ORCID
https://orcid.org/0000-0002-3899-5651
Date of Award
12-19-2023
Degree Name
Doctor of Philosophy (PhD)
Degree Type
Dissertation
Abstract
Apolipoprotein E (ApoE) is the major lipoprotein in the central nervous system and forms high density lipoprotein (HDL) like lipoprotein. ApoE is primarily secreted by astrocytes in the central nervous system and is thought to play an important role in cholesterol and lipid transport. The APOE gene has three major alleles, APOE2, APOE3, and APOE4. These alleles encode for 3 common APOE protein isoforms, ApoE2, ApoE3, and ApoE4. APOE3 is most common isoform of APOE with the other two isoforms differing by the loss (ApoE4, Cys112Arg) or gain (ApoE2, Arg158Cys) of a cysteine residue. APOE4 is the largest risk factor for late onset Alzheimer’s disease (LOAD) with one allele increasing LOAD risk by ~4-fold and two alleles increasing LOAD risk by ~12-fold. APOE2 is moderately protective for LOAD but is a risk factor for Type III Hyperlipoproteinemia, due to the reduced binding of ApoE2 to the low density lipoprotein receptor (LDLR). ApoE4 is known to increase amyloid-β (Aβ) plaque accumulation in murine models of Aβ amyloidosis and to increase neurodegeneration in murine models of tauopathy. However, the mechanism by which APOE4 increases LOAD risk is not understood. Current understanding of the ApoE4 structure is based off of X-ray crystallography of the nonlipidated, N-terminal fragment of ApoE. This structure revealed that nonlipidated ApoE adopts an elongated α-helical 4-helix bundle. This structure has led to the hypothesis that salt bridge rearrangements that occur in ApoE2 may lead to a decrease in the positive potential of the LDLR binding region, explaining the decreased affinity of ApoE2 for LDLR. However, the insights from this model are limited as ApoE is lipidated in its physiological context. Upon lipidation, the 4-helix bundle characteristic of the nonlipidated form opens as the hydrophobic portions of the amphipathic helix dissociate and associate with the hydrophobic tail groups of lipids. However, previous techniques which investigated the structure of lipidated ApoE have been limited. Initial attempts to use X-ray crystallography to solve the structure of lipidated ApoE were hampered by low resolution. To investigate the structure and function of lipidated ApoE, I utilized cryogenic electron microscopy (cryoEM). CryoEM allows for structure determination to near angstrom resolution of proteins in their near native state while suspended in vitreous ice. In combination with cryoEM, I utilized anti-ApoE monoclonal Fab and F(ab’)2 fragments to characterize ApoE on discoidal lipoprotein. ApoE lipoproteins immunopurified from astrocyte conditioned media were found to form discoidal lipoprotein containing two ApoE molecules in an antiparallel conformation. These results expand upon previous research that reported astrocytes secrete discoidal nascent HDL-like lipoprotein containing ApoE. In order to obtain higher resolution of ApoE by cryoEM, I generated artificially lipidated, recombinant ApoE lipoprotein. The sodium cholate dialysis method was able to produce ApoE lipoproteins that resembled astrocyte secreted lipoproteins. The artificial lipoproteins were discoidal in nature and ranged in size from ~8nm to ~20nm depending on the lipid composition of the lipoproteins. ApoE was present as an antiparallel dimer on both 1,2-dimyristoyl-sn-glycero-3-phosphocholine (DMPC) and 1,2-dipalmitoyl-sn-glycero-3-phosphocholine (DPPC) discoidal lipoproteins. These results suggest a new paradigm for understanding the conformation of ApoE in a physiological context. The importance of the physiological conformation of ApoE is emphasized by the finding that interaction between the two ApoE molecules on the nanodisc alters the functionality of the ApoE lipoprotein. Similar to ApoA-I, ApoE can adopt different registries on discoidal lipoprotein. Registry refers to the alignment of different helical domains between the two apolipoproteins on the discoidal lipoprotein. In the context of ApoA-I, ApoA-I primarily adopts the L5/L5 registry with a minor fraction adopting an L5/L2 registry. ApoA-I in both registries can mediate cholesterol efflux, however, the L5/L2 registry had dramatically reduced lecithin–cholesterol acyltransferase (LCAT) activity compared to the L5/L5 registry. Similar to ApoA-I, ApoE appears to adopt different registries, however, the presence of a Cys residue in ApoE allows for ApoE to become locked in a particular registry. ApoE3 contains a single Cys residue, ApoE2 two Cys residues, and ApoE4 contains no Cys residues. ApoE3 is able to form a disulfide bridge between the two ApoE molecules on discoidal lipoprotein, locking ApoE into a registry where residue 112 is aligned. Uptake by Neuro2a cells was dramatically increased for disulfide linked ApoE3 lipoprotein compared to ApoE2, ApoE4, and non-disulfide linked ApoE3 lipoprotein. ApoE uptake was normalized after reduction of the disulfide bond, demonstrating that disulfide linkage is a key regulator in ApoE lipoprotein function. In addition to ApoE structure, ApoE lipoprotein function is modified by additional protein and lipid cargo. Knockout of astrocyte and microglia ApoE has been shown to have differential effects on Aβ and tau pathology. Lipidomics of immunopurified ApoE lipoprotein showed that astrocyte ApoE lipoproteins were enriched in phosphatidylcholine species while microglia ApoE lipoproteins were enriched with sphingomyelin. Previous results showed a dramatic effect of lipid composition on ApoE lipoprotein size suggesting that differential lipid content between astrocyte and microglia lipoproteins could have functional differences. Proteomics of ApoE lipoproteins identified differential abundance of proteins in either astrocyte or microglia secreted ApoE lipoproteins. Of particular interest is Gpmnb which is predominantly secreted by microglia. Gpmnb has been implicated in the fibrilization of α-synuclein, potentially driving Lewy body formation. Together, these results suggest that lipid and protein content specific to certain cell types could lead to altered function of otherwise identical ApoE lipoproteins. Together this research yields important mechanistic insights into the physiological function of ApoE in its lipidated state. Future research will be necessary to achieve a high resolution structure of lipidated ApoE and better understand how ApoE interacts with its receptors. Importantly, the presence of ApoE as an antiparallel, dimer in discoidal lipoprotein drives its ability to bind to its receptors. These results demonstrate that the interaction between the two ApoE molecules have important functional consequences. Of particular interest is that the single nucleotide polymorphisms that define the different APOE alleles all impact Cys residues in the ApoE protein. Formation of disulfide bridges between the two ApoE molecules provides a novel regulatory mechanism for “locking” ApoE into particular registries. This mechanism is loss in ApoE4, providing a potential mechanistic explanation for why ApoE4 has detrimental impacts on AD and atherosclerosis progression. Additionally, beyond ApoE itself, the secretion of particular lipids and proteins by different cell types could be an important factor in modulating the function of ApoE lipoproteins from different cell types.
Language
English (en)
Chair and Committee
David Holtzman
Recommended Citation
Strickland, Michael Ray, "Apolipoprotein E Adopts an Antiparallel Dimer in Discoidal Lipoproteins" (2023). Arts & Sciences Electronic Theses and Dissertations. 3206.
https://openscholarship.wustl.edu/art_sci_etds/3206