a dendritic structure can develop two- or three(3) (a) Fielding, C. J.; Fielding, P. E. J. Lipid Res. 1997, 38, 1503. (b) Scheiffele, P.; Roth, M. G.; Simons, K. EMBO J. 1997, 16, 5501. (c) Rietveid, A.; Simons, K. Biochim. Biophys. Acta 1998, 1376, 467. (d) Brown, D. A.; London, E. Annu. Rev. Cell. Dev. Biol. 1998, 14, 111. (e) Brown, R. E. J. Cell Sci. 1998, 111, 1. (f) Scheiffele, P.; Rietveld, A.; Wilk, T.; Simons, K. J. Biol. Chem. 1999, 274, 2038. (g) Mostov, K. E.; Verges, M.; Altschuler, Y. Curr. Opin. Cell Biol. 2000, 12, 483. (h) Janes, P. W.; Ley, S. C.; Magee, A. I.; Kabouridis, P. S. Semin. Immunol. 2000, 12, 23. (i) Nguyen, D. H.; Hildreth, J. E. K. J. Virol. 2000, 74, 3264. (j) Ikonen, E. Curr. Opin. Cell Biol. 2001, 13, 470. (k) van der Goot, F. G.; Harder, T. Semin. Immunol. 2001, 13, 89. (l) Dietrich, C.; Bagatolli, L. A.; Volovyk, Z. N.; Thompson, N. L.; Levi, M.; Jacobson, K.; Gratton, E. Biophys. J. 2001, 80, 1417. (m) Dietrich, C.; Volovyk, Z. N.; Levi, M.; Thompson, N. L.; Jacobson, K. Proc. Natl. Acad. Sci. U.S.A. 2001, 98, 10642. (n) Kobayashi, T.; Yamagishi-Hasegawa, A.; Kiyokawa, E. Semin. Cell Dev. Biol. 2001, 12, 173. (o) Dietrich, C.; Yang, B.; Fujiwara, T.; Kusumi, A.; Jacobson, K. Biophys. J. 2002, 82, 274. (p) Di Guglielmo, G. M.; Le Roy, C.; Goodfellow, A. F.; Wrana, J. L. Nat. Cell Biol. 2003, 5, 410. (q) Ehehalt, R.; Keller, P.; Haass, C.; Thiele, C.; Simons, K. J. Cell Biol. 2003, 160, 113. (r) Abrami, L.; Liu, S.; Cosson, P.; Leppla, S. H.; van der Goot, F. G. J. Cell Biol. 2003, 160, 321. (s) Binder, W. H.; Barragan, V.; Menger, F. M. Angew. Chem., Int. Ed. 2003, 42, 5802. (4) (a) Zana, R. J. Colloid Interface Sci. 2002, 248, 203. (b) Zana, R. Adv. Colloid Interface Sci. 2002, 9, 205. (c) Sumida, Y.; Masuyama, A.; Oki, T.; Kida, T.; Nakatsuji, Y.; Ikeda, I.; Nojima, M. Langmuir 1996, 12, 3986. (d) Torchilin, V. P. J. Controlled Release 2001, 73, 137. (e) Matti, V.; Saily, J.; Ryhanen, S. J.; Holopainen, J. M.; Borocci, S.; Mancini, G.; Kinnuen, P. K. J. Biophys. J. 2001, 81, 2135. (f) Li, F.; Rosen, M. J.; Sulthana, S. B. Langmuir 2001, 17, 1037. (g) Li, Z. X.; Dong, C. C.; Wang, J. B.; Thomas, R. K. Langmuir 2002, 18, 6614. (h) Seredyuk, V.; Alami, E.; Nyden, M.; Holmberg, K.; Peresypkin, A. V.; Menger, F. M. Colloids Surf., A 2002, 203, 245. (i) Kirby, A. J.; Camilleri, P.; Engberts, J. B. F. N.; Feiters, M. C.; Nolte, R. J. M.; Soderman, O.; Bergsma, M.; Bell, P. C.; Fielden, M. L.; Rodriguez, C. L. G.; Guedat, P.; Kremer, A.; McGregor, C.; Perrin, C.; Ronsin, G.; van Eijk, M. C. P. Angew. Chem., Int. Ed. 2003, 42, 1448. (j) Brun, A.; Brezesinski, G.; Mo¨hwald, H.; Blanzat, M.; Perez, E. Rico-Lattes, I. Colloids Surf., A 2003, 228, 3.

