Reversible Control of Spacing in Charged Lamellar Membrane Hydrogels by Hydrophobically Mediated Tethering with Symmetric and Asymmetric Double-End-Anchored Poly(ethylene glycol)s

Abstract

Complex materials often achieve their remarkable functional properties by hierarchical assembly of building blocks via competing and/or synergistic interactions. Here, we describe the properties of new double-end-anchored poly(ethylene glycol)s (DEA-PEGs)-macromolecules designed to impart hydrophobically mediated tethering attractions between charged lipid membranes. We synthesized DEA-PEGs (MW 2000 (2K) and 4.6K) with two double-tail (symmetric) or a double-tail and a single-tail (asymmetric) hydrophobic end anchors and characterized their equilibrium and kinetic properties using small-angle X-ray scattering. Control multilayer membranes without and with PEG lipid (i.e., single-end-anchored PEG) swelled continuously, with the interlayer spacing increasing between 30 and 90 wt % water content due to electrostatic as well as, in the case of PEG lipid, steric repulsion. In contrast, interlayer spacings in lamellar membrane hydrogels containing DEA-PEGs expanded over a limited water dilution range and reached a "locked" state, which displayed a near constant membrane wall-to-wall spacing (δ_w) with further increases in water content. Remarkably, the locked state displays a simple relation to the PEG radius of gyration δ_w ≈ 1.6 R_G for both 2K and 4.6K PEG. Nevertheless, δ_w being considerably less than the physical size of PEG (2(5/3)^1/2 R_G) is highly unexpected and implies that, compared to free PEG, anchoring of the PEG tether at both ends leads to a considerable distortion of the PEG conformation confined between layers. Significantly, the lamellar hydrogel may be designed to reversibly transition from a locked to an unlocked (membrane unbinding) state by variations in the DEA-PEG concentration, controlling the strength of the interlayer attractions due to bridging conformations. The findings with DEA-PEGs have broad implications for hydrophobic-mediated assembly of lipid- or surfactant-coated building blocks with distinct shape and size, at predictable spacing, in aqueous environments.

Keywords: PEG lipid; hydrophobic mediated tethering; lamellar phase; lipid bilayer; lipid membrane; self-assembly.

Figure 1.

Design of single and double-ended PEG lipids and control of inter-membrane interaction by tethering. (a) Schematic depictions of the single and double-end-anchored PEG-lipids used in this study. (b) Schematic depiction of the possible conformations of double-end-anchored PEG-lipids in opposing lipid bilayer membranes. The looping (left) and bridging (center) conformations can interchange through the intermediate state with one anchor in the aqueous layer (right, related to the barrier between conformations). The cross-bridging conformation fixes opposing membranes in close proximity in the presence of excess water.

Figure 2.

SAXS profiles of lyotropic phases of mixtures of DOPC, DOTAP and single- and double- end-anchored PEG2000-lipids (50/45/5 molar ratio) at varying water content (Φ_w) (30 wt% to 80 wt%). (a) Control samples containing SEA-PEG (L-P2K), one month after preparation. (b) Samples containing asymmetric DEA-PEG (L-P2K-L’), one month after preparation. (c) Samples containing symmetric DEA-PEG (L-P2K-L), one month after preparation. (d) Samples containing symmetric L-P2K-L, two months after preparation. (e) Overlay of the data shown in (c) and (d) (maintaining the color scheme; light blue: one month after preparation; dark blue: two months after preparation).

Figure 3.

(a) Schematic illustration of the likely mechanism driving phase separation in PEG-lipid-containing lamellar phases at low water content. Rather than form a single phase with a water layer that is too thin (δ_w < δ_w*) to accommodate the PEG chains without a large energy cost arising from their elastic deformation, the system phase separates into a PEG-lipid-containing phase with a larger spacing (δ_w = δ_w*) and a phase with a thin water layer and no PEG-lipid. (b,c) Images of X-ray samples of mixtures of DOPC, DOTAP and single- and double- end-anchored PEG2000-lipids (50/45/5 molar ratio) at varying water content (Φ_w) (30, 40, 50, 60, 70, and 80 wt%, from left to right). (b) Control samples containing SEA-PEG (L-P2K). (c) Samples containing asymmetric DEA-PEG (L-P2K-L’). The opacity of the samples at high and low water content is indicative of two coexisting phases (small droplets of the minority phase suspended in the majority phase).

