HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 275, Issue 27, 20775-20781, July 7, 2000

This Article Abstract Full Text (PDF) All Versions of this Article: 275/27/20775    most recent M003508200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Lu, B. Articles by Weisgraber, K. H. Search for Related Content PubMed PubMed Citation Articles by Lu, B. Articles by Weisgraber, K. H. Social Bookmarking                      What's this?

Conformational Reorganization of the Four-helix Bundle of Human Apolipoprotein E in Binding to Phospholipid^*

Bin Lu, Julie A. Morrow, and Karl H. Weisgraber^§¶

From the  Gladstone Institute of Cardiovascular Disease, Cardiovascular Research Institute, and ^§ Department of Pathology, University of California, San Francisco, California 94141-9100

Received for publication, April 25, 2000, and in revised form, May 4, 2000

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Conformational reorganization of the amino-terminal four-helix bundle (22-kDa fragment) of apolipoprotein E (apoE) in binding to the phospholipid dimyristoylphosphatidylcholine (DMPC) to form discoidal particles was investigated by introducing single, double, and triple interhelical disulfide bonds to restrict the opening of the bundle. Interaction of apoE with DMPC was assessed by vesicle disruption, turbidimetric clearing, and gel filtration assays. The results indicate that the formation of apoE·DMPC discoidal particles occurs in a series of steps. A triple disulfide mutant, in which all four helices were tethered, did not form complexes but could release encapsulated 5-(6)-carboxylfluorescein from DMPC vesicles, indicating that the initial interaction does not involve major reorganization of the helical bundle. Initial interaction is followed by the opening of the four-helix bundle to expose the hydrophobic faces of the amphipathic helices. In this step, helices 1 and 2 and helices 3 and 4 preferentially remain paired, since these disulfide-linked mutants bound to DMPC in a manner similar to that of the 22-kDa fragment of apoE4. In contrast, mutants in which helices 2 and 3 and/or helices 1 and 4 paired bound poorly to DMPC. However, all single and double helical pairings resulted in the formation of larger discs than were formed by the 22-kDa fragment, indicating that further reorganization of the helices occurs following the initial opening of the four-helix bundle in which the protein assumes its final lipid-bound conformation. In support of this rearrangement, reducing the disulfide bonds converted the large disulfide mutant discs to normal size.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Apolipoprotein E (apoE)¹ plays an important role in lipoprotein metabolism and neurobiology (1-4). Through its high affinity interaction with the low density lipoprotein (LDL) receptor family, apoE regulates plasma triglyceride clearance and contributes to plasma cholesterol homeostasis (1, 5). Association of apoE with lipid is required for many of its functions, including high affinity binding to the LDL receptor (6) and modulating neuron maintenance and regeneration (reviewed in Ref. 3). However, the conformational changes in apoE associated with lipid binding remain unclear.

Intact human apoE (299 amino acids) contains two independently folded domains (7-9): the 22-kDa amino-terminal domain (residues 1-191), which contains the LDL receptor-binding region (vicinity of residues 136-150), and the 10-kDa carboxyl-terminal domain (residues 216-299), which contains the major lipid-binding elements for spherical lipoprotein particles (reviewed in Ref. 2). In the lipid-free state, the 22-kDa domain contains a four-helix bundle (10). Each helix is amphipathic, and its hydrophobic face is oriented toward the center of the bundle. This structure shares the same basic architecture as the five-helix bundle found in apolipophorin III (apoLpIII) from Locusta migratoria (11) and Manduca sexta (12). Amphipathic -helices represent the lipid-binding structural motifs found in plasma apolipoproteins (13).

While the 22-kDa fragment by itself does not bind to spherical lipoprotein particles, it does bind to and reorganize phospholipid vesicles to form bilayer discoidal complexes. In this lipid-associated state, the 22-kDa fragment displays normal high affinity binding to the LDL receptor (14). It was hypothesized that the four-helix bundle undergoes a lipid-triggered reorganization of the helices that exposes the hydrophobic faces to interact with lipid (2), similar to the model proposed for apoLpIII (11). Molecular area measurements at an air-water interface support this hypothesis (15). Recently, fluorescence resonance energy transfer studies showed that the distance between helices 1 and 3 of the bundle undergoes a dramatic change in binding to lipid (16), also suggesting conformational reorganization of the helical bundle. Several studies demonstrated that there is no significant change in -helical content when apoE binds to various lipid surfaces (9, 17). Therefore, it is likely that the conformational changes associated with the lipid binding involve rearrangement of the relative positions of the helices without a major disruption of the secondary structure. Two models of the conformation of apoE on phospholipid discs have been proposed: the "picket fence" model, in which amphipathic helices are aligned parallel to the phospholipid acyl chains (18), and the "belt" model in which the helices are aligned perpendicular to the acyl chains (19).

