Bioconjugate Chem. 2007, 18, 590−596
590
Protein-Liposome Conjugates Using Cysteine-Lipids And Native Chemical Ligation Sanne W. A. Reulen,†,| Wilco W. T. Brusselaars,†,| Sander Langereis,‡ Willem J. M. Mulder,§ Monica Breurken,† and Maarten Merkx*,† Laboratory of Macromolecular and Organic Chemistry, and Biomedical NMR, Department of Biomedical Engineering, Eindhoven University of Technology, and SyMO-Chem BV, P.O. Box 513, 5600 MB Eindhoven, The Netherlands. Received September 7, 2006; Revised Manuscript Received December 19, 2006
Liposomes have become popular drug delivery vehicles and have more recently also been applied as contrast agents for molecular imaging. Most current methods for functionalization of liposomes with targeting proteins rely on reactions of amine or thiol groups at the protein exterior, which generally result in nonspecific conjugation at multiple sites on the protein. In this study, we present native chemical ligation (NCL) as a general method to covalently couple recombinant proteins in a highly specific and chemoselective way to liposomes containing cysteine-functionalized phospholipids. A cysteine-functionalized phospholipid (Cys-PEG-DSPE) was prepared and shown to readily react with the MESNA thioester of EYFP, which was used as a model protein. Characterization of the EYFP-liposomes using fluorescence spectroscopy showed full retention of the fluorescent properties of conjugated EYFP and provides a lower limit of 120 proteins per liposome. The general applicability of NCL was further tested using CNA35, a collagen-binding protein recently applied in fluorescent imaging of collagen. NCL of CNA35 thioester yielded liposomes containing ∼100 copies of CNA35 per liposome. The CNA35-liposomes were shown to be fully functional and bind collagen with a 150-fold higher affinity compared to CNA35. Our results show that NCL is an attractive addition to existing conjugation methods that allows direct, covalent, and highly specific coupling of recombinant proteins to liposomes and other lipid-based assemblies.
INTRODUCTION Liposomes and micelles have attracted a lot of interest as drug delivery vehicles (1, 2) and more recently as carriers of contrast agents (MRI, ultrasound) in molecular imaging (3-5). An important development in these fields has been the introduction of “sterically stabilized” or “stealth” liposomes that have an increased circulation time and altered biodistributions (6). These liposomes typically contain ∼5% PEGylated phospholipids resulting in a liposome surface with low immunogenicity and increased liposome stability (7-9). For applications in molecular imaging and targeted drug delivery, the liposomes also need to be functionalized with targeting ligands, which in most cases are proteins (e.g., antibodies) or peptides. Most current synthetic strategies to covalently couple proteins to liposomes involve either amine or thiol groups at the exterior of the protein (10, 11). Although methods for direct coupling to liposomes have been reported (12), amine groups are commonly converted to thiol groups, which are subsequently reacted with phospholipids containing thiol reactive groups such as maleimides, activated disulfides, or iodoacetyl groups (2). All of these conjugation methods are nonspecific and result in conjugation at multiple sites on the protein, which sometimes also results in protein inactivation. Due to this lack of control over the conjugation site, protein conjugation is still, to some extent, a process of trial and error. An approach to circumvent the nonspecific modification of proteins is the use of so-called * Corresponding author. E-mail:
[email protected]. Fax: (+31) 40245-1036. † Laboratory of Macromolecular and Organic Chemistry, Department of Biomedical Engineering, Eindhoven University of Technology. ‡ SyMO-Chem BV. § Biomedical NMR, Department of Biomedical Engineering, Eindhoven University of Technology. | Both authors contributed equally to the work presented.
docking proteins as intermediates between the liposome and the targeting protein. Examples include the use of protein G as a docking site for IgG (13) and the use of RNase I as a docking site of proteins with a C-peptide tag (14-16). Although these approaches result in a more homogeneous presentation of targeting ligands on the liposomes, they require additional conjugation steps and depend on a noncovalent bond between targeting ligand and docking protein. To date, no general strategy is available that allows the direct covalent conjugation of liposomes to a single, precisely defined site on a recombinant protein. Native chemical ligation (NCL) was first reported by Dawson et al. as a unique method to ligate two unprotected peptide fragments to form a native peptide bond, thereby allowing the complete chemical synthesis of large proteins (17, 18). NCL is a chemoselective reaction that occurs spontaneously between a peptide with a C-terminal thioester and a peptide with an N-terminal cysteine under aqueous conditions at neutral pH. The possibilities of NCL have been extended by the development of expression systems that use self-cleavable intein domains to generate recombinant proteins with C-terminal thioester groups (19). NCL has been applied to attach synthetic moieties such as fluorescent dyes, biotin, isoprenyl groups, and dendrimers to recombinant proteins, and has also been used in the site-specific immobilization of proteins to surfaces (2023). Here, we report on the application of NCL as a method to covalently couple recombinantly expressed proteins in a highly specific and chemoselective way to liposomes containing cysteine-functionalized phospholipids. While this work was in progress, Bertozzi and co-workers reported the novel synthesis of cysteine-functionalized phospholipids for the lipidation of green fluorescent protein (GFP) and subsequent incorporation of the lipidated GFP in supported lipid bilayers (24). The focus of our work was to explore the suitability of NCL in the
10.1021/bc0602782 CCC: $37.00 © 2007 American Chemical Society Published on Web 02/22/2007
Technical Notes
synthesis of protein-liposome conjugates, however, demonstrating for the first time that protein thioesters can be directly conjugated to cysteine-functionalized phospholipids embedded in liposomes.
