Noncovalent Cell Surface Engineering with Cationic Graft Copolymers John T. Wilson, Venkata R. Krishnamurthy, Wanxing Cui, Zheng Qu, and Elliot L. Chaikof* Departments of Biomedical Engineering and Surgery, Georgia Institute of Technology and Emory UniVersity, 101 Woodruff Circle, Suite 5105 WMRB, Atlanta, Georgia 30322 Received October 18, 2009; E-mail: echaiko@emory.edu Chemical approaches to cell surface engineering have emerged as powerful tools for resurfacing the molecular landscape of cells and tissues. 1 Introduction of exogenously derived molecules alongside native cell surface constituents affords opportunities to control biochemical and cellular responses, with important implica- tions for drug delivery, cell-based therapy, and tissue engineering. Most cell surface modification strategies utilize covalent chemistries that target native cell surface constituents (e.g., lysine residues) or noncanonical reactive moieties introduced by metabolic or genetic engineering approaches. 1,2 However, the perturbation of cellular physiology inherent to such strategies can interfere with important cellular functions governed by cell surface molecules. 1a,2a,3 Hence, a need exists for noncovalent cell surface modification strategies. Here we report a versatile and facile noncovalent approach to cell surface engineering achieved through electrostatic adsorption of poly(L-lysine)-graft-poly(ethylene glycol) (PLL-g-PEG) copoly- mers to cellular interfaces. Isolated pancreatic islets were used as a model system in these investigations owing to the widespread use of cell surface modification in improving the outcome of islet transplantation, a promising treatment for diabetes. 4 PLL-g-PEG copolymers were rendered cytocompatible through appropriate control of the grafting ratio, and used as ‘cell surface active’ polymeric carriers for ligands and reactive groups. Copolymers bearing terminally functionalized PEG grafts were used to display biotin, hydrazide, and azide moieties, which selectively captured streptavidin-, aldehyde-, and cyclooctyne-labeled probes, respec- tively, on the islet surface (Scheme 1). Additionally, coadsorption of polymers enabled simultaneous display of multiple functional groups on cell surfaces. The cytotoxicity inherent to most polycations poses a significant molecular hurdle to the assembly of electrostatically stabilized cell surface-supported monolayers. 5 Indeed, incubation of islets with poly(L- lysine) (PLL) significantly reduced islet viability (Figure 1a). Polycation toxicity is dependent, in part, on charge density, 5 and we therefore postulated that PLL cytotoxicity could be attenuated by grafting methoxy terminated tetra(ethylene glycol) chains (PEG 4 (OCH 3 )) to a critical number of lysine residues. PLL-g-PEG 4 (OCH 3 ) copolymers with different extents of backbone modification were synthesized via active ester coupling between N-hydroxysuccinimide (NHS)-function- alized PEG and ε-amino groups of PLL and copolymer toxicity was assessed. As shown in Figure 1a, polymer cytotoxicity decreased as the extent of PEG grafting increased, with modification of ∼40% of lysine residues resulting in a copolymer that did not adversely influence islet viability (referred to herein as PP-OCH 3 ). Hence, a critical threshold for PEG grafting exists, below which polycation cytotoxicity is considerable. While the adsorption of PLL-g-PEG polymers to abiotic surfaces has been extensively studied, 6 the behavior of this class of copolymers at viable cellular interfaces has not been rigorously investigated. Many polycations induce the formation of pores in the plasma membrane, a phenomenon that mediates cell death and enables transport of molecules, including the polycation itself, across the cell membrane. 5 Indeed, as shown in Figure 1b, FITC-labeled PLL in contact with islets translocated across cell membranes and into the cytoplasm of individual cells. Conversely, AlexaFluor488-labeled PP-OCH 3 was distributed in a pattern consistent with the extracellular architecture of isolated pancreatic islets (Figure 1c), indicating maintenance of cell membrane integrity upon adsorption of the copolymer to cell surfaces. An unexpected finding was that a PEG-free analogue of PP-OCH 3 , synthesized by acetylating ∼40% of lysine monomers, exerted significant toxicity and localized intracellularly (see Supporting Information (SI)). Collectively, these results suggest synergism between decreased charge density and PEG grafting in attenuating PLL cytotoxicity, likely through inhibiting membrane pore formation. Therefore, only PLL-g-PEG copolymers with a unique balance of grafted PEG chains and free lysine monomers adsorb to cell surfaces without compromising cell viability. Based on these findings, structurally comparable copolymers were synthesized with PEG 4 (biotin), PEG 4 (hydrazide), or PEG 12 (azide) grafts Scheme 1. Cell Surface Engineering with PLL-g-PEG Copolymers Figure 1. (a) Islet viability after exposure to PLL and PLL-g-PEG 4 (OCH 3 ) copolymers with variable degrees of backbone PEGylation. Confocal micrographs of islets after incubation with FITC-labeled PLL (b) and AF488-labeled PP-OCH 3 (c). Scale bars: 50 μm (left), 10 μm (right). Published on Web 12/04/2009 10.1021/ja908887v  2009 American Chemical Society 18228 9 J. AM. CHEM. SOC. 2009, 131, 18228–18229