Glycolipid transfer protein interaction with bilayer vesicles: modulation by changing lipid composition

Abstract

Glycosphingolipids (GSLs) are important constituents of lipid rafts and caveolae, are essential for the normal development of cells, and are adhesion sites for various infectious agents. One strategy for modulating GSL composition in lipid rafts is to selectively transfer GSL to or from these putative membrane microdomains. Glycolipid transfer protein (GLTP) catalyzes selective intermembrane transfer of GSLs. To enable effective use of GLTP as a tool to modify the glycolipid content of membranes, it is imperative to understand how the membrane regulates GLTP action. In this study, GLTP partitioning to membranes was analyzed by monitoring the fluorescence resonance energy transfer from tryptophans and tyrosines of GLTP to N-(5-dimethyl-aminonaphthalene-1-sulfonyl)-1,2-dihexadecanoyl-sn-glycero-3-phospho-ethanolamine present in bilayer vesicles. GLTP partitioned to POPC vesicles even when no GSL was present. GLTP interaction with model membranes was nonpenetrating, as assessed by protein-induced changes in lipid monolayer surface pressure, and nonperturbing in that neither membrane fluidity nor order were affected, as monitored by anisotropy of 1,6-diphenyl-1,3,5-hexatriene and 6-dodecanoyl-N,N-dimethyl-2-naphthylamine, even though the tryptophan anisotropy of GLTP increased in the presence of vesicles. Ionic strength, vesicle packing, and vesicle lipid composition affected GLTP partitioning to the membrane and led to the following conclusion: Conditions that increase the ratio of bound/unbound GLTP do not guarantee increased transfer activity, but conditions that decrease the ratio of bound/unbound GLTP always diminish transfer. A model of GLTP interaction with the membrane, based on the partitioning equilibrium data and consistent with the kinetics of GSL transfer, is presented and solved mathematically.

FIGURE 1

(A) RET between GLTP and PC vesicles containing dansyl-DHPE and glycolipid. Emission wavelength spectra were acquired while exciting at 285 nm. Samples were stirred continuously before and after adding GLTP to SUVs composed of POPC/GalCer/dansyl-DHPE (70:20:10 mol%). The increasing fluorescence intensity at 513 nm is indicative of dansyl emission via RET from the Trp and Tyr residues of the protein. (B) Partitioning isotherm for GLTP to PC vesicles containing GalCer and dansyl-DHPE (70:20:10 mol%). Data in panel A representing the emission intensity measured at 513 nm were used to calculate the binding isotherm for GLTP. Protein bound to the vesicles is presented as relative %RET (F_b − F_i/F_i) where F_b and F_i are fluorescence emission intensities of vesicles in the presence and absence of protein, respectively. The error bars represent standard deviation of three experiments performed at 37°C in phosphate buffered saline (2.5 ml) at pH 7.4.

FIGURE 2

Effect of increasing GalCer membrane concentration on GLTP partitioning at low and physiological salt concentrations. The error bars represent standard deviation of three experiments performed at 37°C at pH 7.4.

FIGURE 3

Schematic of the proposed mechanism of GLTP binding to vesicles containing GalCer. Free GLTP from the bulk (P_f) partitions nonspecifically into the lipid phase (P_L). P_L then interacts with the GalCer (L_L) at the interface. The GLTP bound to GSL at the interface may be released into the bulk (P_fL_f). The interaction parameter K_a is a function of K₁, K₂, and K_s. In the absence of GSL at the interface,

FIGURE 4

Anisotropy of DPH (A) and Laurdan (B) in bilayer membranes in the presence and in the absence of GLTP. Experimental conditions are described in Materials and Methods. (C) The steady-state anisotropy of the tryptophans residues of GLTP in the absence and presence of POPC and POPC/GalCer (90:10) vesicles. Experimental conditions are described in Materials and Methods. In each panel, the error bars represent average of three experiments done at 37°C at pH 7.4.

FIGURE 5

Penetration capacity of GLTP for lipid monolayers poised at different initial surface pressures. POPC films are denoted by solid squares. The x-intercept denoting πc = 23.1 mN/m. Linear regression analysis (dotted line) resulted in a correlation coefficient of 0.99745. POPC/GalCer films are denoted by open triangles. The x-intercept denoting πc = 21.9 mN/m. Linear regression analysis (solid line) resulted in a correlation coefficient of 0.9923.

FIGURE 6

Effect of increasing concentration of negatively charged phosphoglyceride in the membrane on GLTP partitioning at low and physiological ionic strength. The error bars represent standard deviation of three experiments performed at 37°C at pH 7.4.

FIGURE 7

Effect of increasing sphingomyelin concentration in the membrane on GLTP partitioning at low and physiological ionic strength. The error bars represent standard deviation of three experiments performed at 37°C at pH 7.4.