EXPERIMENTAL PROCEDURES General. Unless stated otherwise, all reagents and chemicals were obtained from commercial sources and used without further purification. Dichloromethane (DCM) was obtained by distillation from P2O5. 1,2-distearoyl-sn-glycero-3-phosphoethanolamine-N-[amino(poly(ethylene glycol))2000] (NH2-PEGDSPE), 1,2-distearoyl-sn-glycero-3-phosphocholine (DSPC), 1,2-distearoyl-sn-glycero-3-phosphoethanolamine-N-[methoxy(poly(ethylene glycol))2000] (PEG-DSPE), and 1,2-dipalmitoyl-sn-glycero-3-phosphoethanolamine-N-(lissamine rhodamine B sulfonyl) (rhodamine-DPPE) were purchased from Avanti Polar Lipids (Albaster, U.S.A.). Gd-DTPA-bis(stearylamide) was purchased from Gateway Chemical Technology (St. Louis, MO). Trityl-protected cysteine (Tr-Cys(Tr)-OH) was obtained from Bachem (Bubendorf, Switzerland). Trityl-protected succinimidyl-activated cysteine (Tr-Cys(Tr)-OSu) was prepared according to a literature procedure (20). UV-vis spectra were recorded on a Shimadzu Multispec 1501 spectrophotometer. Fluorescence spectra were obtained on an Edinburgh Instruments FS920 double-monochromator spectrophotometer. Primers used for all the cloning procedures were supplied by MWG (Ebersberg, Germany). Synthesis of Cysteine-Functionalized 1,2-Distearoyl-snglycero-3-phosphoethanolamine-N-[amino(poly(ethylene glycol))2000] (Cys-PEG-DSPE). NH2-PEG-DSPE (1) (100 mg, 35.8 µmol) was dissolved in DCM (1 mL) under an atmosphere of argon. Tr-Cys(Tr)-OSu (30 mg, 43 µmol) and triethylamine (10 µL, 71 µmol) were added to the solution. The reaction proceeded overnight at room temperature. The solution was concentrated under reduced pressure, and the crude product was dissolved in CHCl3 (0.5 mL). The crude product was purified by column chromatography (silica, CHCl3/MeOH, 19:1 v/v f 9:1 v/v), and trityl-protected Cys-PEG-DSPE (2) (44 mg, 13 µmol) was obtained in 36% yield. Compound 2 was dissolved in triethyl silane (25 µL, 158 mmol). Subsequently, a solution of trifluoroacetic acid (1 mL) and DCM (1 mL) was added. The obtained solution was stirred for 2 h at room temperature. The solution was concentrated under reduced pressure, and the crude product was precipitated in diethyl ether. The product was filtrated and dried under reduced pressure to give Cys-PEG-DSPE lipid (3) (34 mg, 12 µmol, 92%). Compounds 1-3 were analyzed in detail with RP-HPLC and MALDI-TOF (see Supporting Information). Plasmid Constructs. The EYFP gene was amplified by PCR from vector pEYFP-N1 (Clontech) using the primers 5′-GTG GTC ATA TGG TGA GCA AGG GCG AG-3′ and 5′-GTG GTG AAT TCC TTG TAC AGC TCG TCC ATG C-3′. The CNA35 gene was amplified from pQE30CNA35 (a kind gift from Dr. Magnus Ho¨o¨k, Texas A & M University, U.S.A. (25)) using the primers 5′-GTG GTC ATA TGG GAT CCG CAC GAG ATA TTT C-3′ and 5′-GTG GTT GCT CTT CCG CAT GCC TTG GTA TCT TTA TCC TGT TTT AAA AC-3′. The PCR products and the pTXB1 vector (IMPACT system, New England Biolabs) were double-digested with the restriction endonucleases Nde I and Sap I (CNA35) or Nde I and EcoR I (EYFP) followed by ligation of the amplified DNA fragments in the open plasmids to yield pTXB1-CNA35 and pTXB1EYFP, respectively. DNA sequencing using T7 promoter and intein-specific reversed primers (New England Biolabs) confirmed the correct in-frame fusion of the proteins with the intein sequence. Protein Expression and Purification. The expression plasmids pTXB1-CNA35 and pTXB1-EYFP were transformed
Bioconjugate Chem., Vol. 18, No. 2, 2007 591