10.1021/la0490238 CCC: $27.50 © 2004 American Chemical Society Published on Web 06/26/2004

Spider-Web Amphiphiles as Artificial Lipid Clusters

Langmuir, Vol. 20, No. 16, 2004 6763

dimensionally with controllable size and possibly shape. Dendrimers are also known to be capable of incorporating hydrophilic, hydrophobic, and inorganic guests.5-9 Amphiphilc dendrimers are regarded as unique building blocks of supramolecular assemblies and were subjected to monolayer research at the air-water interface.10-14 Some of them are synthesized by introducing hydrophobic tails to the outermost generation of a hydrophilic dendrimer or hydrophilic tails to a hydrophobic core. In other examples, a hydrophilic tail is attached to a hydrophobic dendrimer. Diblock designs with hydrophilic and hydrophobic cores and other designs were also proposed. However, the dendrimers used in these approaches lack the structural characteristics of a lipid cluster. If a (5) (a) Baars, M. W. P. L.; Meijer, E. W. Top. Curr. Chem. 2000, 210, 131. (b) Bosman, A. W.; Janssen, H. M.; Meijer, E. W. Chem. Rev. 1999, 99, 1665. (6) (a) Kukowska-Latallo, J. F.; Bielinska, A. U.; Johnson, J.; Spindler, R.; Tomalia, D. A.; Baker, J. R., Jr. Proc. Natl. Acad. Sci. U.S.A. 1996, 93, 4897. (b) Liu, M.; Kono, K.; Fre´chet, J. M. J. J. Polym. Sci., Part A: Polym. Chem. 1999, 37, 3492. (c) Bielinska, A. U.; Yen, A.; Wu, H. L.; Zahos, K. M.; Sun, R.; Weiner, N. D.; Baker, J. R., Jr.; Roessler, B. J. Biomaterials 2000, 21, 877. (d) Esfand, R.; Tomalia, D. A. Drug Discovery Today 2001, 6, 427. (e) Grinstaff, M. W. Chem.sEur. J. 2002, 8, 2839. (f) Stiriba, S.-E.; Frey, H.; Haag, R. Angew. Chem., Int. Ed. 2002, 41, 1329. (7) (a) Jansen, J. F. G. A.; de Brabander-van den Berg, E. M. M.; Meijer, E. W. Science 1994, 266, 1226. (b) Jansen, J. F. G. A.; Meijier, E. W. J. Am. Chem. Soc. 1995, 117, 4417. (c) Dykes, G. M.; Brierley, L. J.; Smith, D. K.; McGrail, P. T.; Seeley, G. J. Chem.sEur. J. 2001, 7, 4730. (8) (a) Ko¨hn, F.; Hofkens, J.; Wiesler, U.-M.; Cotlet, M.; van der Auweraer, M.; Mu¨llen, K.; De Schryver, F. C. Chem.sEur. J. 2001, 7, 4126. (b) Morgan, M. T.; Carnahan, M. A.; Immoos, C. E.; Ribeiro, A. A.; Finkelstein, S.; Lee, S. J.; Grinstaff, M. W. J. Am. Chem. Soc. 2003, 125, 15485. (9) (a) Niu, Y.; Yeung, L. K.; Crooks, R. M. J. Am. Chem. Soc. 2001, 123, 6840. (b) Crooks, R. M.; Zhao, M.; Sun, L.; Chechik, V.; Yeung, L. K. Acc. Chem. Res. 2001, 34, 181. (c) Yamamoto, K.; Higuchi, M.; Shiki, S.; Tsuruta, M.; Chiba, H. Nature 2002, 415, 509. (d) Ooe, M.; Murata, M.; Mizugaki, T.; Ebitani, K.; Kaneda, K. J. Am. Chem. Soc. 2004, 126, 1604. (10) (a) Fre´chet, J. M. Science 1994, 263, 1710. (b) Zeng, F.; Zimmerman, S. C. Chem. Rev. 1997, 97, 1681. (c) Tsukruk, V. V. Adv. Mater. 