Intrasequence GFP in class I MHC molecules, a rigid probe for fluorescence anisotropy measurements of the membrane environment

Abstract

Fluorescence anisotropy measurements can elucidate the microenvironment of a membrane protein in terms of its rotational diffusion, interactions, and proximity to other proteins. However, use of this approach requires a fluorescent probe that is rigidly attached to the protein of interest. Here we describe the use of one such probe, a green fluorescent protein (GFP) expressed and rigidly held within the amino acid sequence of a major histocompatibility complex (MHC) class I molecule, H2L(d). We contrast the anisotropy of this GFP-tagged MHC molecule, H2L(d)GFPout, with that of an H2L(d) that was GFP-tagged at its C-terminus, H2L(d)GFPin. Both molecules fold properly, reach the cell surface, and are recognized by specific antibodies and T-cell receptors. We found that polarized fluorescence images of H2L(d)GFPout in plasma membrane blebs show intensity variations that depend on the relative orientation of the polarizers and the membrane normal, thus demonstrating that the GFP is oriented with respect to the membrane. These variations were not seen for H2L(d)GFPin. Before transport to the membrane surface, MHC class I associates with the transporter associated with antigen processing complex in the endoplasmic reticulum. The intensity-dependent steady-state anisotropy in the ER of H2L(d)GFPout was consistent with FRET homotransfer, which indicates that a significant fraction of these molecules were clustered. After MCMV-peptide loading, which supplies antigenic peptide to the MHC class I releasing it from the antigen processing complex, the anisotropy of H2L(d)GFPout was independent of intensity, suggesting that the MHC proteins were no longer clustered. These results demonstrate the feasibility and usefulness of a GFP moiety rigidly attached to the protein of interest as a probe for molecular motion and proximity in cell membranes.

FIGURE 1

Definition of angles used to describe the geometries obtained from a confocal image of a cell bleb. (A) Image plane of a spherical membrane. Excitation polarizer was kept horizontal while images were collected with emission polarizer parallel and perpendicular to this. Intensity is extracted from these images as a function of the angle around the membrane (β). (B) Parameters used in geometric model. N is the membrane normal axis. The absorption and emission dipole vectors, a and e, are described by the angles between N and the vector (θ_a and θ_e) as well as the rotation of e around N(ψ).

FIGURE 2

Fluorescence polarization images of H2L^dGFPin and H2L^dGFPout cell blebs. The excitation polarizer was kept horizontal for the collection of these images. These images were collected on H2L^dGFPin (A, C, and E) and H2L^dGFPout (B, D, and F) cells that were induced to bleb using 1 mM H₂O₂. The emission polarizer was placed parallel (hh) for the collection of A and B. The emission polarizer was placed perpendicular (hv) for the collection of C and D. These images are also shown superimposed upon one another with pseudocolored green (hh) and red (hv). The bar shown in F is 5 μm.

FIGURE 3

Fluorescence polarization intensities obtained from the periphery of cell blebs. Intensity was extracted at 10° increments from >100 membrane blebs for both H2L^dGFPin (A) and H2L^dGFPout (B). The H2L^dGFPout intensity varied substantially in both the I_hh and I_hv images and this allowed least-squares analysis fitting using the model described (Materials and Methods). Error bars indicate the standard error of the mean for each angle (β) measured.

FIGURE 4

Steady-state fluorescence anisotropy obtained from the ER of H2L^dGFPin (A and B) and H2L^dGFPout (C and D). The anisotropy was calculated using the correction factors described (Materials and Methods) from intensities extracted from confocal images. The intensities were extracted from individual cells from regions thought to be the ER. MCMV treatment was done at a concentration of 100 μg/ml in cell media overnight (B and D). In the bottom right corner is the average steady-state anisotropy measured (mean ± SE) from H2L^dGFPin (N = 262), H2L^dGFPin + MCMV (N = 318), H2L^dGFPout (N = 196), and H2L^dGFPout + MCMV (N = 212).

FIGURE 5

Controlled bleach experiments of H2L^dGFPin and H2L^dGFPout cells. Images were collected sequentially as described (Materials and Methods), which resulted in gradual bleaching of the GFP fluorescence. Bleach measurements were done on H2L^dGFPin (A, N = 5), H2L^dGFPin + MCMV (B, N = 5), H2L^dGFPout (C, N = 5), and H2L^dGFPout + MCMV (D, N = 10). N represents the number of fields of view used in each bleach curve and each field of view contained from 3–6 cells. These curves were fit linearly and the y-intercept is shown in the bottom left of each graph (± 95% confidence interval). This value is referred to as the null homotransfer anisotropy.

FIGURE 6

Controlled bleach experiment with lactacystin-treated H2L^dGFPout cells. Images were collected sequentially as described (Materials and Methods), which resulted in gradual bleaching of the GFP fluorescence. Bleach measurements were done on H2L^dGFPout (▪, N = 5) and H2L^dGFPout + lactacystin treatment (•, N = 10).