Adhesion of Antibody-Functionalized Polymersomes - ACS Publications

polymersomes with an anti-ICAM-1 antibody, using modular biotin-avidin chemistry. ... an ICAM-1-coated polystyrene microsphere at various surface dens...
0 downloads 0 Views 195KB Size
Download PDF
Langmuir 2006, 22, 3975-3979

3975

Adhesion of Antibody-Functionalized Polymersomes John J. Lin,†,‡ P. Peter Ghoroghchian,‡,§ Ying Zhang,†,‡ and Daniel A. Hammer*,†,‡,§ Department of Chemical and Biomolecular Engineering, Institute for Medicine and Engineering, and Department of Bioengineering, UniVersity of PennsylVania, Philadelphia, PA, 19104 ReceiVed September 7, 2005. In Final Form: January 15, 2006 Polymersomes are vesicles made from synthetic block copolymers. The adhesiveness of micron-sized polymersomes, functionalized with antibodies that bind to vascular cell adhesion molecules, which could be useful for vascular targeting, was measured. Intercellular adhesion molecule-1 (ICAM-1) is an endothelial cell adhesion molecule whose expression increases during inflammatory disease, and is therefore a natural target for vascular delivery. We functionalized polymersomes with an anti-ICAM-1 antibody, using modular biotin-avidin chemistry. Micropipet aspiration was used to confirm specific adhesion and measure the adhesion strength between an anti-ICAM-1-coated polymersome and an ICAM-1-coated polystyrene microsphere at various surface densities of adhesion molecules. The adhesion is kinetically trapped, and adhesion strength is quantified by the critical tension for detachment. The adhesion strength increases in proportion to the surface density of anti-ICAM-1 molecules, in contrast to results seen previously when measuring adhesion between biotinylated vesicles and avidin-coated beads (Lin et al. Langmuir 2004, 20, 5493). The difference in dependence on the density of functional groups is likely due to the molecular presentation at the vesicle surface; in the current study, the presentation of biotinylated anti-ICAM-1 on a layer of avidin leads to the effective presentation of the anti-ICAM-1 and, thus, a monotonic increase in adhesiveness with antibody density.

Introduction In aqueous solutions, synthetic amphiphilic block copolymers of the appropriate composition spontaneously self-assemble into polymersomes1 (polymer vesicles) with sizes ranging from tens of nanometers to tens of microns in diameter. Depending on the phase behavior of the polymer, the polymersome bilayer membranes are fluid, similar to the phospholipid membrane of cells, and provide the opportunity to use polymersomes as model systems to investigate membrane-mediated processes. The mechanical properties of polymersome and phospholipid membranes, such as the area elastic and bending moduli, have been measured1-17 and compared2 and have shown that polymersomes in the fluid state can be designed to be 5-50 times tougher than * To whom correspondence should be addressed. Current address: Department of Bioengineering, University of Pennsylvania, 120 Hayden Hall, 3320 Smith Walk, Philadelphia, PA 19104. Phone: 215-573-6761. Fax: 215-573-2071. E-mail: [email protected]. † Department of Chemical and Biomolecular Engineering. ‡ Institute for Medicine and Engineering. § Department of Bioengineering. (1) Hammer, D. A.; Discher, D. E. Annu. ReV. Mater. Res. 2001, 31, 387-404. (2) Discher, B. M.; Won, Y. Y.; Ege, D. S.; Lee, J. C.; Bates, F. S.; Discher, D. E.; Hammer, D. A. Science 1999, 284, 1143-1146. (3) Lee, J. C.; Bermudez, H.; Discher, B. M.; Sheehan, M. A.; Won, Y. Y.; Bates, F. S.; Discher, D. E. Biotechnol. Bioeng. 2001, 73, 135-145. (4) Evans, E.; Yeung, A. Chem. Phys. Lipids 1994, 73, 39-56. (5) Evans, E.; Needham, D. J. Phys. Chem. 1987, 91, 4219-4228. (6) Bo, L.; Waugh, R. E. Biophys. J. 1989, 55, 509-517. (7) Bozic, B.; Svetina, S.; Zeks, B.; Waugh, R. E. Biophys. J. 1992, 61, 963973. (8) Waugh, R. E.; Song, J.; Svetina, S.; Zeks, B. Biophys. J. 1992, 61, 974982. (9) Heinrich, V.; Waugh, R. E. Ann. Biomed. Eng. 1996, 24, 595-605. (10) Evans, E.; Rawicz, W. Phys. ReV. Lett. 1997, 79, 2379-2382. (11) Rawicz, W.; Olbrich, K. C.; McIntosh, T.; Needham, D.; Evans, E. Biophys. J. 2000, 79, 328-339. (12) Evans, E.; Rawicz, W. Phys. ReV. Lett. 1990, 64, 2094-2097. (13) Zhelev, D. V.; Needham, D.; Hochmuth, R. M. Biophys. J. 1994, 67, 720-727. (14) Bermudez, H.; Brannan, A. K.; Hammer, D. A.; Bates, F. S.; Discher, D. E. Macromolecules 2002, 35, 8203-8208. (15) Bermudez, H.; Hammer, D. A.; Discher, D. E. Langmuir 2004, 20, 540543. (16) Lin, J. J.; Bates, F. S.; Hammer, D. A.; Silas, J. A. Phys. ReV. Lett. 2005, 95, 026101. (17) Ghoroghchian, P. P.; Lin, J. J.; Brannan, A. K.; Bates, F. S.; Therien, M. J.; Hammer, D. A. AdV. Mater., submitted for publication.

