562
Bioconjugate Chem. 1999, 10, 562−571
ARTICLES Grafting Protein Ligand Monolayers onto the Surface of Microparticles for Probing the Accessibility of Cell Surface Receptors Andreas Frey,* Barbara Meckelein, and M. Alexander Schmidt Institut fu¨r Infektiologie, Zentrum fu¨r Molekularbiologie der Entzu¨ndung (ZMBE), Westfa¨lische Wilhelms-Universita¨t Mu¨nster, von-Esmarch-Strasse 56, D-48129 Mu¨nster, Germany. Received October 16, 1998; Revised Manuscript Received April 28, 1999
Coupling of a specific ligand to vaccines or drugs can be a powerful aid to route these compounds to a certain target cell population. However, if the targeted receptor is buried in a glycocalyx, binding of the ligand may be sterically hindered or even abolished, especially when the ligand is attached to bulky payloads. The antigen-transporting M cells that cover the gut-associated lymphoid tissue have a less pronounced glycocalyx than neighboring enterocytes. Such architectural differences might provide a possibility for targeting micro- or nanoparticulate vaccines to the mucosal immune system. To investigate the influence of the glycocalyx on the accessibility of cell surface receptors, we developed a system where a monolayer of ligand molecules is coupled in spatially aligned manner onto the surface of microparticles. On the basis of fluorescent carboxylate-modified particles of 1 µm diameter, different synthetic strategies were tested. Particles were first modified to display aldehyde functions on their surface, then protein ligands were coupled via Schiff base formation. The performance of the particles was tested on cultured mouse fibroblasts using the B subunit of cholera toxin as ligand and the plasma membrane glycolipid ganglioside GM1 as receptor. Cholera toxin B subunit-coated microparticles generated by one of our synthetic pathways exhibited specific binding to fibroblasts which could be blocked with soluble cholera toxin B subunit. As particles as small as 50 nm and any proteinaceous ligand may be used, this system provides a versatile means for monitoring receptor accessibilities in vitro and in vivo.
INTRODUCTION
Ligands which specifically bind to certain cell types are deemed proverbial “magic bullets” for delivering pharmaceuticals to their desired site of action. The concept of targeting a compound to a given cell type is of particular interest for the development of mucosal vaccines. The chief task of a mucosal vaccine is to induce a secretory immunoglobulin A response at the mucosal surfaces and thus intercept microbial pathogens in the run-up of infection (Kraehenbuhl and Neutra, 1992). Such a secretory immunoglobulin A response is induced only when the antigen is delivered to the mucosal immune system via the organized mucosa-associated lymphoid tissue, such as the Peyer’s patches in the small intestine, the tonsils, and adenoids in the nasopharyngeal area or the appendix (Neutra et al., 1996b). At these sites, the antigen is taken up by M cells, a specialized epithelial cell type which is able to endocytose particulate antigens of up to 10 µm in diameter (Eldridge et al., 1989; Ermak et al., 1995) and translocate them to the underlying lymphoid tissue where an immune response is mounted (Kraehenbuhl and Neutra, 1992; Neutra et al., 1996b). M cells are therefore prime candidates for the targeted * To whom correspondence should be addressed. Phone: +49251-835-6477. Fax: +49-251-835-6467. E-mail:
[email protected].
