Search

The apolipoprotein family is structurally defined by amphipathic α-helical regions that interact with lipid surfaces. In the absence of lipid, human apolipoprotein (apo) C-II also forms well-defined amyloid fibrils with cross-β structure. Formation of this β-structure is accompanied by the burial of two charged residues, K30 and D69, that form an ion-pair within the amyloid fibril core. Molecular dynamics (MD) simulations indicate these buried residues form both intra- and intersubunit ion-pair interactions that stabilize the fibril. Mutations of the ion-pair (either K30D or D69K) reduce fibril stability and prevent fibril formation by K30D apoC-II under standard conditions. We investigated whether mixing K30D apoC-II with other mutants would overcome this loss of fibril forming ability. Co-incubation of equimolar mixtures of K30D apoC-II with wild-type, D69K, or double-mutant (K30D/D69K) apoC-II promoted the incorporation of K30D apoC-II into hybrid fibrils with increased stability. MD simulations showed an increase in the number of intersubunit ion-pair interactions accompanied the increased stability of the hybrid fibrils. These results demonstrate the important role of both intra- and intersubunit charge interactions in stabilizing apoC-II amyloid fibrils, a process that may be a key factor in determining the general ability of proteins to form amyloid fibrils.

Apolipoproteins form amphipathic helical structures that bind lipid surfaces. Paradoxically, lipid-free apolipoproteins display a strong propensity to form cross-β structure and self-associate into disease-related amyloid fibrils. Studies of apolipoprotein C-II (apoC-II) amyloid fibrils suggest that a K30-D69 ion pair accounts for the dual abilities to form helix and cross-β structure. Consistent with this is the observation that a K30D mutation prevents fibril formation under standard fibril forming conditions. However, we found that fibril formation by K30D apoC-II proceeded readily at low pH and a higher salt or protein concentration. Structural analysis demonstrated that K30D apoC-II fibrils at pH 7 have a structure similar to that of the wild-type fibrils but are less stable. Molecular dynamics simulations of the wild-type apoC-II fibril model at pH 7 and 3 showed that the loss of charge on D69 at pH 3 leads to greater separation between residues K30 and D69 within the fibril with a corresponding reduction in β-strand content around residue 30. In contrast, in simulations of the K30D mutant model at pH 7 and 3, residues D30 and D69 moved closer at pH 3, accompanied by an increase in β-strand content around residue 30. The simulations also demonstrated a strong dominance of inter- over intramolecular contacts between ionic residues of apoC-II and suggested a cooperative mechanism for forming favorable interactions between the individual strands under different conditions. These observations demonstrate the important role of the buried K30-D69 ion pair in the stability and solution properties of apoC-II amyloid fibrils.

Plasma apolipoproteins form amphipathic α helices in lipid environments but in the lipid-free state show a high propensity to form β structure and self-associate into amyloid fibrils. The widespread occurrence of apolipoproteins in amyloid plaques suggests disease-related roles, specifically in atherosclerosis. To reconcile the dual abilities of apolipoproteins to form either α helices or cross-β sheet structures, we examined fibrils formed by human apolipoprotein C-II (apoC-II). A structural model for apoC-II fibrils shows a cross-β core with parallel β strands, including a buried K30-D69 charge pair. We investigated the effect of abolishing this charge pair in mutant D69K apoC-II. Fluorescence studies indicated more rapid fibril formation and less solvent accessibility of tryptophan (W26) in D69K apoC-II fibrils than in wild-type (WT) fibrils. X-ray diffraction data of aligned D69K apoC-II fibrils yielded a typical cross-β structure with increased β sheet spacing compared to that of WT fibrils. Hydrogen/deuterium (H/D) exchange patterns were similar for D69K apoC-II fibrils compared to WT fibrils, albeit with an overall reduction in the level of slow H/D exchange, particularly around residues 29-32. Molecular dynamics simulations indicated reduced β strand content for a model D69K apoC-II tetramer compared to the WT tetramer and confirmed an expansion of the cross-β spacing that contributed to the formation of a stable charge pair between K69 and E27. The results highlight the importance of charge-pair interactions within the apoC-II fibril core, which together with numerous salt bridges in the flexible connecting loop play a major role in the ability of lipid-free apoC-II to form stable cross-β fibrils.