In the present study, lipid-triggered conformational unfolding of apoE was investigated using a disulfide bond engineering approach. The results demonstrate that the interactions of apoE with dimyristoylphosphatidylcholine (DMPC) can be viewed as a series of steps, including initial interaction, lipid-triggered opening of the four-helix bundle, and further reorganization of the helices to achieve the final lipid-bound conformation.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Site-directed Mutagenesis and Protein Expression-- A thioredoxin fusion expression vector, pET32-E4NT (20), containing the full-length human apoE4 cDNA lacking cysteine codons, served as the starting template. Using site-directed mutagenesis with the polymerase chain reaction (20, 21), we introduced a stop codon immediately after codon 191 to generate the construct expressing the amino-terminal 22-kDa fragment. Further polymerase chain reaction mutagenesis with mutagenic oligonucleotide primers (Oligos Etc., Wilsonville, OR) was performed to introduce cysteine residues at selected positions. Cloned Pfu DNA polymerase (Stratagene) was used for high fidelity amplifications. All constructs were sequenced to confirm the presence of the desired mutations and to ensure that no other mutations were introduced during amplification.

The apoE4 22-kDa fragment and the cysteine-containing mutants were expressed by transforming the appropriate expression constructs into a T7 expression strain Escherichia coli BL21 (DE3) (Novagen). Expression of thioredoxin-apoE fusion protein was induced for 2 h at 37 °C by adding isopropyl -D-thioglactopyranoside (100 mg/liter final concentration) to a culture at its midlog phase (A₆₆₀ ~ 0.6). The induced cells were pelleted by centrifugation and resuspended in a small volume of extraction buffer containing 20 mM Tris-HCl, pH 7.9, 5 mM imidazole, and 500 mM NaCl. The fusion protein was extracted by sonication (Branson Sonifier 450) and purified by His·Bind resin according to the manufacture's procedures (Novagen). Fusion proteins were cleaved by incubation with thrombin (Hematologic Technologies) for 2 h at room temperature at a ratio of 1:100 (thrombin/fusion protein (w/w)). Recombinant proteins were denatured in 6 M guanidine-HCl (Roche Molecular Biochemicals), 100 mM Tris-HCl, pH 7.4, 1 mM EDTA, and 1% -mercaptoethenol (BME) (Sigma) and purified on a Sephacryl S-300 HR (Amersham Pharmacia Biotech) column (4 × 120 cm) eluted with 4 M guanidine-HCl, 100 mM Tris-HCl, pH 7.4, 1 mM EDTA, and 0.1% BME. Fractions containing apoE were identified by SDS-polyacrylamide gel electrophoresis (SDS-PAGE) and Western blotting with a rabbit anti-human apoE polyclonal antibody, as described previously (20).

Protein Refolding and Disulfide Bond Formation-- Refolding of mutant proteins was carried out according to our established procedure for apoE 22-kDa fragments used in receptor-binding assays (20) and x-ray crystallography (22). Briefly, the apoE-containing fractions, at concentrations less than 0.01 mg/ml, were dialyzed extensively against 5 mM NH₄HCO₃, 0.1% BME, to allow refolding of the protein into its native conformation. The BME was subsequently removed by dialysis against 5 mM NH₄HCO₃ to allow air oxidization and disulfide bond formation. The disulfide-linked forms were separated from mutant proteins species containing free sulfhydryl groups by activated Thiol-6B chromatography (Amersham Pharmacia Biotech). Final purification and concentration was performed by DEAE chromatography (Seperco) on a BioCAD 700E workstation (Perspective Systems). Protein concentration was determined by the method of Lowry et al. (23).

Structural Characterization and Disulfide Bond Detection-- The mutant proteins were characterized by circular dichroism in both the free cysteine and disulfide-linked forms with an apoE4 22-kDa fragment as a control. Proteins (250 µg/ml) were dialyzed in 20 mM sodium phosphate buffer, pH 7.4, and circular dichroism measurements were made on a JASCO 715 spectropolarometer at 222 nm ([]₂₂₂) at 25 °C. The -helical content was calculated from the molar elipticity values derived from the following equation (24).