into E. coli BL21 (DE3) cells. The same expression conditions were used for both proteins. Bacteria were grown in LB medium containing 100 µg/mL ampicillin at 37 °C and 250 rpm to an optical density (OD600nm) between 0.6 and 0.8. Protein expression was induced with 0.5 mM IPTG, and the cultures were incubated overnight at 15 °C. Cells were harvested by centrifugation for 30 min at 8000 g at 4 °C. The supernatant was removed, and the cell pellet was resuspended using the BugBuster protocol (Novagen). After incubation for 20 min at room temperature, the cell suspension was centrifuged at 16 000 g for 20 min at 4 °C. The supernatant was directly applied to a column of chitin beads and equilibrated with 10 column volumes of column buffer (20 mM sodium phosphate, 0.1 mM EDTA, 0.5 M NaCl, pH 8). The column was washed with 10 volumes of column buffer after which the column was quickly flushed with 3 column volumes of cleavage buffer (20 mM sodium phosphate, 0.1 mM EDTA, 0.5 M NaCl, pH 6) containing 50 mM sodium 2-mercaptoethanesulfonate (MESNA) and incubated overnight at room temperature. Elution fractions were collected and pooled, after which the cleavage step was repeated to gain more thioester-terminated proteins. The proteins were buffer-exchanged into 10 mM HEPES, 135 mM NaCl, pH 8.0 (HBS) using Amicon ultracentrifuge tubes (MWCO 10 kDa). The concentrations of EYFP protein with a C-terminal MESNA thioester (EYFP-MESNA) and CNA35 protein with a Cterminal MESNA thioester (CNA35-MESNA) were determined by UV-vis using 514nm ) 84 000 M-1 cm-1 (26) and 280nm ) 33 167 M-1 cm-1, respectively. 1 L E. coli culture typically yielded 20 mg of EYFP-MESNA and 40 mg of CNA35MESNA. Ligation of EYFP to Pure Cysteine-Functionalized Phospholipids. Cys-PEG-DSPE lipid 3 was dissolved in buffer containing 200 mM sodium phosphate, 200 mM NaCl, pH 7.2. Native chemical ligation reactions were performed in 200 µL with a final concentration of EYFP-COSR of 87 µM. A 10fold molar excess of compound 3 (final concentration 870 µM) was used. 1% (v/v) benzyl mercaptan and 1% (v/v) thiophenol, or 100 mM MESNA, were added to the ligation mixtures. After overnight incubation at room temperature, the samples were centrifuged to remove precipitate, and the supernatant was analyzed using SDS-PAGE. Liposome Preparation. Liposomes were prepared by lipid film hydration as described previously (27). A mixture of DSPC (37 µmol), Gd-DTPA-bis(stearylamide) (25 µmol), cholesterol (33 µmol), PEG-DSPE (2.5 µmol), rhodamine-DPPE (0.1 µmol), and Cys-PEG-DSPE (2.5 µmol) was dissolved in CHCl3/MeOH 1:1 (v/v) and concentrated under reduced pressure at room temperature. The obtained lipid film was hydrated in HBS buffer (4 mL). This dispersion was extruded five times at 65 °C through polycarbonate membrane filters with pores of 100 or 200 nm. Phospholipid concentrations were determined by phosphate analysis according to Rouser (28). The amount of lipids per liposome was calculated using a lipid surface area of 0.6 nm2 (29) and assuming unilamellar liposomes. Ligation of EYFP to Cysteine-Liposomes. Native chemical ligation of cysteine-liposomes (450 µM Cys-PEG-DSPE) with EYFP-MESNA (33 µM) was performed for 48 h at 20 °C in HBS, pH 8, containing either 100 mM MESNA or 1% (v/v) thiophenol and 1% (v/v) benzyl mercaptan. The coupling efficiency was monitored via SDS-PAGE analysis. The amount of lipidated EYFP was calculated by scanning of the SDSPAGE gel and integration of the bands of unreacted and reacted protein. The liposome concentration was calculated from the lipid concentration that was determined via phosphate analysis using the assumption of unilamellar liposomes. The number of EYFP proteins per liposome was then obtained by dividing the concentration of lipidated EYFP by the liposome concentration.