liposomes (vesicles comprised of natural phospholipids);2 the areal strain achievable by a polymersome is an order of magnitude higher than that of a liposome made of stearoyl-oleoylphosphatidylcholine (SOPC) and depends on the molecular weight of the polymers that comprise the membrane. When compared to liposomes, polymersomes have superior and tunable material properties for storage and stability3 that may prove advantageous in their employment as intravascular drug-delivery vehicles. To this end, the engineering of targeted adhesion is a critical issue that must be addressed. Micropipet aspiration is a useful technique to study the adhesion of bilayer membranes.18-26 The adhesion of red blood cells and liposomes via receptor-ligand interactions have been measured using micropipet aspiration.20-24 We showed previously that biotinylated polymersomes form kinetically trapped crossbridges to superavidin-coated microspheres;25 biotin-avidin bonds impart little tension as they form, but require significantly more tension to break. In the study conducted by Lin et al.,25 biotin was directly coupled to the end terminal hydroxide on a poly(ethylene oxide)terminated block copolymer. The length of the biotin-containing diblock, the length of the surrounding nonfunctionalized diblock, and the density of the functionalized polymer all had an effect on adhesion strength.25 The strongest adhesion was seen when the functionalized polymer was substantially longer than the surrounding nonfunctionalized polymer, in roughly a 1:1 ratio. This is likely because the functionalized polymer must be substantially longer than the surrounding polymer for ready access to the opposing surface and because functionalized polymers of the same length compete for access to opposing sites, thus limiting the effectiveness of adhesion at high functional densities. (18) Evans, E. A. Biophys. J. 1985, 48, 175-183. (19) Evans, E. A. Biophys. J. 1985, 48, 184-192. (20) Evans, E.; Berk, D.; Leung, A. Biophys. J. 1991, 59, 838-848. (21) Evans, E.; Berk, D.; Leung, A.; Mohandas, N. Biophys. J. 1991, 59, 849-860. (22) Berk, D.; Evans, E. Biophys. J. 1991, 59, 861-872. (23) Evans, E.; Leung, A. J. Cell Biol. 1984, 98, 1201-1208. (24) Noppl-Simson, D. A.; Needham, D. Biophys. J. 1996, 70, 1391-1401. (25) Lin, J. J.; Silas, J. A.; Bermudez, H.; Milam, V. T.; Bates, F. S.; Hammer, D. A. Langmuir 2004, 20, 5493-5500. (26) Evans, E.; Klingenberg D. J.; Rawicz, W.; Szoka, F. Langmuir 1996, 12, 3031-3037.

