Conformational analysis of apolipoprotein E3/E4 heteromerization

Abstract

Apolipoprotein E (apoE) is a 299 residue, exchangeable apolipoprotein that has essential roles in cholesterol homeostasis and reverse cholesterol transport. It is a two-domain protein with the C-terminal (CT) domain mediating protein self-association via helix-helix interactions. In humans, the APOE gene is polymorphic with three common alleles, ε2, ε3, and ε4, occurring in frequencies of ~ 5%, 77%, and 18%, respectively. Heterozygotes expressing apoE3 and apoE4 isoforms, which differ in residue at position 112 in the N-terminal domain (C112 in apoE3 and R112 in apoE4), represent the highest population of ε4 carriers, an allele highly associated with Alzheimer's disease. The objective of this study was to determine if apoE3 and apoE4 have the ability to hybridize to form a heteromer in lipid-free state. Refolding an equimolar mixture of His-apoE3 and FLAG-apoE4 (or vice versa) followed by pull-down and immunoblotting indicated formation of apoE3/apoE4 heteromers. Förster resonance energy transfer between donor fluorophore on one isoform and acceptor on the other, both located in the respective CT domains, revealed a distance of separation of ~ 46 Å between the donor/acceptor pair. Similarly, a quencher placed on one was able to mediate significant quenching of fluorescence emission on the other, indicative of spatial proximity within collisional distance between the two. ApoE3/apoE4 heteromer association was also noted in lipid-associated state in reconstituted lipoprotein particles. The possibility of heteromerization of apoE3/apoE4 bears implications in the potential mitigating role of apoE3 on the folding and physiological behavior of apoE4 and its role in maintaining cholesterol homeostasis.

Keywords: FRET; apolipoprotein E3; apolipoprotein E4; heteromerization; lipoproteins.

Figure 1.. GdnHCl-induced denaturation profile of His-apoE3/His-apoE4 mixtures.

His-apoE3 (black circles), His-apoE4 (red triangles) and His-apoE3/His-apoE4 mixtures (green squares) (0.2 mg/ml) were treated with 0–6 M GdnHCl in 10 mM ammonium bicarbonate buffer, pH 7.4 in the presence of 2x molar excess of TCEP and the % maximal change calculated from the ellipticity values at 222 nm as described under Methods. Data shown are average ± SD (n=3).

Figure 2.. Pull-down of His-apoE3 and FLAG-apoE4 mixture reveals heteromer formation.

Equimolar amounts of His-apoE3 and FLAG-apoE4 were mixed, unfolded, and refolded as described under Materials and Methods, captured by anti-FLAG-antibody (Panels A and B) or Co-Dynabeads (Panels C and D) under non-denaturing conditions. In control reactions, His-apoE3 or FLAG-apoE4 alone were subjected to capture treatment as described above. The bound samples were probed by WB with HRP-anti-His antibody (Panels A and C) or with HRP-anti-FLAG antibody (Panels B and D). Lane 1, His-apoE3; lane 2, FLAG-apoE4; lane 3, His-apoE3/FLAG-apoE4 mixture. His-apoE3 and FLAG-apoE4 alone were loaded as WB control in lanes 4 and 5, respectively.

Figure 3.. FRET between AEDANS- or F5M-labeled-apoE3 and apoE4 heteromers in lipid-free state and response to dilution.

Fluorescence emission spectra of 100 μg/ml (Panel A) or 50 μg/ml (Panel C) F5M-apoE3/AEDANS-apoE4 mixtures (dashed line) or AEDANS-apoE4 alone (solid line) were recorded following excitation at 340 nm. The relative fluorescence intensity is shown to facilitate comparison and assess FRET response to dilution. Fluorescence emission spectra of 100 μg/ml (Panel B) or 50 μg/ml (Panel D) AEDANS-apoE3/F5M-apoE4 mixtures (dashed line) or AEDANS-apoE3 alone (solid line) were recorded as above.

Figure 4.. Quenching of fluorescence from F5M-labeled apoE4 or apoE3 by MTSL-labeled apoE3 or apoE4.

Fluorescence emission spectra of MTSL-apoE3/F5M-apoE4 and F5M-apoE3/MTSL-apoE4 are shown in Left and Right panels, respectively). The spectra were recorded following excitation at 490 nm. Also shown are spectra of F5M-apoE4 alone, Left, and F5M-apoE3 alone, Right, in the absence of quencher.

Figure 5.. Size exclusion chromatography to assess oligomeric state of His-apoE3/His-apoE4 mixtures.

FPLC elution profile following size exclusion chromatography of His-apoE3/His-apoE4 (~10 mg/ml concentration) on a Superdex 200 Increase 10/300 GL column in 20 mM sodium phosphate buffer, pH 7.4, containing 150 mM NaCl (phosphate-buffered saline, PBS). His-apoE4 was used as a control. FPLC was carried out at a flow rate of 0.5 ml/min. The standards used and their corresponding molecular masses were: horse heart myoglobin, bovine serum albumin and bovine immunoglobulin (~16,950, 66,430 and 150,000 Da, respectively).

Figure 6.. FRET between AEDANS-apoE4 and F5M-apoE3 in lipid-associated state as rHDL.

The emission spectra of rHDL samples in PBS, pH 7.4 (200 µg/ml protein complexed with POPC) were recorded following excitation at 340 nm.

Figure 7.. Schematic model depicting apoE3/apoE4 heteromerization.

The schematic illustration is a hypothetical model of protein-protein interaction between the two isoforms forming apoE3/apoE4 heteromer. The NT domain helix bundle is based on X-ray structure of apoE3(1–191) (blue) (PDB ID# 1NFN) and apoE4(1–191) (red) (PDB ID# 1LE4). Segments that were unstructured due to high mobility are either not shown (residues 1–22) or schematically modeled (residues 165–191). The linker loop and the CT domain depictions (residues 192–299) are based on secondary structure predictions and other biophysical studies. The dashed line in apoE4 represents the proposed salt bridge between R61 and E255. The segments that likely mediate heteromerization between the two isoforms are indicated as helices with oval patches towards the C-terminal end.