The formation of interhelical disulfide bonds was assessed by two independent methods: fluorescence chemical modification and electrophoretic mobility shift assays. Mutant proteins (5 µg) in their free cysteine and disulfide-linked forms were incubated with equal volumes of SDS loading buffer (62.5 mM Tris-HCl, pH 6.8, 25% glycerol, and 2% SDS), containing 20 mM fluorescein 5-maleimide (Molecular Probes, Inc., Eugene, OR), at room temperature for 30 min and subjected to nonreducing SDS-PAGE on 10-20% gels. After the electrophoresis, the gel was visualized with an ultraviolet transluminator and photographed. Free sulfhydryl groups were detected by fluorescence reactivity (25). The same gel was then stained with Coomassie Blue R-250 to visualize all of the protein bands and to determine the mobility shifts resulting from interhelical disulfide bonds (26). Purified apoE3 22-kDa fragment, which contains a single cysteine residue, was used as a control.

Preparation of DMPC Vesicles-- DMPC (Avanti Polar-Lipids) was resuspended in benzene and lyophilized into dry powder, which was dispersed in a buffer containing 20 mM Tris-HCl, pH 7.4, 0.15 M NaCl, and 1 mM EDTA, with and without the presence of 100 mM 5-(6)-carboxylfluorescein (CF) (Molecular Probes) to a final lipid concentration of 10 mg/ml. Small unilamellar vesicles were prepared by sonicating the DMPC solution at 24 °C for 20 min. The CF-containing vesicles were purified by gel filtration on a Superdex 200 column (Amersham Pharmacia Biotech) to remove nonencapsulated CF. The concentration of DMPC was determined with a phospholipid assay kit (Wako Chemicals).

Fluorescence Release-- The release of fluorescent dye from CF-containing DMPC vesicles was monitored on a Hitachi spectrofluorometer F-2000, with excitation and emission set at 480 and 518 nm, respectively. The increase in fluorescence due to the release of dye from the DMPC vesicles was recorded after adding 10 µl of protein (1 µM final concentration) to a jacketed cuvette containing 500 µl of DMPC solution (60 µM final concentration) at 23.9 °C (DMPC transition temperature). All solutions and protein samples were preincubated at 23.9 °C.

Turbidimetric Clearing Assay-- The abilities of the mutants to reorganize DMPC vesicles into apoE·DMPC discoidal complexes was determined by the decrease in turbidity (clearing) of DMPC solutions, as monitored by the decrease in 90° light scattering on a Hitachi spectrofluorometer F-2000 with both excitation and emission set at 600 nm (27, 28). The samples were maintained at 23.9 °C in a jacketed cuvette and continuously stirred with a magnetic bar. All solutions and protein samples were preincubated at 23.9 °C. The kinetic curves were fitted to a single exponential decay function (GraphPad Prism, version 2.0) from which the t_1/2, defined as the time required for a 50% decrease of the initial turbidity, was calculated (27).

Gel Filtration Binding Assay-- The 22-kDa fragments of apoE4 and the cysteine-containing mutants (100 µg) were added to DMPC vesicles (375 µg) (3.75:1, DMPC/apoE, w/w) with and without reducing reagents (1% BME and 100 mM dithiothreitol) and incubated at 23.9 °C for 10 h to allow apoE·DMPC complexes to form. The reaction mixture was applied to a Superdex 200 gel filtration column (Amersham Pharmacia Biotech) to separate the apoE·DMPC complexes from unbound protein. The column was eluted with 20 mM Tris-HCl, pH 7.4, 150 mM NaCl, and 1 mM EDTA at a flow rate of 0.5 ml/min. The unbound protein was quantified by area integration based on a standard curve obtained from applying known amounts of apoE 22-kDa fragment to the column.

Determination of ApoE·DMPC Disc Size-- Relative apoE·DMPC disc sizes were compared by the elution time of the complex on Superdex 200 gel filtration column (Amersham Pharmacia Biotech) (29) eluted as described above. The purified apoE·DMPC complexes were dialyzed against 125 mM NH₄OAc, 2.6 mM NH₄HCO₂, and 0.26 mM Na₄EDTA, pH 7.4 (30) and negatively stained on the surface of carbon grids. Electron micrographs were made at a magnification of × 200,000 and imported into an Image 1/AT image analysis system. Particle size was analyzed by automated sizing and counting programs available on system software (Universal Imaging). A total of 350 discs from multiple areas were sampled.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Production and Characterization of Cysteine Mutants-- Based on computer modeling of the x-ray crystal structure of the 22-kDa fragment of apoE4 (22), sites for pairs of cysteine mutations were selected based on favorable interhelical distances (C-C, 4-7 Å) and steriochemistry for disulfide bond formation (31). Four pairs of potential linkage positions were selected: single disulfide mutants 30C-67C linking helices 1 and 2, 57C-112C linking helices 2 and 3, 30C-152C linking helices 1 and 4, and 122C-130C linking helices 3 and 4; double disulfide mutants linking helices 1 and 2 and 3 and 4 or helices 1 and 4 and 2 and 3; and a triple disulfide mutant linking all four helices (Fig. 1A). These disulfide bonds would either pair two helices together or, in the case of triple pair mutant, prevent the helical bundle from opening (Fig. 1B).