10.1021/la052445c CCC: $33.50 © 2006 American Chemical Society Published on Web 03/24/2006

3976 Langmuir, Vol. 22, No. 9, 2006

Letters

Table 1. Polymersome Formation Block Copolymers designation

block copolymer

AMWa(Da)

fEOb

OB-2 OB-18

PEO26-PBD46 PEO80-PEB125

3600 10 400

0.28 0.29

a Average molecular weight. b Volume fraction of poly(ethylene oxide).

In this paper, we used micropipet aspiration to measure the adhesion strength between an anti-intercellular adhesion molecule-1 (ICAM-1)-coated polymersome and ICAM-1-coated polystyrene microsphere. We focused on ICAM-1, since it is an endothelial cell surface molecule whose expression is increased during inflammatory disease; correspondingly, when leukocytes are homing to inflammatory sites, they employ a specific adhesion receptor, β2-integrins, to bind to sites of high ICAM-1 expression. The availability of ICAM-1 on the endothelium has been used to target phospholipid vesicles to inflammatory sites using antiICAM-1 antibodies.27 We previously used anti-ICAM-1 to functionalize colloidal particles and found these beads to specifically bind to ICAM-1 surfaces under flow.28 Eniola and co-workers showed that the kinetics of association between an anti-ICAM-1 (BBIG-1A) and ICAM-1 is similar to that of activated integrin and ICAM-1;28 therefore, this is the antiICAM-1 antibody we employ here. In our model system, polymersomes were functionalized with biotin, then labeled with avidin, and then labeled with a biotinylated BBIG-1A anti-ICAM-1 antibody. Biotin-avidin binding is a well-known modular chemistry for assembling biomolecules on surfaces. These functionalized polymersomes should be seen as the first step in the construction of “leukopolymersomes”, which are polymersomes with the adhesive or targeting properties of leukocytes. Micropipet aspiration was then used to disengage an anti-ICAM-1-coated polymersome from an ICAM-1-coated polystyrene microsphere. The adhesion measurements between anti-ICAM-1-coated polymersomes and ICAM-1-coated microspheres show that the adhesion strength scales linearly with the surface density of the anti-ICAM-1 molecules on the vesicle surface, in contrast to that shown previously when the adhesion between biotinylated vesicles and avidin beads was measured. The results suggest that the engineered adhesiveness of a polymer vesicle depends critically on the molecular presentation on the polymersome surface. Experimental Methods Poly(ethylene oxide)-polybutadiene block copolymers OB-2 (EO26-BD46, molecular weight (Mr) ) 3600 Da) and OB-18 (EO80BD125, Mr ) 10400 Da) (Table 1) were synthesized according to the protocol from Hillmyer and Bates.29 The protocol used to functionalize the surface of polymersome with anti-ICAM-1 molecules is as follows. The hydroxyl terminus of the block copolymer OB-18 is first labeled using tresyl chloride chemistry adopted from Nillson and Mosbach30-32 to create a leaving group for the subsequent addition of biocytin (biotin-lysine) to the hydrophilic end of the polymer OB-18; detailed experimental procedures of forming biocytin-coated polymersomes are described in the study of Lin et al.25 In brief, 500 µg of polymer, with a mass ratio of 20/80 of OB-2 and tresylated (27) Spragg, D. D.; Alford, D. R.; Greferath, R.; Larsen, C. E.; Lee, K. D.; Gurtner, G. C.; Cybulsky, M. I.; Tosi, P. F.; Nicolau, C.; Gimbrone, M. A. Proc. Natl. Acad. Sci. U.S.A. 1997, 94, 8795-8800. (28) Eniola, A. O.; Willcox, P. J.; Hammer, D. A. Biophys. J. 2003, 85, 37202731. (29) Hillmyer, M. A.; Bates, F. S. Macromolecules 1996, 29, 6994-7002. (30) Nilsson, K.; Mosbach, K. Methods Enzymol. 1984, 104, 56-69. (31) Delgado, C.; Patel, J. N.; Francis, G. E.; Fisher, D. Biotechnol. Appl. Biochem. 1990, 12, 119-128. (32) Delgado, C.; Francis, G. E.; Fisher, D. Crit. ReV. Ther. Drug Carrier Syst. 1992, 9, 249-304.