Protein misfolding and aggregation, leading to amyloid fibril formation, are characteristic of many devastating and debilitating amyloid diseases. Accordingly, there is significant interest in the mechanisms underlying amyloid fibril formation and identification of possible intervention tools. Small molecule drug compounds approved for human use or for use in phase I-III clinical trials were investigated for their effects on amyloid formation by human apolipoprotein (apo) C-II. Several of these compounds modulated the rate of amyloid formation by apoC-II. Epigallocatechin gallate (EGCG), a green tea catechin, was an effective inhibitor of apoC-II fibril formation, and the antipsychotic drug, fluphenazine·HCl, was a potent activator. Both EGCG and fluphenazine·HCl exerted concentration-dependent effects on the rate of fibril formation, bound to apoC-II fibrils with high affinity, and competitively reduced thioflavin T binding. EGCG significantly altered the size distribution of fibrils, most likely by promoting the lateral association of fibrils and subsequent formation of large aggregates. Fluphenazine·HCl did not significantly alter the size distribution of fibrils, but it may induce the formation of a small population of rod-like fibrils that differ from the characteristic ribbon-like fibrils normally observed for apoC-II. The findings of this study emphasize the effects of small molecule drugs on the kinetics of amyloid fibril formation and their roles in determining fibril structure and aggregate size.

The misfolding, aggregation, and accumulation of proteins as amyloid fibrils is a defining characteristic of several debilitating diseases. Human apolipoprotein C-II (apoC-II) amyloid fibrils are representative of the fibrils formed by a number of plasma apolipoproteins implicated in amyloid-related disease. Previous structural analyses identified a buried charge pair between residues K30 and D69 within apoC-II amyloid fibrils. We have investigated the effects of amino acid substitutions of these residues on apoC-II fibril formation. Two point mutations of apoC-II, D69K and K30D, as well as a reversed ion-pair mutant containing both mutations (KDDK) were generated. Fibril formation by the double mutant, apoC-II KDDK, and apoC-II D69K was enhanced compared to that of wild-type (WT) apoC-II, while apoC-II K30D lacked the ability to form fibrils under standard conditions. Structural analyses showed that WT apoC-II, apoC-II D69K, and apoC-II KDDK fibrils have similar secondary structures and morphologies. Size distribution analyses revealed that apoC-II D69K fibrils have a broader range of fibril sizes while apoC-II KDDK fibrils showed an increased frequency of closed fibrillar loops. ApoC-II D69K fibrils exhibited reduced thioflavin T binding capacity compared to that of fibrils formed by WT apoC-II and apoC-II KDDK. These results indicate that specific charge and charge-pair mutations within apoC-II significantly alter the ability to form fibrils and that position 69 within apoC-II plays a key role in the rate-limiting step of apoC-II fibril formation.

Human apolipoprotein (apo) C-II is one of several lipid-binding proteins that self-assemble into fibrils and accumulate in disease-related amyloid deposits. A general characteristic of these amyloid deposits is the presence of lipids, known to modulate individual steps in amyloid fibril formation. ApoC-II fibril formation is activated by submicellar phospholipids but inhibited by micellar lipids. We examined the mechanism for the activation by submicellar lipids using the fluorescently labeled, short-chain phospholipid 1-dodecyl-[(7-nitro-2-1,3-benzoxadiazol-4-yl)amino]-2-hydroxyglycero-3-phosphocholine (NBD-lyso-12-PC). Addition of submicellar NBD-lyso-12-PC increased the rate of fibril formation by apoC-II approximately 2-fold. Stopped flow kinetic analysis using fluorescence detection and low, non-fibril-forming concentrations of apoC-II indicated NBD-lyso-12-PC binds rapidly, on the millisecond time scale, followed by the slower formation of discrete apoC-II tetramers. Sedimentation velocity analysis showed NBD-lyso-12-PC binds to both apoC-II monomers and tetramers at approximately five sites per monomer with an average dissociation constant of approximately 10 μM. Mature apoC-II fibrils formed in the presence of NBD-lyso-12-PC were devoid of lipid, indicating a purely catalytic role for submicellar lipids in the activation of apoC-II fibril formation. These studies demonstrate the catalytic potential of small amphiphilic molecules in controlling protein folding and fibril assembly pathways.