592 Bioconjugate Chem., Vol. 18, No. 2, 2007
The EYFP-liposomes were ultracentrifuged in a Kontron Centrikon T-2060 ultracentrifuge with a TFT 70.38 rotor for 1 h at 270 000 g and 20 °C, after which the liposomal pellet was resuspended in HBS. Pellet and supernatant were analyzed using SDS-PAGE to confirm the separation between reacted and unreacted protein. The incorporation of EYFP was also established using fluorescence spectroscopy. Emission spectra of EYFP-liposomes were measured using an excitation wavelength of 490 nm. The ratio of the EYFP (527 nm) and rhodamine (590 nm) peaks in the emission spectrum was compared to a calibration curve with known molar ratios of EYFP to rhodamine, thus providing an alternative method to determine the amount of EYFP per liposome. Ligation of CNA35 to Cysteine-Liposomes. Native chemical ligation of cysteine-liposomes (180 µM Cys-PEG-DSPE) with CNA35 thioester (50 µM) was performed for 24 h at 20 °C in HBS, pH 8, containing 100 mM MESNA. The amount of lipidated CNA35 was calculated by scanning the SDS-PAGE gel and integrating the bands of unreacted and reacted protein. The liposome concentration was calculated from the lipid concentration that was determined via phosphate analysis using the assumption of unilamellar liposomes. The amount of lipidated CNA35 was divided over the liposome concentration to determine the number of CNA35 proteins per liposome. Ultracentrifugation of CNA35-liposomes was performed in a Kontron Centrikon T-2060 ultracentrifuge with a TFT 70.38 rotor for 1 h at 270 000 g and 20 °C. The obtained liposomal pellet was resuspended in HBS. Pellet and supernatant were analyzed using SDS-PAGE to confirm the separation between reacted and unreacted protein. The concentration of CNA35 in the liposome fraction was determined with the Quant-iT Protein Assay Kit (Invitrogen) according to manufacturer instructions. This amount of lipidated CNA35 was divided by the liposome concentration to determine the number of CNA35 proteins per liposome. Collagen Binding Assay. 96 well Corning EIA/RIA microplates were coated overnight at 4 °C with 1.4 µg/well (45 µL) rat tail collagen type I (Sigma, C7661) in TBS (50 mM Tris, 150 mM NaCl, pH 7.5). After overnight incubation, the plates were blocked with 100 µL TBS containing 5% (w/v) skim milk powder for 2 h at room temperature. After washing the plates 3 times with 300 µL TBS, the plates were incubated with CNA35-liposomes or nonmodified liposomes in HBS supplemented with 5% (w/v) skim milk powder for 3 h at room temperature. Plates were washed 5 times with 50 mM Tris, 500 mM NaCl, pH 7.5, and subsequently washed 2 times with TBS. The fluorescence of the rhodamine-containing liposomes was measured at 620 nm in triplicate on a Thermo Fluoroskan Ascent FL plate reader after excitation at 578 nm.
RESULTS AND DISCUSSION In order to use native chemical ligation as a site-specific method to prepare protein-liposomes, the PEG terminus of PEG-DSPE was functionalized with a cysteine containing a free amine group. The introduction of the reactive group at the end of the PEG chain (far removed from the lipid tail) has been reported to enhance ligation efficiencies (30) and was also applied in previous work using maleimide-functionalized phospholipids (27, 31). Cys-PEG-DSPE (3) was obtained in a two-step synthesis starting from the commercially available 1,2-distearoyl-sn-glycero-3-phosphoethanolamine-N-[amino(poly(ethylene glycol))2000] (1) (Scheme 1). In the first step, the amine end group of 1 was reacted with succinimidyl-activated trityl-protected cysteine to yield intermediate 2 in 36% yield. The second step involved deprotection of the cysteine using trifluoroacetic acid (TFA) to give Cys-PEG-DSPE 3 in 92% yield, which was confirmed using RP-HPLC and MALDI-TOF (see Supporting Information).
Reulen et al. Scheme 1. Synthesis of Cys-PEG-DSPE (3)a
a (a) DCM, triethylamine, Tr-Cys(Tr)-OSu; (b) triethylsilane, TFA, DCM.
Scheme 2. Synthesis of Thioester-Terminated Proteins Using Self-Cleavable Intein Domains
Enhanced yellow fluorescent protein (EYFP) was chosen as a model protein to study the performance of Cys-PEG-DSPE in native chemical ligation reactions. The fluorescence of EYFP depends on proper folding of the protein and can thus be used to assess protein stability after ligation. In addition, fluorescence can be used to determine the amount of proteins conjugated per liposome. A C-terminal fusion protein of EYFP with intein and chitin binding domains was expressed in E. coli using the IMPACT system. The fusion protein was purified on a chitin resin and treated with 50 mM MESNA to induce the intein-catalyzed cleavage of the fusion protein and the formation of EYFP with a C-terminal MESNA thioester (EYFP-MESNA) (Scheme 2). SDS-PAGE and ESI-MS analysis (see Supporting Information) showed the presence of a single protein with a mass of 27 849 Da that corresponds to the calculated mass of EYFP-MESNA (theoretical mass: 27845.2 Da). We first studied the reaction of pure Cys-PEG-DSPE (not incorporated in a liposome) with EYFP-MESNA in the presence of two different thiol catalysts, thiophenol/benzyl mercaptan and MESNA. Since ligation of Cys-PEG-DSPE to EYFP increases its molecular weight by approximately 3 kDa, the ligation reaction could be monitored
Technical Notes
Bioconjugate Chem., Vol. 18, No. 2, 2007 593
Figure 1. Native chemical ligation of Cys-PEG-DSPE 3 (870 µM) with EYFP-MESNA (87 µM) in 0.2 M sodium phosphate pH 7.2 with either 100 mM MESNA or 1% (v/v) thiophenol + 1% benzyl mercaptan (THIOPHENOL).