Figure 4.

Evolution of the interlamellar distance d with water content for lamellar phases of membranes containing 5 mol% of double-end-anchored PEG2K-lipids and controls (single-end-anchored PEG2K-lipids and no PEG-lipid). Membrane compositions: DOTAP/DOPC/PEG2K-lipid=45/50/5 (molar ratio) or DOTAP/DOPC 1:1 (molar ratio) (control, no PEG-lipid). (a) Plot showing the entire range of water contents investigated. (b) Enlarged view of the data shown in (a) for low water content, where two lamellar phases are observed for samples containing PEG-lipids.

Figure 5.

Evolution of the interlamellar distance d with water content (Φ_w) for lamellar phases of membranes containing 2.5 mol% of double-end-anchored PEG2K-lipids and controls (single-end-anchored PEG2K-lipids and no PEG-lipid). Membrane compositions: DOTAP/DOPC/PEG2K-lipid=50/47.5/2.5 (molar ratio) or DOTAP/DOPC 1:1 (molar ratio) (control with no PEG-lipid). The suffix “(50/80)” in the legend designates samples prepared by mixing appropriate amounts of lamellar phases with 50 and 80 wt% water. The suffix “(q001’)” in the legend indicates data for the second lamellar phase that was observed for the respective PEG-lipid (at low water content).

Figure 6.

Evolution of the interlamellar distance d with water content for lamellar phases of membranes containing 5 mol% of double-end-anchored PEG4.6K-lipids and controls (single-end-anchored PEG5K-lipids and no PEG-lipid). Membrane compositions: DOTAP/DOPC/PEG-lipid=50/45/5 (molar ratio) or DOTAP/DOPC 1:1 (molar ratio) (control with no PEG-lipid). The suffix “(q001’)” in the legend indicates data for the second lamellar phase that was observed for the respective PEG-lipid (at low water content). (a) Plot showing the entire range of water contents investigated. (b) Enlarged view of the data shown in (a) for low water content, where two lamellar phases are observed for samples containing PEG-lipids.

Figure 7.

Evolution of the interlamellar spacing d with water content for lamellar phases containing 2.5 mol% of double-end-anchored PEG4.6K-lipids and controls (no PEG-lipid). Membrane compositions: DOTAP/DOPC/PEG-lipid=50/47.5/2.5 (molar ratio) or DOTAP/DOPC 1:1 (molar ratio) (control, no PEG-lipid). (a) Plot showing the entire range of water contents investigated. (b) Enlarged view of the data shown in (a) for low water content. The suffix “(60/90)” in the legend designates samples prepared by mixing appropriate amounts of lamellar phases with 60 and 90 wt% water. The suffix “(q001’)” in the legend indicates data for the second lamellar phase that was observed for the respective PEG-lipid (at low water content).

Scheme 1.

Synthesis of single- and double-end-anchored PEG-lipids; n=45 (PEG MW=2000 g/mol), n=104 (PEG MW=4600 g/mol), or n=113 PEG MW=5000 g/mol) Reagents and conditions: (i) methanesulfonyl chloride, Et₃N; (ii) concentrated aqueous ammonia solution; (iii) HBTU, DIEA; (iv) pentafluorophenol, DCC. DCC: dicyclohexylcarbodiimide; DIEA: diisopropylethylamine; Et₃N: triethylamine; HBTU: N,N,N’,N’-tetramethyl-O-(1H-benzotriazol-1-yl) uronium hexafluorophosphate.