1998, 10, 253. (d) Grayson, S. M.; Fre´chet, J. M. Chem. Rev. 2001, 101, 3819. (e) Tully, D. C.; Fre´chet, J. M. Chem. Commun. 2001, 1229. (f) Morgan, J. R.; Cloninger, M. J. Curr. Opin. Drug Discovery Dev. 2002, 5, 966. (11) (a) Schenning, A. P. H. J.; Elissen-Roma´n, C.; Weener, J.-W.; Baars, M. W. P. L.; van der Gaast, S. J.; Meijer, E. W. J. Am. Chem. Soc. 1998, 120, 8199. (b) Cardullo, F.; Diederich, F.; Echegoyen, L.; Habicher, T.; Jayaraman, N.; Leblanc, R. M.; Stoddart, J. F.; Wang, S. Langmuir 1988, 4, 1955. (c) Kirton, G. F.; Brown, A. S.; Hawker, C. J.; Reynolds, P. A.; White, J. W. Physica B 1998, 248, 184. (d) Weener, J.-W.; Meijer, E. W. Adv. Mater. 2000, 12, 741. (e) Schenning, A. P. H. J.; Peeters, E.; Meijer, E. W. J. Am. Chem. Soc. 2000, 122, 4489. (f) Sui, G.; Micic, M.; Huo, Q.; Leblanc, R. M. Langmuir 2000, 16, 7847. (g) Sui, G.; Micic, M.; Huo, Q.; Leblanc, R. M. Colloids Surf., A 2000, 171, 185. (h) Felder, D.; Gallani, J.-L.; Guillon, D.; Heinrich, B.; Nicoud, J.-F.; Nierengarten, J.-F. Angew. Chem., Int. Ed. 2000, 39, 201. (i) Yoon, D. K.; Jung, H.-T. Langmuir 2003, 19, 1154. (12) (a) Aoi, K.; Motoda, A.; Okada, M.; Imae, T. Macromol. Rapid Commun. 1997, 18, 945. (b) Kampf, J. P. Frank, C. W.; Malmstro¨m, E. E.; Hawker, C. J. Science 1999, 283, 1730. (c) Kampf, J. P. Frank, C. W.; Malmstro¨m, E. E.; Hawker, C. J. Langmuir 1999, 15, 227. (d) Iyle, J.; Hammond, P. T. Langmuir 1999, 15, 1299. (d) Bo, Z.; Zhang, C.; Severin, N.; Rabe, J. P.; Schlu¨ter, A. D. Macromolecules 2000, 33, 2688. (13) (a) Maraval, V.; Laurent, R.; Donnadieu, B,; Mauzac, M.; Caminade, A. M.; Majoral, J. P. J. Am. Chem. Soc. 2000, 122, 2499. (b) Nierengarten, J.-F.; Eckert, J.-F.; Rio, Y.; del Pilar Carreon, M.; Gallani, J.-L.; Guillon, D. J. Am. Chem. Soc. 2001, 123, 9743. (14) (a) Saville, O. M.; Reynolds, P. A.; White, J. W.; Hawker, C. J.; Fre´chet, J. M. J.; Wooley, K. L.; Penfold, J.; Webster, J. R. P. J. Phys. Chem. 1995, 99, 8283. (b) Karthaus, O. Ijiro, K. Shimomura, M. Langmuir 1996, 12, 6714. (c) Tsukruk, V. V.; Rinderspacher, F.; Bliznyuk, V. N. Langmuir 1997, 13, 2171. (d) Mulders, S. J. E.; Brouwer, A. J.; Kimkes, P.; Sudho¨lter, E. J. R.; Liskamp, R. M. J. J. Chem. Soc., Perkin Trans. 2 1998, 1535. (e) Sheiko, S. S.; Buzin, A. I.; Muzafarov, A. M.; Rebrov, E. A.; Getmanova, E. V. Langmuir 1998, 14, 7468. (f) Ariga, K.; Urakawa, T.; Michiue, A.; Sasaki, Y.; Kikuchi, J. Langmuir 2000, 16, 9147. (g) Higuchi, M.; Shiki, S.; Ariga, K.; Yamamoto, K. J. Am. Chem. Soc. 2001, 123, 4414. (h) Oertel, U.; Appelhans, D.; Friedel, P.; Jehnichen, D.; Komber, H.; Pilch, B.; Ha¨nel, B.; Voit, B. Langmuir 2002, 18, 105.