Figure 2. Native chemical ligation of liposomes containing Cys-PEG-DSPE (450 µM cysteine-lipid) with EYFP-MESNA (33 µM) in HBS pH 8 with either 100 mM MESNA or 1% (v/v) thiophenol + 1% benzyl mercaptan (THIOPHENOL).
using SDS-PAGE. Nearly quantitative reaction of EYFP to the lipidated form of EYFP (EYFP-PEG-DSPE) was observed in the presence of thiophenol, whereas a much lower conversion (∼5%) was obtained using MESNA as the catalyst (Figure 1). Since PEGylated phospholipids such as Cys-PEG-DSPE are known to form relatively stable micelles (32), the ligation reaction probably resulted in the formation of EYFP micelles. We are currently characterizing the properties of these putative protein micelles. Having established the suitability of Cys-PEG-DSPE in native chemical ligation reactions, we next tested whether the same native chemical ligation can also be performed when CysPEG-DSPE is incorporated into liposomes. Liposomes containing 2.5% of Cys-PEG-DSPE, Gd-DTPA-bis(stearylamide) lipids, and rhodamine-functionalized lipids were prepared by lipid film hydration as described previously (27). We again compared thiophenol/benzyl mercaptan and MESNA as catalysts
and monitored the reaction using SDS-PAGE. Figure 2 clearly shows the presence of lipidated EYFP proteins upon reaction of 33 µM EYFP-thioester with 450 µM liposome-embedded Cys-PEG-DSPE. To the best of our knowledge, this is the first report that describes the direct coupling of proteins to liposomes via native chemical ligation. Although the ligation reaction is more efficient in the presence of thiophenol/benzyl mercaptan (conversion ≈ 30%) than in the presence of MESNA (conversion ≈ 10%), the use of MESNA has important advantages, in particular, when working with lipid-based systems. Thiophenol and benzyl mercaptan are poorly water soluble and are likely to accumulate in the phospholipid bilayer of liposomes, making it almost impossible to remove these toxic compounds after the ligation reaction. MESNA is a watersoluble thiol with a nonoffensive odor compared to thiophenol and benzyl mercaptan, and is easily removed after ligation via centrifugation.
594 Bioconjugate Chem., Vol. 18, No. 2, 2007
Reulen et al.
Figure 3. Emission spectrum of EYFP-liposome conjugates using an excitation wavelength of 490 nm (solid line). The emission spectra of EYFP-MESNA (dashed line) and nonmodified rhodamine-containing liposomes (dotted line) are also shown for comparison.
To prove that the lipidated EYFP observed in SDS-PAGE was still attached to the liposomes, liposomes were separated from unreacted EYFP via ultracentrifugation. SDS-PAGE analysis showed that all of the unreacted EYFP was present in the supernatant, whereas most of the lipidated EYFP was found associated with the liposomal pellet (see Supporting Information). Fluorescence spectroscopy was used to determine the number and proper folding of the conjugated EYFP proteins. The fluorescence emission spectrum of the EYFP-liposomes showed peaks at 527 and 590 nm, corresponding to EYFP and rhodamine, respectively (Figure 3). The observation of EYFP fluorescence not only demonstrates that the protein kept its native fold after ligation, but the relative intensity of the EYFP fluorescence compared to the rhodamine fluorescence can also be used to estimate the number of EYFP proteins per liposome. The ratio of EYFP to rhodamine was compared to a calibration curve with known molar ratios of EYFP to rhodamine, yielding approximately 120 proteins per 200 nm liposome. Another method to estimate the number of proteins per liposome is to calculate the amount of lipidated protein from the relative intensities of lipidated and non-lipidated protein bands on the SDS-PAGE gel obtained after NCL, and divide this number by the original liposome concentration. This calculation assumes that all lipidated protein is incorporated into liposomes and yielded 180 EYFP per 200 nm liposome. Although EYFP is a good model protein to prove that proteins can be conjugated to liposomes via native chemical ligation, we also wanted to use a biomedically relevant protein with a specific binding function. We recently developed a collagenspecific fluorescent probe based on a collagen-binding protein domain (CNA35) from the bacterial adhesion protein of Staphylococcus aureus (33). Solid-phase binding assays showed a dissociation constant of approximately 0.5 µM for collagen type I. In addition to providing an example of a targeted liposomal contrast agent, we were also interested to see whether attachment of multiple CNA35 proteins to a liposomal scaffold would yield a probe with a higher affinity for collagen due to multivalent interactions between CNA35 and collagen. The IMPACT expression vector pTXB1 was again used to obtain recombinant CNA35 with a C-terminal MESNA thioester (Scheme 2). An alanine residue was added to the C-terminus of CNA35 during cloning, as this residue is known to enhance the rate of native chemical ligation reactions (18). Affinity purification on a chitin column and subsequent cleavage with MESNA yielded CNA35-MESNA in excellent yield. SDSPAGE analysis typically showed the presence of several smaller bands besides the major band at 35 kDa, which are probably degradation products of CNA35 (Figure 4). ESI-MS showed a
Figure 4. SDS-PAGE analysis of CNA35-MESNA before and after native chemical ligation to liposomes containing Cys-PEG-DSPE.