delivery of vaccines to the mucosal immune system. A major obstacle along that line, however, is the apparent lack of specific receptors on the apical M cell surface. No universal M-cell specific surface molecule has been identified as yet, and only a few cytoplasmic M cell markers have been described so far (Jepson et al., 1992; Gebert et al., 1994; Rautenberg et al., 1996). Nonetheless, M cells differ from the neighboring epithelial lining cells. For example, M cells lack an organized brush border and, in the gut, they display a less pronounced glycocalyx than the adjacent enterocytes (Kato, 1990; Gebert and Bartels, 1995; Frey et al., 1996; Neutra et al., 1996a). Consequently, plasma membrane receptors may be more accessible on M cells than on enterocytes. This opens up the possibility to target a ubiquitous receptor selectively on M cells simply because it may be shielded by the glycocalyx on neighboring enterocytes. We provided the proof of principle for this concept in a previous study in which we could demonstrate that binding of the B 1 Abbreviations: CTB, cholera toxin B subunit; DMEM, Dulbecco’s modified Eagle’s medium; EDC, 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide; Hepes, 4-(2-hydroxyethyl)1-piperazineethanesulfonic acid; MES, 4-morpholinoethanesulfonic acid; OVA, ovalbumin; PBS, phosphate buffered saline (2.7 mM KCl, 1.5 mM KH2PO4, 136 mM NaCl, and 8.1 mM Na2HPO4); RT, room temperature; S-NHS, N-hydroxysulfonylsuccinimide; TEG-biotin, biotinyl-3,6,9-trioxaundecanediamine.
10.1021/bc980124p CCC: $18.00 © 1999 American Chemical Society Published on Web 06/10/1999
Grafting Protein Monolayers onto Microparticles
Figure 1. Influence of protein multilayers on probe size. Relative increase in total hydrodynamic diameter of a nano- or microparticulate probe caused by multilayers of an average protein (hydrodynamic diameter 7 nm) when compared to the hydrodynamic diameter of the same probe carrying a protein monolayer only.
subunit of cholera toxin (CTB)1 to the plasma membrane glycolipid GM1 in the small intestine becomes M-cell specific when CTB is offered as a particulate probe of approximately 30 nm diameter (Frey et al., 1996). Soluble CTB of 7 nm hydrodynamic diameter bound ubiquitously, and CTB-coated particles exceeding 1 µm in diameter did not bind at all to the small intestinal epithelium. Consequently, there must be a size window in which selective ligand-mediated targeting of intestinal M cells is possible without the need for a specific receptor on the M-cell surface. To define the exact size of this window, we were in demand of easy-to-detect, preferably fluorescent nanoparticulate probes on which a monolayer of proteinaceous ligands can be coupled covalently. The formation of a single ligand layer on the probe core is essential because the final probe size will be affected strongly by ligand multilayers or spikes consisting of ligand molecules which are coupled to each other (Figure 1). Currently, there are only few companies who offer derivatizable fluorescent nanoparticles of a well-defined size. Most of these products carry carboxylate groups on their surface intended for attaching proteins via their amino functions by use of carbodiimides (Molday et al., 1975; Illum and Jones, 1985). A direct coupling of protein ligands to carboxylate-modified microparticles in the presence of carbodiimide cross-linker, however, also enables protein molecules to react with each other, thereby generating the undesired multilayers. We were therefore seeking alternative strategies in which the particles are preactivated such that they will react with ligand molecules without the need for additional crosslinkers. In this report, we describe a system where a monolayer of proteinaceous ligands is coupled in a spatially aligned manner onto the surface of fluorescent microparticles. The resulting probes were shown to be versatile tools for assessing the accessibility of receptor molecules in complex biological settings. MATERIALS AND METHODS