OB-18 polymer, is reacted with a 2:1 excess of biocytin in anhydrous methanol (Sigma-Aldrich, St. Louis, MO) overnight at 4 °C. A thin film of polymer was deposited on the bottom of a vial by evaporation, and polymersomes were formed by rehydration with 2 mL (300 mOsm) of sucrose solution (osmometer model 3300, Advance Instruments, Norwood, MA). Excess biocytin molecules in the external solution were removed by dialysis, using 10K Mr cutoff Slide-a-Lyzer dialysis cassettes (Pierce, Rockford, IL) before incubating biocytin-coated polymersomes with neutravidin (Pierce, Rockford, IL) to make neutravidin-coated polymersomes. The surface of the neutravidin-coated polymersome for the binding of biotinylated molecules was confirmed using Alexa Fluor 488-biocytin (Molecular Probes, Eugene, OR). The neutravidin-coated polymersomes, after excess neutravidin molecules were purged using an Amicon stirred ultrafiltration cell (Millipore, Bedford, MA), were then incubated with various mixtures of biotin-sialyl Lewisx (sLex) (GlycoTech, Gaithersburg, MD) and biotinylated anti-ICAM-1 BBIG-1A (R&D Systems, Minneapolis, MN) molecules. The density of anti-ICAM-1 molecules on the polymersome surface was tuned using various ratios of biotin-sLex and biotinylated anti-ICAM-1 BBIG-1A. Various ratios of biotin-sLex and biotinylated anti-ICAM-1 molecules were incubated with neutravidin-coated microspheres (Bangs Laboratory, Fisher, IN) to illustrate the feasibility of using biotin-sLex to tune the surface density of biotinylated anti-ICAM1; fluorescently labeled antibodies are used as markers for detection: fluorescein isothiocyanate (FITC)-conjugated rat anti-mouse IgG1 monocolonal antibody (PharMingen, San Diego, CA) was used to detect the IgG domain part of anti-ICAM-1, and FITC anti-human cutaneous lymphocyte antigen (CLA) (PharMingen, San Diego, CA) was used to detect sLex. ICAM-1-coated polystyrene microspheres were made by incubating recombinant human ICAM-1/Fc Chimera (R&D Systems, Minneapolis, MN) with polystyrene microspheres (Polysciences, Warrington, PA) overnight at 4 °C. Excess recombinant human ICAM-1/Fc Chimera molecules were removed by washing the microspheres with phosphate-buffered saline (PBS) solution via a centrifugation method in which microspheres are spun down into a pellet in a 1.5 mL centrifuge tube at 10 000 rpm for 3 min, followed by the removal of the supernatant and its replacement with fresh media. The process is repeated three times. Flow cytometry data confirm the coating of the polystyrene microsphere surface with approximately 4300 ICAM-1 molecules/µm2 using monoclonal ICAM-1 fluorescein (R&D Systems, Minneapolis, MN). Micropipet aspiration18-25 was used to measure the adhesion strength between polymersomes coated with various surface densities of anti-ICAM-1 molecules and ICAM-1-coated polystyrene microspheres. Adhesion experiments were performed inside a chamber made from microscope cover glasses, the same as those described in the study by Lin et al.25 The surfaces of the chamber were first coated with a 0.5% bovine serum albumin (BSA) solution, preventing polymersomes from nonspecifically adhering to the glass walls, before the microsphere/polymersome solution was introduced. A polymersome and a microsphere were aspirated using two micropipets that were mounted coaxially and facing each other. After the polymersome and microsphere pair was brought into contact, the tension on the polymersome was decreased gradually to allow the contact zone to form, and the surfaces were left undisturbed for 15 min to allow the polymersome surface ligand anti-ICAM-1 to bind to the complementary surface groups on the microsphere ICAM-1. After 15 min, the tension on the polymersome was increased gradually while one micropipet held the ICAM-1-coated microsphere statically. As the suction pressure increased, the polymersome was aspirated into the micropipet with increasing projected length. As the polymersome was initially aspirated, a small portion of the adhering membrane was peeled from contact with the microsphere with a small tension. After this initial decrease in the contact distance, the contact distance decreased slowly as the tension applied on the polymersome membrane increased, until the critical tension was reached. At the critical tension, all the anti-ICAM-1/ICAM-1 bonds connecting the polymersome and the microsphere are broken, and the contact distance decreases to a single contact point and breaks.