View larger version (68K): [in this window] [in a new window]   Fig. 1.   Engineering of intrahelical disulfide bonds by site-directed mutagenesis. Cysteine residues were introduced into the four-helix bundle, where they were predicted to form intramolecular disulfide bonds based on computer modeling of the x-ray crystal structure. A, single disulfide mutants 30C-67C linking helices 1 and 2, 57C-112C linking helices 2 and 3, 30C-152C linking helices 1 and 4, and 122C-130C linking helices 3 and 4; double disulfide mutants linking helices 1 and 2 and helices 3 and 4 and linking helices 1 and 4 and helices 2 and 3. B, a triple disulfide variant linking all four helices. Recombinant proteins were expressed in a T7 expression strain Escherichia coli BL21 (DE3), refolded, and air-oxidized to allow disulfide bond formation.

The guanidine-denatured cysteine-containing mutants were allowed to refold at low concentrations into the four-helix bundle structures under reducing conditions before allowing air oxidation to form the disulfide bonds. This procedure minimized the possibility that intermolecular disulfide bonds would form and, in cases of four and six cysteines, that the designed pairs would most likely be formed. In addition, any misfolded structures containing either intermolecular or nonexpected intramolecular disulfide linkages would probably contain free sulfhydryl groups. Free sulfhydryl groups in fact were observed in high molecular mass bands (data not shown). These proteins were removed by thiopropyl affinity chromotography. Secondary structures of apoE and the cysteine mutant 22-kDa fragments in the reduced and disulfide-linked states were determined by circular dichroism measurements. The -helical content of the mutants was 56-64%, identical to the 60% content in the apoE4 22-kDa fragment. These results demonstrate that the introduced cysteine residues and the formation of disulfide bonds do not significantly affect the overall four-helix bundle structure.

Disulfide bond formation was assessed with two independent methods: SDS-PAGE gel shifts (26) and a chemical modification method with the cysteine-specific reagent fluorescein 5-maleimide (25). Intramolecular disulfide bonds increase the mobility of proteins in SDS-PAGE compared with reduced forms (26), probably resulting from a more compact structure. With one exception, the nonreduced disulfide-linked mutants displayed greater mobility than reduced samples and apoE4 22-kDa fragment (Fig. 2A). Mutant 122C-130C did not display a shift, probably due to the small loop size of eight amino acid residues formed between the two cysteine residues (Fig. 2A). All of the double and triple disulfide bond mutants showed similar shifts (data not shown). The formation of disulfide bonds was verified by lack of reactivity with fluorescein 5-maleimide. None of the nonreduced mutants reacted with the fluorescein 5-maleimide (Fig. 2B), demonstrating the absence of free sulfhydryl groups in these proteins.

View larger version (31K): [in this window] [in a new window]   Fig. 2.   Demonstration of disulfide bond formation. Mutant proteins (5 µg), in both their free cysteine and disulfide-linked forms, were incubated with equal volumes of SDS loading buffer (62.5 mM Tris-HCl, pH 6.8, 25% glycerol, and 2% SDS) containing 20 mM fluorescein 5-maleimide at room temperature for 30 min and analyzed by a nonreducing 10-20% SDS-PAGE. A, electrophoretic mobility of the reduced and disulfide-linked mutants, stained with Coomassie Blue R-250 to visualize all of the protein bands. B, fluorescence of proteins containing free sulfhydryl groups under UV illumination.

Release of Fluorescent Dye Encapsulated in DMPC Vesicles-- Phase transition release of a self-quenching fluorescent dye, such as CF, has been used to monitor interactions of apolipoproteins with phospholipid vesicles (32). When CF is encapsulated in phospholipid vesicles at a sufficiently high concentration (~100 mM), its fluorescence is self-quenched (32). Interaction of apolipoproteins with the phospholipid vesicles disrupts the vesicles, releasing and diluting the CF, which results in a fluorescent signal. Therefore, the initial interaction between protein and lipid can be assessed from the increase in fluorescence.