Materials and Cell Lines. Red and green fluorescent, carboxylate-modified microparticles of 1.0 µm nominal diameter (FluoSpheres) were from Molecular Probes Europe (Leiden, The Netherlands). These microparticles
Bioconjugate Chem., Vol. 10, No. 4, 1999 563
consist of a polystyrene core grafted with poly(acrylic acid) and acrylic acid. They possess a specific surface load of 0.1286 mequiv of carboxylate/g (corresponding to 7.76 × 10-14 mequiv of carboxylate/particle). Cholera toxin B subunit (CTB) (low salt) was purchased from List Biological Laboratories (Campbell, CA; via Quadratech, Epsom, U.K.), avidin (g13.5 Units/mg) from Molecular Probes and chicken egg white ovalbumin (OVA) from Calbiochem-Novabiochem (Bad Soden, FRG). Biotinyl-3,6,9-trioxaundecanediamine (TEG-biotin) was obtained from Molecular Biosciences (Boulder, CO). The BALB/c 3T3 fibroblast cell line clone A 31 was obtained from the American Type Culture Collection (Rockville, MD). Defined bovine calf serum (iron-supplemented) was from Hyclone (Logan, UT), Dulbecco’s modified Eagle’s medium (DMEM), glutamine, penicillin/ streptomycin, and 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (Hepes) were from PAA Laboratories (Linz, Austria). Round glass coverslips (13 mm diameter) were obtained from Bellco Glass (Vineland, NJ; via Dunn Labortechnik, Asbach, FRG). Poly(vinyl formal) (Formvar 15/95) and 150 square mesh copper grids were purchased from Polysciences Europe (Eppelheim, FRG). Mowiol 4-88 was from Calbiochem-Novabiochem. FilterCount scintillation liquid was obtained from Canberra Packard (Dreieich, FRG), [14C]ethanolamine was from Moravek Biochemicals (Brea, CA; via Hartmann Analytic, Braunschweig, FRG), and [14C]biotin from Amersham Buchler (Braunschweig, FRG). Nitrocellulose filter membranes were from Schleicher & Schuell (Dassel, FRG). Micro BCA Protein Assay Reagent was from Pierce (Rockland, IL; via KMF, Sankt Augustin, FRG). Purpald and aminoacetaldehyde dimethylacetal were purchased from Aldrich (Deisenhofen, FRG). All other chemicals and solvents were from various sources in the highest quality commercially available. Activation of Carboxylate-Modified Microparticles (1). A total of 1.29 × 109 carboxylate-modified microparticles (corresponding to 0.1 µmol of carboxylate groups according to the specific surface load given by the supplier) were washed twice by suspending in 1.8 mL of H2O followed by two times 20 min centrifugation at 10000g and removal of the supernatant to less than 20 µL. The microparticles were suspended in 65 µL of H2O before 100 µL of 10 mM 4-morpholinoethanesulfonic acid (MES)-NaOH, pH 6.0, 15 µL of 0.1 M N-hydroxysulfonylsuccinimide (S-NHS) (Staros et al., 1986), and 20 µL of 0.1 M 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDC) were added. The suspension was sonicated for 30 s in a Sonorex RK 510S sonifier (Bandelin Electronic, Berlin, FRG) and allowed to react for 20 min at room temperature (RT). The activated microparticles (1) were immediately used for the subsequent derivatization reactions without further purification. Synthesis of Aldehyde-Derivatized Microparticles (2, 3). Method A: via Aminoacetaldehyde Dimethylacetal. To the suspension of activated microparticles (1) were added 200 µL of H2O, 500 µL of 0.1 M HepesNaOH, pH 7.4, and 100 µL of 1 M aminoacetaldehyde dimethylacetal. The suspension was incubated for 16 h at RT in the dark on an end-over-end mixer rotating with 15 rpm. The particles (2a) were washed four times with 1.8 mL of H2O each and resuspended in 1 mL of 1 M HCl. In some experiments, 0.1 M citric acid or 3.25 M TFA was used instead of 1 M HCl. The suspension was incubated for 4 h under rotation as described above, with 2 × 30 s sonication every 30 min. The aldehyde-deriva-
564 Bioconjugate Chem., Vol. 10, No. 4, 1999
Frey et al.