Letters

Langmuir, Vol. 22, No. 9, 2006 3977

Figure 1. Fluorescent confocal image of neutravidin-coated polymersomes and the flow cytometry measurement of anti-ICAM-1 on microsphere surfaces. (A) Fluorescent confocal image of neutravidin-coated polymersomes with Alexa Fluor 488-biocytin as a marker. The fluorescence from the edge of the polymersome indicates the neutravidin molecules on the surface of the polymersomes are readily accessible for binding biotinylated anti-ICAM-1 and biotin-sLex molecules. (B) Flow cytometry demonstrates that the surface density of anti-ICAM-1 at the surface of microspheres can be titrated down using biotin-sLex molecules. Each peak illustrates a different mixture of biotinylated anti-ICAM-1 and biotinsLex. From the most right peak to the left, the concentration mixtures (by mass) of biotinylated anti-ICAM-1 and biotin-sLex are 100% biotinylated anti-ICAM-1, 90% biotinylated anti-ICAM-1, 80% biotinylated anti-ICAM-1, 50% biotinylated anti-ICAM-1, 20% biotinylated anti-ICAM-1, and 100% biotin-sLex. The events were recorded using a digital cooled CCD camera (model Retiga Exi Fast 1394, QImaging, Burnaby, B.C., Canada).

Results and Discussion We are able to demonstrate that neutravidin-coated polymersomes are readily available for the binding of biotinylated molecules using Alexa Fluor 488-biocytin, as illustrated in Figure 1A. In Figure 1, the fluorescence from Alexa 488-labeled vesicles is imaged using confocal microscopy. Anti-ICAM-1-coated polymersomes were made by adding biotinylated anti-ICAM-1 molecules to neutravidin-coated polymersomes, and then removing the unbound antibody by dialysis. The surface density of anti-ICAM-1 is varied by mixing biotin-sLex with biotinylated anti-ICAM-1 at various ratios with neutravidin-coated polymersomes. Since the flow cytometry technique cannot be used