The addition of both apoE4 and disulfide-linked 22-kDa mutants resulted in a rapid release of the encapsulated CF (Fig. 3). Although we cannot rule out the possibility of slight differences among the mutants, these results indicate that pairing of the helices or even tethering the whole helical bundle (in the case of triple disulfide mutant) did not appreciably affect the ability of apoE to interact with DMPC vesicles. The triple mutant results demonstrate that initial interaction with DMPC vesicles does not require the complete opening of the four-helix bundle.

View larger version (17K): [in this window] [in a new window]   Fig. 3.   Phase transition release of fluorescent dye encapsulated in DMPC vesicles. The ability of the apoE and the disulfide mutants to induce release of encapsulated 5-(6)-carboxyfluorescein was monitored by increasing fluorescence intensity for 1 min (x-axis) after proteins (1 µM) were added (indicated by the dotted lines) to the DMPC (60 µM) solutions at the DMPC transition temperature (23.9 °C). A-H, the cross-link's for each are denoted at the bottom of each panel.

Opening of the Four-helix Bundle Is Required to Form ApoE·DMPC Complexes-- The triple disulfide mutant, which effectively prevents the four-helix bundle from opening (Fig. 1B), was designed to test the hypothesis that opening of the bundle is required to form apoE·DMPC discoidal complexes (2). Complex formation was monitored with a turbidimetric clearing assay (Fig. 4). The ability of the reduced form of the triple disulfide mutant to remodel DMPC vesicles into discs was similar to that of the apoE4 22-kDa fragment (t_1/2 = 4.5 ± 0.3 versus 4.2 ± 0.3 min), demonstrating that the cysteine substitutions did not significantly affect lipid binding. However, the triple disulfide mutant was severely defective in its ability to clear the DMPC solution (Fig. 4), supporting the hypothesis that opening of the four-helix bundle of apoE is required to form discoidal DMPC complexes.

View larger version (21K): [in this window] [in a new window]   Fig. 4.   The ability of the triple disulfide mutant to clear DMPC solutions. The reduction in turbidity of DMPC vesicles was monitored as a function of time following the addition of buffer only, the apoE4 22-kDa fragment, and nonreduced and reduced triple disulfide mutants. Incubations were performed at 23.9 °C at a 1:1 (w/w) ratio of apoE to DMPC.

Initial Opening of the Four-helix Bundle Occurs in a Preferred Manner Followed by Helical Rearrangement-- To investigate the possibility that helices remain paired in binding to DMPC, two groups of pairing were examined: group I, helices 1 and 2 and helices 3 and 4 remained paired; group II, helices 1 and 4 and helices 2 and 3 remained paired. Both groups were tested for their lipid binding abilities. In a reduced form, all of the mutants displayed identical abilities as the apoE4 22-kDa fragment to clear DMPC solutions (data not shown). Although the disulfide-linked forms exhibited slower rates of clearing, group II mutants were much less efficient in clearing than group I mutants (Fig. 5, Table I). As assessed in the gel filtration assay by the amount of free unbound protein, the ability of the group I mutants to form complexes was more similar to that of the apoE4 22-kDa fragment (Fig. 6, Table I). The group II mutants had significantly more unbound protein. Taken together, these results demonstrated that helices 1 and 2 and helices 3 and 4 are likely to remain in close proximity following the initial binding of apoE to the DMPC vesicles in the early stage of complex formation.

View larger version (29K): [in this window] [in a new window]   Fig. 5.   Comparison of the abilities of disulfide mutants to form apoE·DMPC discs. The turbidimetric clearing assays were performed as described in the legend to Fig. 4. Single and double mutants are grouped as favoring the pairing of helices 1 and 2 and/or 3 and 4 (group I) or helices 1 and 4 and/or 2 and 3 (group II).

View larger version (18K): [in this window] [in a new window]   Fig. 6.   Gel filtration chromatography of apoE·DMPC discoidal complexes. Superdex 200 gel filtration chromatography was used to separate apoE·DMPC discoidal complexes from unbound protein. ApoE (100 µg) and DMPC (375 µg) vesicles were incubated at 23.9 °C for 10 h. Incubation mixtures (100 µl) were eluted at 0.5 ml/min with 20 mM Tris-HCl, pH 7.4, 150 mM NaCl, and 1 mM EDTA. A, apoE4 22-kDa fragment. B, group I mutants pairing helices 1 and 2 and/or helices 3 and 4. C, group II mutants pairing helices 1 and 4 and/or helices 2 and 3.

However, gel filtration analysis demonstrated that all of the disulfide-linked mutants formed larger discs than did the apoE4 22-kDa fragment (Fig. 6, Table I), suggesting that nonrestricted helices undergo additional rearrangement following the initial bundle opening. The larger disc sizes of the disulfide mutants were confirmed by negative staining electron microscopy (Fig. 7). Disc diameters were 120-165% of these formed with the apoE4 22-kDa fragment complexes (19.8-32.5 versus 16.7 ± 1.9 nm).