Figure 1. Design of spider-web amphiphiles.

dendrimer were to have an amphiphilic structure at each unit, a more straightforward design for the artificial lipid clusters would be achieved in which desirable amphiphilic structures can be embedded in defined sequences. In this study, we developed this concept to provide a novel category of two-dimensional lipid clusters. For this purpose, amphiphilic units based on a Lys-Lys-Glu tripeptide with hydrophobic tails at the C-terminal and a polar head at the N-terminal are dendrically connected through stepwise peptide coupling (Figure 1). The obtained amphiphiles can two-dimensionally spread like a spider web. The design, synthesis, and monolayer behaviors of the “spider-web amphiphiles” are reported here. We also demonstrate the stoichiometric accommodation of lipid components in the core of the spider-web amphiphile. The obtained results would provide a significant contribution to practical applications such as the development of membrane-based drug delivery systems as well as the design of artificial lipid rafts for a fundamental understanding of biological phenomena at the cell surface. Experimental Section 1. Materials. Water used for the subphase was distilled in an Autostill WG220 (Yamato) and deionized by a Milli-Q lab water system (Millipore). Its specific resistance is more than 18 MΩ cm. Spectroscopic grade CHCl3, benzene, and ethanol (Wako Pure Chemicals) were used as the spreading solvents. Gold (99.999%)

6764

Langmuir, Vol. 20, No. 16, 2004

Ariga et al.

Figure 2. π-A isotherms of (a) G0, (b) G1, (c) G2, and (d) G3 amphiphiles: (A) AcC16 type at 20 °C; (B) N+C16 type at 20 °C; (C) AcC18 type at 20 °C; (D) AcC18 type at 10 °C. and chromium (99.99%) were used for the surface modification of the substrates. 2. Surface Pressure-Area (π-A) Isotherms. The π-A isotherms were measured using an FSD-300 computer-controlled film balance system (USI system). A mixture of CHCl3/benzene (80/20 v/v) or benzene/ethanol (80/20 v/v) was used as the spreading solvent. For the mixed monolayers, solutions of individual components were first prepared and then mixed at appropriate ratios to attain the desired mixing ratios of the amphiphile. Compression was started about 10 min after spreading at a rate of 0.2 mm s-1 (or 20 mm2 s-1 based on area). Fluctuation of the subphase temperature was within (0.2 °C. 3. Syntheses. See the Supporting Information.

Results and Discussion 1. Design and Syntheses. The synthetic reactions of the amphiphilic units and the spider-web amphiphiles are summarized in Schemes 1 and 2. The amphiphilic unit has a tripeptide moiety of Lys-Lys-Glu, which has two amino groups and one carboxyl group at their side chains along with a polar head at the N-terminal and hydrophobic tails at the C-terminal of the main chain. Branching and expansion of the web structure of the amphiphiles were achieved by the stepwise deprotection and coupling between the Lys and Glu residues on the basis of conventional peptide chemistry. All the spiderweb amphiphiles were synthesized in satisfactory yields, and characterized by 1H NMR, MALDI-TOFMS, GPC, and elemental analyses (see the Supporting Information). This synthetic strategy enables us to introduce an amphiphilic structure into each unit unlike the previously reported amphiphilic dendrimers where hydrophobic tails were at the outermost generation of a hydrophilic dendrimer or a hydrophilic tail was attached to a hydrophobic dendrimer. Therefore, the spider-web amphiphiles have structural characteristics of an amphiphile (lipid) cluster rather than polymeric bulk surfactants. Since a polar head and hydrophobic tails can be separately introduced into the unit structure, three series with different heads and tails were demonstrated as denoted by AcC16-GX, AcC18-GX, and N+C16-GX, where GX (X ) 0, 1, 2, and 3) represents the generation of the dendritic structures with one, two, six, and fourteen amphiphilic units, respectively (see Scheme 2). Furthermore, the introduction of different amphiphilic units can