to successfully determine the density of molecules on the surface of the polymersomes (because of the inability to make a sufficient number of polymersomes for flow cytometry), to illustrate the feasibility of the protocol, various ratios of biotin-sLex and biotinylated anti-ICAM-1 molecules are incubated with neutravidin-coated microspheres and fluorescently labeled antibodies are used as markers for detection. FITC-conjugated rat antimouse IgG1 monocolonal antibody is used to detect the IgG domain part of anti-ICAM-1. On microspheres, flow cytometry data confirm that mixing various ratios of biotin-sLex with biotinylated anti-ICAM-1 molecules in solution before incubating the solution with neutravidin-coated microspheres produces various surface densities of anti-ICAM-1 on the microsphere surface (Figure 1B). Thus, we infer that we can systematically vary the amount of anti-ICAM-1 on the polymersome surface using ratios of biotinylated anti-ICAM-1 to biotin-sLex. Note, we cannot report the absolute amount of anti-ICAM-1 on the polymersome surface; rather, only the percentage of avidin sites occupied by biotinylated anti-ICAM-1. Adhesion strength between an anti-ICAM-1-labeled vesicle and an ICAM-1-coated bead is measured using micropipet aspiration. To prove that the bonds between anti-ICAM-1 and ICAM-1 formed within the contact area are responsible for holding the polymersome to the microsphere, a control experiment with a nonfunctionalized polymersome results in a contact area between the polymersome and microsphere that decreases rapidly with a negligible applied tension (data not shown). A similar result is observed if the adhesion experiment is repeated with an excess of soluble ICAM-1 molecules (data not shown), which demonstrates the adhesion is specifically due to anti-ICAM-1 and ICAM-1 binding. No adhesion was observed between neutravidincoated polymersomes with ICAM-1-coated microspheres. OB-2 and OB-18 block copolymers are used in this study; they both consist of poly(ethylene oxide)-polybutadiene with an ethylene oxide volume fraction of about 0.30. The molecular weight and the hydrophilic and hydrophobic portions of OB-18 (EO80-BD125, Mr ) 10400 Da) are larger and longer than that of OB-2 (EO26-BD46, Mr ) 3600 Da). The hydrophilic end of OB-18 is functionalized with biocytin molecules to create a molecule called OB-18b. We made vesicles with 55 mol % of OB-18b in OB-2, bound these with avidin, and then labeled them with anti-ICAM-1, as described in the Experimental Methods section. The critical tension needed to separate the polymersomes and microspheres is plotted against the molar percentage of anti-ICAM-1 on the surface of the polymersome in Figure 2B. The adhesion of anti-ICAM-1-coated polymersomes to ICAM-1-coated polystyrene microspheres is kinetically trapped.19 This system behaves differently from an equilibrium system; in kinetically trapped adhesion, the force needed to form the contact is negligible; however, the force required to subsequently separate the contact is large.19 Thus, the appropriate metric of the adhesion strength is critical tension, rather than adhesion energy. A video micrograph of the time progression of the experiment is shown in Figure 2A. In Figure 2, the critical tension increases linearly as the surface density of anti-ICAM-1 molecules increases. The adhesiveness derived from the anti-ICAM-1-labeled vesicles binding to ICAM-1 microspheres differs from the behavior of biotin-labeled vesicles to avidin-coated microspheres, as published by Lin et al.25 A comparison of these results is illustrated in Figure 3. Although the maximal level of adhesion is of the same order of magnitude, the dependence of adhesion on the functional density of the adhesion molecule is very different in the two presentations. In this paper, we have utilized avidin-

3978 Langmuir, Vol. 22, No. 9, 2006

Letters

Figure 3. Critical tension measurements of two different receptorligand pairs: biotin-avidin (open circle, dashed line) and antiICAM-1-ICAM-1 (closed circle, solid line). The discrepancy between the two critical tension curves results from the different ways of presenting the adhesion molecules. In the biotin-avidin system, the adhesion molecules, biocytin, are linked directly to the hydrophilic ends of the OB-18 polymer (OB-18b), and various compositions of biotinylated polymersomes are made by mixing OB-18b with OB-2 or OB-18 in different percentages. The surface topology, membrane composition, mechanical properties of the membrane, or surface density of biotin all play roles in the adhesion strength. However, anti-ICAM-1 molecules are presented on a neutravidin plateau in which the contributions from surface topology, membrane composition, and mechanical properties of the membrane are eliminated by the coating of neutravidin molecules onto the surface of biotinylated polymersomes. When neutravidin molecules are adsorbed onto the surface of the biotinylated polymersomes, the flexible polymer brush is buried beneath by neutravidin molecules; hence, the adhesion strength is affected only by surface density of anti-ICAM-1 molecules. This, depending on the topology, maximal adhesion is achieved at intermediate biotin compositions in biotinavidin adhesion where biotin is linked directly to the chains at high concentration and where anti-ICAM-1 is linked to the avidin. All points have error bars. Figure 2. Critical tension measurements of an anti-ICAM-1-coated polymersome to an ICAM-1-coated polystyrene microsphere. (A) The critical tension required to break all the anti-ICAM-1 and ICAM-1 bonds formed between a polymersome and a microsphere scales proportional to the surface density of anti-ICAM-1. The closed circles are anti-ICAM-1-coated polymersomes using 55 mol % OB-18b in OB-2 as the underlying membrane, and the star symbol represents the critical tension measured at 100% surface anti-ICAM-1 using 10 mol % OB-18b in OB-2 and 100 mol % OB-18b as the underlying membrane. All points have error bars. (B) Representative sequence of the critical tension experiment. The anti-ICAM-1-coated polymersome was first allowed to adhere to an ICAM-1-coated microsphere. The micropipet on the left-hand side held the microsphere statically, while the one on the right-hand side come in to just touch the polymersome. As the suction pressure on the polymersome increased, the contact distance between the polymersome and the microsphere decreased. The scale bar is 10 µm.