View larger version (48K): [in this window] [in a new window]   Fig. 7.   Size comparison of negatively stained apoE·DMPC complexes. A, apoE4 22-kDa fragment. B, double disulfide mutant 30C-67C/122C-130C. C, double disulfide mutant 30C-152C/57C-112C. The apoE·DMPC particles were dialyzed in 125 mM ammonium acetate, 2.6 mM ammonium carbonate, and 0.26 mM EDTA, pH 7.4, and stained on carbon film grids. Photographs are shown at a magnification of × 200,000. Particle size was determined as described under "Experimental Procedures."

To examine the suggestion that the helices underwent a further rearrangement in assuming their final conformation on discs, the complexes were isolated by gel filtration, treated with reducing reagents (1% BME and 100 mM dithiothreitol), and repassed on a gel filtration column. Disulfide bond reduction converted the large mutant complexes into normal size complexes (Fig. 8), demonstrating that the helices are capable of reorganizing while bound to DMPC. These results suggest that following initial interaction with DMPC, in which the helices are paired, a further rearrangement occurs as apoE assumes its final conformation on the discs.

View larger version (21K): [in this window] [in a new window]   Fig. 8.   Effect of reducing disulfide bonds on the size of apoE·DMPC particles. Large apoE·DMPC disulfide mutant discs (solid lines) were isolated, incubated with reducing reagents, and eluted from a Superdex 200 gel filtration column (dashed lines) (as described in the legend to Fig. 7) to determine the change in their relative sizes. A-F, the specific cross-links are indicated for each.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Using engineered interhelical disulfide bonds, we investigated the molecular interactions and conformational changes of the four-helix bundle of apoE in binding to DMPC to form discoidal particles. The data support a multistep process that begins with the initial interaction of apoE with the DMPC vesicle surface. This step is followed by the opening of the four-helix bundle as the protein begins to remodel the lipid vesicles into discs. In the final step, the helices undergo additional rearrangement as they assume their final conformation on a disc.

Previous studies demonstrated that human apoA-1 adopts a partially unfolded, molten globular-like state that was suggested to play an important role in lipid binding (33). Recently, it was suggested that a partially unfolded conformation of apoLpIII, in which the hydrophobic core of the bundle is hydrated, is associated with increased lipid binding (34). Although the final conformations of both apoLpIII (11, 35, 36) and apoE (15) bound to lipid appear to involve open helical bundle conformations, it was not clear whether apoE or apoLpIII undergoes a major unfolding in the initial interaction with lipid. Two possibilities have been suggested for apoLpIII. First, a short stretch of hydrophobic residues in the loops connecting helices 1 and 2 and helices 3 and 4 initiates binding (11, 36, 37). Second, electrostatic interactions between charged residues on the protein exterior and phospholipid head groups are responsible for the initial interaction (38). The results from our fluorescence release studies (Fig. 3) in which the triple disulfide mutant of apoE 22-kDa fragment effectively prevented the four-helix bundle from completely opening, indicated that exposing the entire hydrophobic core is not required for initial lipid interaction in the case of apoE. However, since the 122-130 link is at the base of helices 3 and 4 (Fig. 1), which would allow some flexibility at the opposite end of the bundle, we cannot completely rule out the possibility of partial exposure of hydrophobic residues at this end of the molecule in the initial interaction.

Recent x-ray crystallography studies of multiple apoE crystal forms revealed a conformational flexibility at one end of the four-helix bundle (39). The flexible regions included the loop linking helices 2 and 3 (residues 78-90, referred as the 80s loop) and distortions in the ends of helices 3 and 4. It was suggested that this highly flexible end of the bundle plays an important role in initiation of lipid binding and initial unfolding of the bundle (39). Interestingly, the 80s loop is enriched in negatively charged acidic residues (Glu⁷⁹, Glu⁸⁰, Glu⁸⁷, and Glu⁸⁸); one possibility is that these negative charges initiate interaction with positive charges on the phospholipid headgroups. Alternatively, the conformational flexibility at this end of apoE might also partially expose the hydrophobic core and trigger the bundle opening. Thus, at this point, it is not clear if the initial interaction is mediated by ionic or hydrophobic interactions or both. Resolving this issue will require additional studies, including site-directed mutagenesis of the 80s loop.