tized particles (2) were washed three times with 1.8 mL of H2O and once with 1.8 mL of 50 mM Hepes-NaOH, pH 7.4. Method B: via Ethylenediamine-Glutaraldehyde Bridge. To the suspension of activated microparticles (1) were added 200 µL of H2O, 500 µL of 0.1 M Hepes-NaOH, pH 7.4, and 100 µL of 1 M ethylenediamine-HCl, pH 8. The suspension was incubated for 16 h under continuous rotation as described above. The particles (3a) were washed four times with 1.8 mL of H2O each, suspended in 1 mL of 50 mM Hepes-NaOH, pH 7.4, and 0.1 M glutaraldehyde, sonicated 2 × 30 s, and rotated as described above for 4 h. The aldehyde-derivatized particles (3) were washed four times with 1.8 mL of H2O and used immediately for subsequent coupling reactions. Synthesis of Biotin-Derivatized Microparticles (4). To the suspension of activated microparticles (1) were added 275 µL of H2O, 500 µL of 0.1 M Hepes-NaOH, pH 7.4, and 25 µL of 20 mM TEG-biotin. The suspension was incubated for 16 h under continuous rotation, and the biotinylated particles (4) were washed four times with 1.8 mL of H2O as described above. Protein Coupling to Aldehyde-Derivatized Microparticles. For the protein coupling reaction, 1.29 × 109 aldehyde-derivatized microparticles (2 or 3) were suspended in 500 µL of 0.1 M Hepes-NaOH, pH 7.4, and, in case of avidin coupling reactions, 100 µL of 5 M NaCl. To this suspension, 176 µL of 0.5 mg/mL (8.6 µM) CTB (1.51 nmol) or 88 µL of 1 mg/mL (22.2 µM) OVA (1.96 nmol) or 116 µL of 1 mg/mL (14.7 µM) avidin (1.70 nmol) were added, and the volume was adjusted to 1 mL with H2O. The suspension was rotated at 4 °C for 1 h, then 50 µL of 1 M NaCNBH3 was added, and the incubation was continued for 36 h under continuous rotation at 4 °C. The particles were washed three times with 1.8 mL of 10 mM Hepes-NaOH, pH 7.4, 0.2 M NaCl (avidinderivatized particles: once with 10 mM Hepes-NaOH, pH 7.4, 0.2 M NaCl, twice with 250 mM sodium carbonate, pH 9.5, 0.5 M NaCl), with at least 30 min rotation at 4 °C for each washing step. For blocking of the remaining free aldehyde functions, the particles were suspended in 1 mL of 50 mM Hepes-NaOH, pH 7.4, containing 0.1 M ethanolamine (or 0.1 M glycine) and 50 mM NaCNBH3, before they were incubated under continuous rotation at 4 °C for 16 h, washed three times with 1.8 mL of 10 mM Hepes-NaOH, pH 7.4, and stored in this buffer at 4 °C in the dark. Determination of Maximum Ligand Load. The maximum number of protein molecules that can be bound to one individual microparticle in a monolayer was estimated with a model which assumes tight packing of globular protein molecules on the surface of a spherical core (Frey et al., 1996). The maximum ligand load in a protein monolayer is then given by eq 1:
nmolecules )
(
2π dparticle +1 x3 dprotein
)
2
(1)
where dparticle is the diameter of the microparticle and dprotein is the hydrodynamic diameter of the ligand. To determine the hydrodynamic diameters of CTB, avidin, and OVA, the respective diffusion coefficients were measured by dynamic light scattering as described before (Frey et al., 1996) and the hydrodynamic diameters dh were calculated with the Stokes-Einstein eq 2:
dh )
2kT 6πηD
(2)
Table 1. Maximum Protein Load of 1 µm Microparticle
protein
diffusion coefficienta (10-13 m2/s)
diameterb (nm)
loadc (molecules/ particle)
CTB avidin ovalbumin
634 ( 22 662 ( 17 679 ( 6
7.4 7.0 6.4
71 000 80 000 92 000
a Diffusion coefficient at 24 °C in PBS as determined by dynamic light scattering. Data represent mean ( standard deviation of at least six independent experiments. b Hydrodynamic diameters were calculated from the mean diffusion coefficients using the Stokes-Einstein equation (see text). c The maximum number of protein molecules in a monolayer on the surface of one individual microparticle (diameter ) 1030 nm) was calculated using eq 1 (see text).