biotin modular chemistry to present anti-ICAM-1 molecules to bead surfaces; anti-ICAM-1 molecules are linked to the surface of polymersomes on a plateau of avidin, as illustrated in Figure 4. However, in our previous measurements of the adhesion between biotinylated vesicles and avidin-coated beads, the adhesion molecules, biocytin, are linked directly to the hydrophilic ends of the OB-18 polymer. The biotinylated polymersomes are made with various compositions of OB-18b with OB-2. As also illustrated in the Figure 4, anti-ICAM-1 molecules are linked to the surface of polymersomes on an avidin plateau to make antiICAM-1-coated polymersomes. This fundamental difference in molecular orientation is likely responsible for the different responses, as explained below. Our previous work on the adhesion of polymersomes through biotin-avidin interaction has shown that surface topology,

Figure 4. Schematic illustrations of biotinylated polymersome and anti-ICAM-1-coated polymersome. In this paper, the comparison of critical tension behaviors are drawn between biotinylated polymersomes made with OB-18b in OB-2 (functionalized polymers are surrounded by a shorter polymer brush, OB-2), as illustrated in the schematic on the left-hand side, and anti-ICAM-1-coated polymersomes. Avidin-biotin modular chemistry was used to present the anti-ICAM-1 molecules. As illustrated in the schematic on the righthand side, anti-ICAM-1 molecules are linked to the surface of polymersomes on an avidin plateau to make anti-ICAM-1-coated polymersomes.

membrane composition, and the surface density of biotin all play important roles in the adhesion of biotinylated polymersomes to a neutravidin-coated microsphere surface.25 When biotinylated polymersomes are made from OB-18b and pure OB-18, the critical tension increases with biotinylated polymer concentration up to 10 mol % of OB-18b, then reaches a plateau. However, distinct adhesion behavior is observed when using biotinylated poly-