According to the proposed hypothesis for the interaction of apoE with DMPC, the four-helix bundle undergoes a lipid-triggered conformational opening (2, 15). The fact that the triple disulfide mutant could not remodel DMPC vesicles into discoidal complexes (Fig. 4) provides direct evidence that opening of the bundle is required for disc formation. The results with the single and double disulfide mutants (Figs. 6 and 7) indicate that, at least initially, the opening happens in a manner in which helices 1 and 2 and helices 3 and 4 remain paired. However, the fact that the DMPC discs formed by these mutants are larger than those formed by apoE4 (Figs. 6 and 7, Table I) indicates an inability to achieve a final lipid-bound conformation, suggesting that the paired helices undergo additional rearrangement in assuming the final conformation. This additional rearrangement of the helices was demonstrated when the large sized discs formed by the single and double disulfide mutants were reduced and shown to convert to normal size (Fig. 8). Thus, restricting helical movement or influencing structure by inserting additional residues, as occurs in apoE-Leiden (29), alters disc size. Thus, subtle differences in apoE can have a dramatic effect on disc size, suggesting that the final conformation of apoE is an important determinant of disc size.

Although our studies indicate that helical reorganization is a major component in the multiple steps of interaction of apoE with DMPC, they do not directly address the issue of the final conformation of apoE on DMPC discs. Thus, the results do not allow us to distinguish between the "picket fence" (18) and the "belt" models (19). The final reorganization of helices is compatible with either model. Resolution of this issue will require additional studies, perhaps including crystallization of an apoE·phospholipid complex.

	ACKNOWLEDGEMENTS

We thank Drs. Stanley Rall, Robert Raffaï, Paul Hopkins, and Brent Segelke for helpful discussions; Dr. David Sanan and Dale Newland for negative staining electron microscopy; Yvonne Newhouse for technical assistance; John W. C. Carroll for graphic arts; Brian Auerbach for manuscript preparation; and Stephen Ordway and Gary Howard for editorial assistance.

	FOOTNOTES

^* This work was supported in part by National Institutes of Health Grant HL41633.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

^¶ To whom correspondence should be addressed: Gladstone Institute of Cardiovascular Disease, P.O. Box 419100, San Francisco, CA 94141-9100. Tel.: 415-826-7500; Fax: 415-285-5632; E-mail: kweisgraber@gladstone.ucsf.edu.

Published, JBC Papers in Press, May 8, 2000, DOI 10.1074/jbc.M003508200

	ABBREVIATIONS

The abbreviations used are: apoE, apolipoprotein E; apoLpIII, apolipophorin III; BME, -mercaptoethenol; CF, 5-(6)-carboxylfluorescein; DMPC, dimyristoylphosphatidylcholine; LDL, low density lipoprotein; PAGE, polyacrylamide gel electrophoresis.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				C. D. Blanchette, R. Law, W. H. Benner, J. B. Pesavento, J. A. Cappuccio, V. Walsworth, E. A. Kuhn, M. Corzett, B. A. Chromy, B. W. Segelke, et al. Quantifying size distributions of nanolipoprotein particles with single-particle analysis and molecular dynamic simulations J. Lipid Res., July 1, 2008; 49(7): 1420 - 1430. [Abstract] [Full Text] [PDF]

				M. J. Gunzburg, M. A. Perugini, and G. J. Howlett Structural Basis for the Recognition and Cross-linking of Amyloid Fibrils by Human Apolipoprotein E J. Biol. Chem., December 7, 2007; 282(49): 35831 - 35841. [Abstract] [Full Text] [PDF]

				M. R. Tubb, R. A. G. D. Silva, K. J. Pearson, P. Tso, M. Liu, and W. S. Davidson Modulation of Apolipoprotein A-IV Lipid Binding by an Interaction between the N and C Termini J. Biol. Chem., September 28, 2007; 282(39): 28385 - 28394. [Abstract] [Full Text] [PDF]

				C. A. Peters-Libeu, Y. Newhouse, S. C. Hall, H. E. Witkowska, and K. H. Weisgraber Apolipoprotein E*dipalmitoylphosphatidylcholine particles are ellipsoidal in solution J. Lipid Res., May 1, 2007; 48(5): 1035 - 1044. [Abstract] [Full Text] [PDF]

				M. Lamant, F. Smih, R. Harmancey, P. Philip-Couderc, A. Pathak, J. Roncalli, M. Galinier, X. Collet, P. Massabuau, J.-M. Senard, et al. ApoO, a Novel Apolipoprotein, Is an Original Glycoprotein Up-regulated by Diabetes in Human Heart J. Biol. Chem., November 24, 2006; 281(47): 36289 - 36302. [Abstract] [Full Text] [PDF]