possibly provide heterogenic amphiphile clusters in desirable sequences, which is now under development for artificial raft synthesis. 2. Fundamental Monolayer Characteristics. π-A isotherms of the spider-web amphiphiles on pure water are summarized in Figure 2. They form expanded monolayers up to their collapse pressure at 20 °C (Figure 2AC) except for AcC18-G2 (curve c in Figure 2C). A decrease in the subphase temperature to 10 °C induced a wellcondensed phase in the monolayers of the AcC18-type amphiphiles (Figure 2D), although AcC16-GX and N+C16GX still form expanded monolayers at the same temperature (data not shown). In the isotherm of the AcC18-G3 monolayer at 10 °C, a transition behavior is clearly observed. For the precise evaluation of the isotherms, the neighboring nine points in a π-A isotherm were fitted to the second-order equation using the Savitzky-Golay method,15 and the mathematically obtained slope of the isotherms was converted to compressibility ((-dA/dπ)/A).14f The molecular area at the most condensed state was calculated to be 4.9 nm2 from a point with the lowest compressibility. The limiting area obtained by extrapolation of a linear region of the condensed phase to 0 mN m-1 is 6.2 nm2. These values are rather close to the crosssectional area of the well-packed 28 alkyl chains (28 × 0.2 ) 5.6 nm2). Therefore, linkages through the peptide side chains would not seriously disturb the alkyl chain condensation in this molecule. Molecular areas in the expanded phase at given pressures were plotted as a function of the number of units (Figure 3). A linear relation was observed, suggesting that the structural restriction through a linkage is a minimum and all the amphiphilic units behave independently. The average molecular areas per chain at 30 mN m-1 were calculated to be 0.47, 0.41, and 0.49 nm2 for AcC16-GX, AcC18-GX, and N+C16-GX, respectively, at 20 °C. In these monolayers, alkyl chains are in a rather disordered state. 3. Accommodation of Lipid Components. Threedimensional dendrimers have the potential to encapsulate guest molecules on the basis of the density difference between the outer surface and inner core and/or using (15) Savitzky, A.; Golay, M. J. E. Anal. Chem. 1964, 36, 1627.

Spider-Web Amphiphiles as Artificial Lipid Clusters

Langmuir, Vol. 20, No. 16, 2004 6765

Scheme 1. Synthetic Scheme of Amphiphilic Unitsa

a Fmoc, Boc, and BOP represent 9-fluorenylmethyloxycarbonyl, tert-butoxycarbonyl, and benzotriazol-1-yloxytris(dimethylamino)phosphonium hexafluorophosphate, respectively.

dendrically immobilized binding sites. As shown above, all the amphiphilic units in the spider-web amphiphiles independently assemble upon compression, showing no influence on the structural restriction from the side chain linkage. The homogeneous expansion of the amphiphilic units suggests that the spider-web amphiphiles in their

expanded phase fit a filled-core model rather than a hollowcore model.16 However, the mixing behavior between the spider-web amphiphiles and other lipid components is a (16) (a) Pickett, G. T. Macromolecules 2002, 35, 1896. (b) Zook, T. C.; Pickett, G. T. Phys. Rev. Lett. 2003, 90, 015502.

6766

Langmuir, Vol. 20, No. 16, 2004

Ariga et al.

Scheme 2. Synthetic Scheme of Spider-Web Amphiphilesa

a

HBTU represents 2-(1H-benzotriazol-1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate.

subject worthy enough to be further investigated, because the accommodation of guest components by the spiderweb amphiphile may happen by a change in the alkyl chain packing. The encapsulation of the external lipid components into host assemblies can be simply evaluated through a dependency of the molecular area on the mixing ratio at the air-water interface. As the first example, the mixing behaviors between AcC16-GX and octadecanoic acid, which is a typical lipid component that forms an insoluble condensed phase, were investigated on pure water at 20 °C. A premixed solution [CHCl3/benzene (8/2 in v/v)] of AcC16-GX and octadecanoic acid at various mixing ratios (mol/mol) was spread on pure water and compressed. The π-A isotherms of the

AcC16-G1/octadecanoic acid, AcC16-G2/octadecanoic acid, and AcC16-G3/octadecanoic acid mixed monolayers are shown in Figure 4, where the abscissa represents the surface area per AcC16-GX molecule. In these plots, the octadecanoic acid molecules located inside the spider-web amphiphile do not cause an area increase, while positive shifts are induced by the area contribution of octadecanoic acid at the outside of the spider-web core. As easily recognized in Figure 4B,C, the addition of octadecanoic acid to the monolayers of AcC16-G2 and AcC16-G3 induced an increase in the collapse pressure, suggesting that the octadecanoic acid interacts with these spider-web amphiphiles to stabilize the two-dimensional assembly against three-dimensional collapse. In contrast,

Spider-Web Amphiphiles as Artificial Lipid Clusters

Langmuir, Vol. 20, No. 16, 2004 6767

Figure 3. Molecular areas in the expanded phase at (a) 10, (b) 20, and (c) 30 mN m-1 plotted against the number of units: (A) AcC16 type at 20 °C; (B) N+C16 type at 20 °C; (C) AcC18 type at 20 °C; (D) AcC16 type at 10 °C.