Figure 5. Solid-phase binding assay of CNA35-liposomes to rat tail collagen type I (solid squares). Liposome binding was monitored by measuring the fluorescence of the rhodamine lipids at 620 nm using an excitation of 578 nm. Control experiments using nonmodified liposomes incubated on rat tail collagen type I (open triangle) and CNA35-liposomes incubated on milk-powder blocked well without collagen (open squares) are also shown for comparison. The solid line represents a fit to a 1:1 binding model using a Kd of 3 nM.
single protein peak with a mass of 34 757 Da (theoretical mass: 34774 Da). Collagen-binding assays using fluorescently labeled CNA35-MESNA showed that the modified C-terminus did not affect the collagen-binding properties. CNA35-MESNA was ligated to liposomes using the same procedure as described for EYFP using MESNA as a catalyst. SDS-PAGE analysis again indicated the formation of lipidated protein by the appearance of a new band at approximately 38 kDa (Figure 4). CNA35-modified liposomes were separated from MESNA and unreacted CNA35 by two ultracentrifugation steps (see Supporting Information). A protein quantification assay was used to determine the amount of protein in the liposome fraction. This protein concentration was divided by the known liposome concentration, yielding ∼100 proteins per 100 nm liposome. This number is in reasonable agreement with the amount of lipidated CNA35 that was detected by SDS-PAGE analysis immediately after ligation, which gave ∼80 CNA35 proteins per 100 nm liposome. The rhodamine fluorescence of the CNA35-functionalized liposomes was used to study their binding to a collagen-coated 96 well plate. Figure 5 shows specific binding of these liposomes to rat tail collagen type I at
Technical Notes
low nanomolar concentrations. Control experiments using nonfunctionalized liposomes or plates without collagen did not show a similar signal. A fit of the binding curve using a simple 1:1 binding model yielded a Kd of 3 ( 1 nM, representing a 150-fold increase in affinity compared to the protein itself. These results show that CNA35 conjugated to liposomes via native chemical ligation is fully active in binding collagen. Whether the observed increase in collagen affinity is due to multiple, simultaneous interactions between probe and collagen or mainly a statistical effect due to the presence of ∼100 copies of protein per liposome remains to be investigated. In summary, we have demonstrated that native chemical ligation is an attractive method to directly conjugate recombinantly expressed proteins to sterically stabilized poly(ethylene glycol) liposomes via cysteine-functionalized phospholipids. In contrast to other conjugation methods such as thiol-maleimide reactions, NCL is highly specific and exclusively occurs at the protein’s C-terminus, which in most proteins is not important for binding activity. An additional advantage of NCL is that only a single site in the protein is available for conjugation, whereas cross-linking is sometimes observed between liposomes and proteins using classical thiol/amine chemistries (34). The strategy reported here for liposomes should be readily applicable to other lipid-based imaging agents such as immunomicelles (35), iron oxide nanoparticles (31, 36), and quantum dots with a lipid coating (37-40).
ACKNOWLEDGMENT This study was funded by the BSIK program entitled Molecular Imaging of Ischemic Heart Disease (project number BSIK03033). The authors thank Anouk Dirksen for help with the synthesis of the cysteine-functionalized lipid, Marieke Rensen for constructing pTXB1-EYFP, Ingrid van Baal for supplying trityl-protected succinimidyl-activated cysteine, Erik Sanders for providing assistance with liposome preparation and characterization, and Bert Meijer, Klaas Nicolay, and Gustav Strijkers for general support. Supporting Information Available: Experimental details including the RP-HPLC and MALDI-TOF analysis of the synthesized phospholipids, ESI-MS spectra of EYFP-MESNA and CNA35-MESNA, and SDS-PAGE analysis of the purification of EYFP-MESNA, CNA35-MESNA, EYFP-liposomes, and CNA35-liposomes. This material is available free of charge via the Internet at http://pubs.acs.org/BC.