The misfolding and self-assembly of proteins into amyloid fibrils that occur in several debilitating diseases are affected by a variety of environmental factors, including mechanical factors associated with shear flow. We examined the effects of shear flow on amyloid fibril formation by human apolipoprotein C-II (apoC-II). Shear fields (150, 300, and 500 s(-1)) accelerated the rate of apoC-II fibril formation (1 mg/mL) approximately 5-10-fold. Fibrils produced at shear rates of 150 and 300 s(-1) were similar to the twisted ribbon fibrils formed in the absence of shear, while at 500 s(-1), tangled ropelike structures were observed. The mechanism of the shear-induced acceleration of amyloid fibril formation was investigated at low apoC-II concentrations (50 μg/mL) where fibril formation does not occur. Circular dichroism and tryptophan fluorescence indicated that shear induced an irreversible change in apoC-II secondary structure. Fluorescence resonance energy transfer experiments using the single tryptophan residue in apoC-II as the donor and covalently attached acceptors showed that shear flow increased the distance between the donor and acceptor molecules. Shear-induced higher-order oligomeric species were identified by sedimentation velocity experiments using fluorescence detection, while fibril seeding experiments showed that species formed during shear flow are on the fibril formation pathway. These studies suggest that physiological shear flow conditions and conditions experienced during protein manufacturing can exert significant effects on protein conformation, leading to protein misfolding, aggregation, and amyloid fibril formation.

A common feature of many amyloid diseases is the appearance of oxidized, aggregated proteins. Methionine is one of the most readily oxidized amino acids, and its oxidative state is regulated in vivo by the methionine sulfoxide reductases (Msr). Here, we have explored the basis by which methionine oxidation is linked to amyloid disease by comparing the reduction of oxidized amyloid fibrils and monomer. We show that oxidized amyloid fibrils are not as effectively reduced by the Msr enzymes as the monomer. This work suggests a mechanism by which oxidized proteins and aggregates can accumulate as a part of degenerative disease.

Methionine residues are linked to the pathogenicity of several amyloid diseases; however, the mechanism of this relationship is largely unknown. These diseases are characterized, in vivo, by the accumulation of insoluble proteinaceous plaques, of which the major constituents are amyloid fibrils. In vitro, methionine oxidation has been shown to modulate fibril assembly in several well-characterized amyloid systems. Human apolipoprotein (apo) C-II contains two methionine residues (Met-9 and Met-60) and readily self-assembles in vitro to form homogeneous amyloid fibrils, thus providing a convenient system to examine the effect of methionine oxidation on amyloid fibril formation and stability. Upon oxidation of the methionine residues of apoC-II with hydrogen peroxide, fibril formation was inhibited. Oxidized apoC-II molecules did not inhibit native apoC-II assembly, indicating that the oxidized molecules had a reduced ability to interact with the growing fibrils. Single Met-Val substitutions were performed and showed that oxidation of Met-60 had a more significant inhibitory effect than oxidation of Met-9. In addition, Met-Gln substitutions designed to mimic the effect of oxidation on side chain hydrophilicity showed that a change in hydrophobicity at position 60 within the core region of the fibril had a potent inhibitory effect. The oxidation of preformed apoC-II fibrils caused their dissociation; however, mutants in which the Met-60 was substituted with a valine were protected from this peroxide-induced dissociation. This work highlights an important role for methionine in the formation of amyloid fibril structure and gives new insight into how oxidation affects the stability of mature fibrils.