Figure 5. Molecular areas plotted as a function of [octadecanoic acid]/[spider-web amphiphile], mol/mol, at (a) 30, (b) 40, and (c) 50 mN m-1: (A) AcC16-G1; (B) AcC16-G2; (C) AcC16-G3. Panel D shows magnified plots of (c) in panel C.

Figure 4. π-A isotherms of mixed monolayers of the spiderweb amphiphiles and octadecanoic acid at 20 °C: (A) AcC16-G1 at a mixing ratio ([octadecanoic acid]/[AcC16-G1], mol/mol) of 0, 0.25, 0.50, 1.0, 2.0, 4.0, 8.0, and 10 (from the left); (B) AcC16G2 at a mixing ratio ([octadecanoic acid]/[AcC16-G2], mol/mol) of 0, 1.5, 3.0, 6.0, 9.0, 12, 24, and 36 (from the left); (C) AcC16G3 at a mixing ratio ([octadecanoic acid]/[AcC16-G3], mol/mol) of 0, 1.0, 2.0, 5.0, 7.0, 10, 14, 20, 30, 40, and 60 (from the left). Abscissas are based on the molecular area per spider-web amphiphile.

an increase in the collapse pressure was hardly detected for mixed monolayers of AcC16-G1 even by mixing with octadecanoic acid. Isotherms with a steep slope and sharp collapse, indicating the formation of a phase-separated condensed phase of octadecanoic acid, were observed at high octadecanoic acid contents (Figure 4A). The latter observation implies the less favorable interaction of these species. For a more quantitative evaluation of the two-dimensional mixing, the surface areas per spider-web amphiphile are plotted versus the ratio of octadecanoic acid to the amphiphile (Figure 5). The mixed monolayer of AcC16G1 and octadecanoic acid shows a linear relation (Figure 5A), indicating that octadecanoic acid mainly locates at the outside of AcC16-G1 and their areas are simply added. The slopes of these plots are ca. 0.2 nm2 per octadecanoic acid, which also confirms that the area increase originated from the addition of the area of the separated octadecanoic acid. In sharp contrast, a clear inflection point can be observed in plots for the mixed monolayer of AcC16-G3 and octadecanoic acid at all of the surface pressures we examined (Figure 5C). The surface areas per AcC16-G3 molecule remain unchanged (or slightly decreased) up to the critical ratio. The condensation effect observed in these mixing ratios suggests that the presence of octadecanoic acid on the surface area does not contribute to the total surface area probably due to the inclusion of the octadecanoic acid in the AcC16-G3 core. As seen in the magnified plot (Figure 5D for data at 50 mN m-1), the inflection

6768

Langmuir, Vol. 20, No. 16, 2004

Ariga et al. Table 1. Calculated Molecular Areas for the Spider-Web Amphiphile (Host) and Octadecanoic Acid (Guest) host AcC16G1 AcC16G2 AcC16G3

π (mN m-1)

A0a (nm2)

A1a (nm2)

A2a (nm2)

guest/host ratiob

30 30 40 30 40 50

2.112 5.730 4.872 12.857 11.193 9.369

2.105 5.306 4.403 11.809 9.938 8.156

0.007 0.424 0.469 1.048 1.255 1.313

0.0 2.1 2.3 5.2 6.3 6.6

a The definitions of A , A , and A are described in the text. b The 0 1 2 number of octadecanoic acids included in one spider-web amphiphile upon assumption of a molecular area of 0.2 nm2 per octadecyl chain.