LITERATURE CITED (1) Allen, T. M., and Cullis, P. R. (2004) Drug delivery systems: entering the mainstream. Science 303, 1818-1822. (2) Torchilin, V. P. (2005) Recent advances with liposomes as pharmaceutical carriers. Nat. ReV. Drug DiscoVery 4, 145-160. (3) Massoud, T. F., and Gambhir, S. S. (2003) Molecular imaging in living subjects: seeing fundamental biological processes in a new light. Genes DeV. 17, 545-580. (4) Weissleder, R., and Mahmood, U. (2001) Molecular imaging. Radiology 219, 316-333. (5) Mulder, W. J., Strijkers, G. J., van Tilborg, G. A., Griffioen, A. W., and Nicolay, K. (2006) Lipid-based nanoparticles for contrastenhanced MRI and molecular imaging. NMR Biomed. 19, 142164. (6) Allen, T. M., and Hansen, C. (1991) Pharmacokinetics of stealth versus conventional liposomes: effect of dose. Biochim. Biophys. Acta 1068, 133-141. (7) Allen, T. M., Hansen, C., Martin, F., Redemann, C., and YauYoung, A. (1991) Liposomes containing synthetic lipid derivatives of poly(ethylene glycol) show prolonged circulation half-lives in vivo. Biochim. Biophys. Acta 1066, 29-36.
Bioconjugate Chem., Vol. 18, No. 2, 2007 595 (8) Klibanov, A. L., Maruyama, K., Torchilin, V. P., and Huang, L. (1990) Amphipathic polyethyleneglycols effectively prolong the circulation time of liposomes. FEBS Lett. 268, 235-237. (9) Papahadjopoulos, D., Allen, T. M., Gabizon, A., Mayhew, E., Matthay, K., Huang, S. K., Lee, K. D., Woodle, M. C., Lasic, D. D., Redemann, C., and Martin, F. J. (1991) Sterically stabilized liposomes: improvements in pharmacokinetics and antitumor therapeutic efficacy. Proc. Natl. Acad. Sci. U.S.A. 88, 11460-11464. (10) Nobs, L., Buchegger, F., Gurny, R., and Allemann, E. (2004) Current methods for attaching targeting ligands to liposomes and nanoparticles. J. Pharm. Sci. 93, 1980-1992. (11) Hermanson, G. T. (1995) Bioconjugate Techniques. pp 556-569, Chapter 12, Academic Press, Inc., San Diego. (12) Weissig, V., Lasch, J., Klibanov, A. L., and Torchilin, V. P. (1986) A new hydrophobic anchor for the attachment of proteins to liposomal membranes. FEBS Lett. 202, 86-90. (13) Chen, C. S., Baeumner, A. J., and Durst, R. A. (2005) Protein G-liposomal nanovesicles as universal reagents for immunoassays. Talanta 67, 205-211. (14) Backer, M. V., Patel, V., Jehning, B. T., and Backer, J. M. (2006) Self-assembled “dock and lock” system for linking payloads to targeting proteins. Bioconjugate Chem. 17, 912-919. (15) Backer, M. V., Gaynutdinov, T. I., Gorshkova, I. I., Crouch, R. J., Hu, T., Aloise, R., Arab, M., Przekop, K., and Backer, J. M. (2003) Humanized docking system for assembly of targeting drug delivery complexes. J. Controlled Release 89, 499-511. (16) Blankenberg, F. G., Mandl, S., Cao, Y. A., O’Connell-Rodwell, C., Contag, C., Mari, C., Gaynutdinov, T. I., Vanderheyden, J. L., Backer, M. V., and Backer, J. M. (2004) Tumor imaging using a standardized radiolabeled adapter protein docked to vascular endothelial growth factor. J. Nucl. Med. 45, 1373-1380. (17) Dawson, P. E., Muir, T. W., Clark-Lewis, I., and Kent, S. B. (1994) Synthesis of proteins by native chemical ligation. Science 266, 776-779. (18) Hackeng, T. M., Griffin, J. H., and Dawson, P. E. (1999) Protein synthesis by native chemical ligation: expanded scope by using straightforward methodology. Proc. Natl. Acad. Sci. U.S.A. 96, 10068-10073. (19) Muir, T. W., Sondhi, D., and Cole, P. A. (1998) Expressed protein ligation: a general method for protein engineering. Proc. Natl. Acad. Sci. U.S.A. 95, 6705-6710. (20) van Baal, I., Malda, H., Synowsky, S. A., van Dongen, J. L., Hackeng, T. M., Merkx, M., and Meijer, E. W. (2005) Multivalent peptide and protein dendrimers using native chemical ligation. Angew. Chem., Int. Ed. Engl. 44, 5052-5057. (21) Lesaicherre, M. L., Lue, R. Y., Chen, G. Y., Zhu, Q., and Yao, S. Q. (2002) Intein-mediated biotinylation of proteins and its application in a protein microarray. J. Am. Chem. Soc. 124, 87688769. (22) Lue, R. Y., Chen, G. Y., Hu, Y., Zhu, Q., and Yao, S. Q. (2004) Versatile protein biotinylation strategies for potential high-throughput proteomics. J. Am. Chem. Soc. 126, 1055-1062. (23) Gottlieb, D., Grunwald, C., Nowak, C., Kuhlmann, J., and Waldmann, H. (2006) Intein-mediated in vitro synthesis of lipidated Ras proteins. Chem. Commun. 260-262. (24) Grogan, M. J., Kaizuka, Y., Conrad, R. M., Groves, J. T., and Bertozzi, C. R. (2005) Synthesis of lipidated green fluorescent protein and its incorporation in supported lipid bilayers. J. Am. Chem. Soc. 127, 14383-14387. (25) Xu, Y., Rivas, J. M., Brown, E. L., Liang, X., and Ho¨o¨k, M. (2004) Virulence potential of the staphylococcal adhesin CNA in experimental arthritis is determined by its affinity for collagen. J. Infect. Dis. 189, 2323-2333. (26) Patterson, G., Day, R. N., and Piston, D. (2001) Fluorescent protein spectra. J. Cell Sci. 114, 837-838. (27) Mulder, W. J., Strijkers, G. J., Griffioen, A. W., van Bloois, L., Molema, G., Storm, G., Koning, G. A., and Nicolay, K. (2004) A liposomal system for contrast-enhanced magnetic resonance imaging of molecular targets. Bioconjugate Chem. 15, 799-806. (28) Rouser, G., Fkeischer, S., and Yamamoto, A. (1970) Two dimensional then layer chromatographic separation of polar lipids and determination of phospholipids by phosphorus analysis of spots. Lipids 5, 494-496.