Letters

mersomes of OB-18b and OB-2: the critical tension goes through a maximum near 55 mol % OB-18b, no plateau of adhesion is observed in the OB-18b/OB-2 system, and the critical tensions measured when OB-18b is mixed with OB-2 are larger than those measured when OB-18b is mixed with pure OB-18. In our previous paper,25 we reasoned that these results were due to the flexibility of the biotinylated polymer. When presented among polymers of the same length, OB-18b was not as effective as when it was much larger than the surrounding polymer. One can envision that the biotinylated polymer exists in a distributed brush which is most effective when most of the biotin is much larger than the surrounding polymer layer. In contrast, building an adhesive layer using a polymer brush, followed by avidin, and then followed by an antibody, creates an extended molecular layer that is much more effective at binding. Neutravidin, which is about 60 000 Da in molecular weight, is a large molecule. When neutravidin molecules are adsorbed onto the surface of the biotinylated polymersomes, the flexible polymer brush is buried beneath by neutravidin molecules. Hence, the outer surface of the polymersome represents a raised plateau, and the adhesion strength is affected only by surface density of anti-ICAM-1 molecules. Since we can argue that the adhesiveness is due to the avidin-antibody layer, and since the avidin is a large molecule, the amount of avidin, and therefore the net adhesiveness of the system, should not be a function of the amount of OB-18b, provided there is enough available to support an adhesive layer. To test this, biotinylated polymersomes made from either 10 mol % of OB-18b in OB-2 and/or pure OB-18b (in contrast to the 55 mol % OB-18b in OB-2) were used as the underlying membrane. Then the adhesion to ICAM-1-coated beads was measured. The average critical tension for polymersomes using 55 mol % OB-18b in OB-2 as the underlying membrane at 100% anti-ICAM-1 is 0.28 mN/m, and the measured critical tensions at 100% surface anti-ICAM-1 using 10 mol % OB-18b in OB-2 or 100 mol % OB-18b as the underlying membrane are both 0.28 mN/m (represented by the star symbol in Figure 2A). Thus, the adhesiveness does not depend on the amount of biotinylated polymer, but rather, the amount of avidin on the vesicle surface. The surface of a polymersome is successfully functionalized with anti-ICAM-1 molecules using biotin-avidin modular chemistry. Biotin-sLex molecules are used to titrate down the surface density of anti-ICAM-1 molecules. However, previous studies have shown that sLex, which was used in this study to modulate the surface density of anti-ICAM-1, can mediate the

Langmuir, Vol. 22, No. 9, 2006 3979

rolling of colloidal particles on selectin surfaces.33-38 Selectins are a second key molecule upregulated during inflammation. Previous studies have demonstrated the ability to reproduce the rolling and firm adhesion phenomena of the recruitment of leukocytes to endothelial cells during the inflammation process using microspheres that are surface-functionalized with antiICAM-1 and sLex molecules;28 hence, anti-ICAM-1 and sLexcoated polymersomes will be used to mimic a leukocyte in a future study. These “leukopolymersomes” should be ideally suited for targeting sites of inflammation and carrying drugs or optical agents to inflammatory sites.

Conclusions Our goal was to make anti-ICAM-1-coated polymersomes and measure the adhesion strength of anti-ICAM-1 and ICAM-1 bonds using micropipet aspiration. Polymersomes made from block copolymers were chosen as a model because of their superior material properties. The current study succeeded in modifying the polymersome surface with anti-ICAM-1 molecules and measuring adhesion strength between anti-ICAM-1-coated polymersomes and ICAM-1-coated polystyrene microspheres. The adhesion strength scales linearly with the surface density of antiICAM-1. This result is in contrast to the adhesion between biotinylated polymersomes and superavidin-coated microspheres25 because of differences in the molecular topology at the vesicle surface. Future work will involve examining the adhesion properties of anti-ICAM-1 and sLe x-coated polymersomes under flow conditions. Acknowledgment. Funding was provided by Penn NSFMERSC, NASA Material Science, and NIH EB-003567 to D.A.H. J.J.L. acknowledges support from the NASA Graduate Student Researchers Program (GSRP) Fellowship. P.P.G. acknowledges fellowship support from the NIH Medical Scientist Training Program and the Whitaker Foundation. LA052445C (33) Brunk, D. K.; Goetz, D. J.; Hammer, D. A. Biophys. J. 1996, 71, 29022907. (34) Rossiter, H.; Alon, R.; Kupper, T. S. Mol. Med. Today 1997, 3, 214-222. (35) Vestweber, D. J. Cell Biochem. 1996, 61, 585-591. (36) Brandley, B. K.; Kiso, M.; Abbas, S.; Nikrad, P.; Srivasatava, O.; Foxall, O.; Oda, Y.; Hasegawa, A. Glycobiology 1993, 3, 633-641. (37) Rodgers, S. D.; Camphausen, R. T.; Hammer, D. A. Biophys. J. 2000, 79, 694-706. (38) Greenberg, A. W.; Brunk, D. K.; Hammer, D. A. Biophys. J. 2000, 79, 2391-2402.