Figure 6. (A) π-A isotherms of mixed monolayers of AcC16G3 and octadecanol at 20 °C at a mixing ratio ([octadecanool]/ [AcC16-G3], mol/mol) of 0, 1.0, 2.0, 5.0, 7.0, 14, 20, 30, 40, and 60 (from the left). Abscissas are based on the molecular area per spider-web amphiphile. (B) Molecular areas plotted as a function of [octadecanol]/[AcC16-G3], mol/mol, at (a) 30, (b) 40, and (c) 50 mN m-1.

point was clearly detected at a ratio ([octadecanoic acid]/ [AcC16-G3]) of 5 or 6. In the case of the mixed monolayer of AcC16-G2 and octadecanoic acid, the inflection point was less clear but appeared at a smaller ratio. The mixed monolayers of AcC16-G3 and octadecanol were similarly investigated (Figure 6). Although the corresponding plots are less clear (Figure 6B), an inflection point can be detected. Interestingly, the inflection point appears at a ratio similar to that observed for the mixed monolayer of AcC16-G3 and octadecanoic acid. If the linear part of these plots is extrapolated to a ratio of [octadecanoic acid]/[spider-web amphiphile] of 0, we can estimate the molecular area occupied by the spiderweb amphiphile in the complex with octadecanoic acid. The areas obtained by extrapolation (A1) are smaller than those observed for only the spider-web amphiphiles (A0). For example, the extrapolated area for the mixed monolayer of AcC16-G3 and octadecanoic acid at 50 mN m-1 is 8.16 nm2, but the surface area of a single-component monolayer of AcC16-G3 was measured as 9.37 nm2. These results mean that molecules on water become more condensed upon mixing with octadecanoic acid. This area shrinkage of the spider-web amphiphile (A2 ) A0 - A1) is expected to supply spaces for accommodation of lipid guest molecules. The calculated results are summarized in Table 1, where A0, A1, and A2 represent the area of the singlecomponent monolayer, the area obtained by extrapolation, and the shrunken area (spaces for guest accommodation), respectively, as mentioned above. Although the area shrinkage upon the addition of octadecanoic acid was not observed for AcC16-G1, the spaces for accommodation of guests are apparently assigned to the other two monolayers. The A2 values for the AcC16-G2/octadecanoic acid and AcC16-G3/octadecanoic acid monolayers are ca. 0.4 and 1.2 nm2, respectively, as seen in Table 1. Therefore, one AcC16-G2 molecule can accommodate two octadecanoic acid molecules, and six molecules of octadecanoic acid are incorporated into the AcC16-G3 core. Images of the capture of octadecanoic acid by the spiderweb amphiphile can be speculatively illustrated in Figure

Figure 7. Plausible models of the two-dimensional capture of lipidic guests by spider-web amphiphiles.

7. Interestingly, the number of accommodated guests is comparable to the number of branching points in one spider-web molecule: 2 for AcC16-G2 and 2 + 4 for AcC16G3. At the branching points of the spider-web amphiphile, the alkyl chains are rather crowded and might cause condensation through positive interaction between the fatty chains. Such a structural characteristic might be related to the stoichiometric capture of the lipidic component by the spider-web amphiphile. However, the detailed mechanism of the lipid accommodation is still not clear and has to be further investigated.

Spider-Web Amphiphiles as Artificial Lipid Clusters

Conclusion In this study, we designed and synthesized spider-web amphiphiles in which the amphiphile units are linked in a two-dimensional network. The formed structure can be regarded as a cluster of amphiphiles, which would be useful for designing artificial lipid rafts. We also demonstrated that the spider-web amphiphiles can accommodate lipid components possibly according to the stoichiometric rule where the number of accommodating guest lipids is consistent with the number of branching points in the spider-web amphiphile. The synthesized amphiphile might be useful for the development of a membrane-based delivery system for lipophilic drugs, which are usually hard to deliver due to their insolubility in water. The design

Langmuir, Vol. 20, No. 16, 2004 6769

proposed for the spider-web amphiphile also allows us to construct amphiphile units with different polar heads and hydrophobic tails in a linked cluster with a desirable connection sequence. Therefore, we can possibly create a more sophisticated design of lipid clusters where regulated flow of energy, electrons, and information would be realized as seen in actual cell membranes. Supporting Information Available: Detailed information on the synthetic procedures and structural characteristics of the spider-web amphiphiles. This material is available free of charge via the Internet at http://pubs.acs.org. LA0490238