596 Bioconjugate Chem., Vol. 18, No. 2, 2007 (29) Strijkers, G. J., Mulder, W. J., van Heeswijk, R. B., Frederik, P. M., Bomans, P., Magusin, P. C., and Nicolay, K. (2005) Relaxivity of liposomal paramagnetic MRI contrast agents. Magma 18, 186-192. (30) Hansen, C. B., Kao, G. Y., Moase, E. H., Zalipsky, S., and Allen, T. M. (1995) Attachment of antibodies to sterically stabilized liposomes: evaluation, comparison and optimization of coupling procedures. Biochim. Biophys. Acta. 1239, 133-144. (31) van Tilborg, G. A., Mulder, W. J., Deckers, N., Storm, G., Reutelingsperger, C. P., Strijkers, G. J., and Nicolay, K. (2006) Annexin A5-functionalized bimodal lipid-based contrast agents for the detection of apoptosis. Bioconjugate Chem. 17, 741-749. (32) Ashok, B., Arleth, L., Hjelm, R. P., Rubinstein, I., and Onyuksel, H. (2004) In vitro characterization of PEGylated phospholipid micelles for improved drug solubilization: effects of PEG chain length and PC incorporation. J. Pharm. Sci. 93, 2476-2487. (33) Krahn, K. N., Bouten, C. V., van Tuijl, S., van Zandvoort, M. A., and Merkx, M. (2006) Fluorescently labeled collagen binding proteins allow specific visualization of collagen in tissues and live cell culture. Anal. Biochem. 350, 177-185. (34) Takasaki, J., and Ansell, S. M. (2006) Micelles as intermediates in the preparation of protein-liposome conjugates. Bioconjugate Chem. 17, 438-450. (35) Torchilin, V. P., Lukyanov, A. N., Gao, Z., and PapahadjopoulosSternberg, B. (2003) Immunomicelles: targeted pharmaceutical
Reulen et al. carriers for poorly soluble drugs. Proc. Natl. Acad. Sci. U.S.A. 100, 6039-6044. (36) Nitin, N., LaConte, L. E., Zurkiya, O., Hu, X., and Bao, G. (2004) Functionalization and peptide-based delivery of magnetic nanoparticles as an intracellular MRI contrast agent. J. Biol. Inorg. Chem. 9, 706-712. (37) Michalet, X., Pinaud, F. F., Bentolila, L. A., Tsay, J. M., Doose, S., Li, J. J., Sundaresan, G., Wu, A. M., Gambhir, S. S., and Weiss, S. (2005) Quantum dots for live cells, in vivo imaging, and diagnostics. Science 307, 538-544. (38) Mulder, W. J., Koole, R., Brandwijk, R. J., Storm, G., Chin, P. T., Strijkers, G. J., de Mello Donega, C., Nicolay, K., and Griffioen, A. W. (2006) Quantum dots with a paramagnetic coating as a bimodal molecular imaging probe. Nano Lett. 6, 1-6. (39) Dubertret, B., Skourides, P., Norris, D. J., Noireaux, V., Brivanlou, A. H., and Libchaber, A. (2002) In vivo imaging of quantum dots encapsulated in phospholipid micelles. Science 298, 1759-1762. (40) van Tilborg, G. A., Mulder, W. J., Chin, P. T., Storm, G., Reutelingsperger, C. P., Nicolay, K., and Strijkers, G. J. (2006) Annexin A5-conjugated quantum dots with a paramagnetic lipidic coating for the multimodal detection of apoptotic cells. Bioconjugate Chem. 17, 865-868. BC0602782