where k is the Boltzmann constant, T is the absolute temperature, η is the viscosity of water, and D the diffusion coefficient. The diameter of the microparticles (1030 nm ( 2.3%) was provided by the manufacturer. The maximum ligand load for the proteins on the microparticles used in this study is given in Table 1. Analytical Procedures. Electron Micoscopy. Carboxylate-modified microparticles were washed three times, suspended in H2O to a concentration of 5 × 104 particles/ µL, and sonicated for 30 s. Of these suspensions, 20 µL was placed on a Formvar-coated copper grid and allowed to settle for 90 min. Particles were contrasted with 4% (w/v) U(OAc)2 for 2 min and photographed at 38000× and 69000× magnification in a Philips 401 transmisson electron microscope (Philips Electron Optics, Eindhoven, The Netherlands). Determination of Coupling Yields. To determine the maximum number of derivatizable functional groups, 1.29 × 109 carboxylate-modified microparticles were activated with EDC and S-NHS to yield the active ester (1) as described above and incubated in 1 mL 50 mM Hepes-NaOH, pH 7.4, containing 2 mM/0.8 µCi/mL [14C]ethanolamine for 16 h at RT under continuous rotation. For quantitation of the product (5), aliquots of the reaction suspension were filtered through a glass frit equipped with a BA 85 nitrocellulose filter membrane (0.1 µm pore size) and washed extensively with 0.2 M NaCl under continuous suction. The filter membranes were airdried, transferred into scintillation vials, and counted in a Beckman LS 6000 scintillation counter (Beckman, Fullerton, CA) 2 h after addition of 5 mL of FilterCount scintillation liquid. The yields of acetal-derivatized (2a) and biotin-derivatized (4) particles were determined indirectly by coupling [14C]ethanolamine to residual unreacted carboxylate groups in the presence of EDC and S-NHS as described above. To determine the yield of aldehyde functionalization, the particles (2, 3) were incubated at RT under continuous rotation in 1 mL of 50 mM Hepes-NaOH, pH 7.4, containing 2 mM/0.8 µCi/mL [14C]ethanolamine. After 1 h of incubation, 2 µmol of NaCNBH3 was added, and the incubation was continued for 16 h under rotation. The amount of radiolabel bound to the particles was determined as described above. Determination of Protein Derivatization Yields. To determine the protein derivatization of microparticles in analytical assays, 1.29 × 109 microparticles either activated with EDC and S-NHS (1), or after aldehyde (2, 3) or biotin derivatization (4) were incubated in 1 mL of 50 mM Hepes-NaOH, pH 7.4, 0.5 M NaCl, and 5 µM (5 nmol) avidin for 36 h at 4 °C under rotation. The particles were washed once with 10 mM Hepes-NaOH, pH 7.4,
Grafting Protein Monolayers onto Microparticles
0.2 M NaCl, and twice with 250 mM sodium carbonate, pH 9.5, 0.5 M NaCl. The avidin-derivatized particles (6, 7, 8, or 9) were suspended in 1 mL of 50 mM HepesNaOH, pH 7.4, 5 µM/0.27 µCi/mL [14C]biotin and incubated, processed, and counted as described above. The protein load of particles derivatized with CTB, OVA, and avidin was also determined with the Micro BCA Protein Assay. The assay protocol provided by the supplier was modified to allow for determination of protein bound covalently to a particulate carrier (Stich, 1990). Suspensions of 1 × 109 particles in 500 µL of 10 mM Hepes-NaOH, pH 7.4, were mixed with 500 µL of Micro BCA working solution and incubated at 50 °C under continuous rotation (180 rpm) for 60 min. The suspensions were quickly cooled to 4 °C and centrifuged at this temperature for 10 min at 14000g. The supernatants were removed and their optical density at 562 nm was determined immediately. Protein solutions of known concentration were treated identically to the samples and used as standards. Acetaldehyde Dimethylacetal Cleavage and Purpald Assay. Acetaldehyde dimethylacetal (final concentration 2 mM) was incubated for 2 h at RT in the following solutions: 3.25 M TFA (pH