				S. D. Tetali, M. S. Budamagunta, J. C. Voss, and J. C. Rutledge C-terminal interactions of apolipoprotein E4 respond to the postprandial state J. Lipid Res., July 1, 2006; 47(7): 1358 - 1365. [Abstract] [Full Text] [PDF]

				C. A. Peters-Libeu, Y. Newhouse, D. M. Hatters, and K. H. Weisgraber Model of Biologically Active Apolipoprotein E Bound to Dipalmitoylphosphatidylcholine J. Biol. Chem., January 13, 2006; 281(2): 1073 - 1079. [Abstract] [Full Text] [PDF]

				D. M. Hatters, M. S. Budamagunta, J. C. Voss, and K. H. Weisgraber Modulation of Apolipoprotein E Structure by Domain Interaction: DIFFERENCES IN LIPID-BOUND AND LIPID-FREE FORMS J. Biol. Chem., October 7, 2005; 280(40): 34288 - 34295. [Abstract] [Full Text] [PDF]

				M. Futamura, P. Dhanasekaran, T. Handa, M. C. Phillips, S. Lund-Katz, and H. Saito Two-step Mechanism of Binding of Apolipoprotein E to Heparin: IMPLICATIONS FOR THE KINETICS OF APOLIPOPROTEIN E-HEPARAN SULFATE PROTEOGLYCAN COMPLEX FORMATION ON CELL SURFACES J. Biol. Chem., February 18, 2005; 280(7): 5414 - 5422. [Abstract] [Full Text] [PDF]

				H. Saito, P. Dhanasekaran, F. Baldwin, K. H. Weisgraber, M. C. Phillips, and S. Lund-Katz Effects of Polymorphism on the Lipid Interaction of Human Apolipoprotein E J. Biol. Chem., October 17, 2003; 278(42): 40723 - 40729. [Abstract] [Full Text] [PDF]

				V. Raussens, C. M. Slupsky, B. D. Sykes, and R. O. Ryan Lipid-bound Structure of an Apolipoprotein E-derived Peptide J. Biol. Chem., July 3, 2003; 278(28): 25998 - 26006. [Abstract] [Full Text] [PDF]

				H. Saito, P. Dhanasekaran, D. Nguyen, F. Baldwin, K. H. Weisgraber, S. Wehrli, M. C. Phillips, and S. Lund-Katz Characterization of the Heparin Binding Sites in Human Apolipoprotein E J. Biol. Chem., April 18, 2003; 278(17): 14782 - 14787. [Abstract] [Full Text] [PDF]

				J. A. Morrow, D. M. Hatters, B. Lu, P. Hochtl, K. A. Oberg, B. Rupp, and K. H. Weisgraber Apolipoprotein E4 Forms a Molten Globule. A POTENTIAL BASIS FOR ITS ASSOCIATION WITH DISEASE J. Biol. Chem., December 20, 2002; 277(52): 50380 - 50385. [Abstract] [Full Text] [PDF]

				M. L. Segall, P. Dhanasekaran, F. Baldwin, G. M. Anantharamaiah, K. H. Weisgraber, M. C. Phillips, and S. Lund-Katz Influence of apoE domain structure and polymorphism on the kinetics of phospholipid vesicle solubilization J. Lipid Res., October 1, 2002; 43(10): 1688 - 1700. [Abstract] [Full Text] [PDF]

				V. Raussens, C. M. Slupsky, R. O. Ryan, and B. D. Sykes NMR Structure and Dynamics of a Receptor-active Apolipoprotein E Peptide J. Biol. Chem., August 2, 2002; 277(32): 29172 - 29180. [Abstract] [Full Text] [PDF]

				Z.-S. Ji, R. D. Miranda, Y. M. Newhouse, K. H. Weisgraber, Y. Huang, and R. W. Mahley Apolipoprotein E4 Potentiates Amyloid beta Peptide-induced Lysosomal Leakage and Apoptosis in Neuronal Cells J. Biol. Chem., June 7, 2002; 277(24): 21821 - 21828. [Abstract] [Full Text] [PDF]

				C. P. Libeu, S. Lund-Katz, M. C. Phillips, S. Wehrli, M. J. Hernaiz, I. Capila, R. J. Linhardt, R. L. Raffai, Y. M. Newhouse, F. Zhou, et al. New Insights into the Heparan Sulfate Proteoglycan-binding Activity of Apolipoprotein E J. Biol. Chem., October 12, 2001; 276(42): 39138 - 39144. [Abstract] [Full Text] [PDF]

S. Lund-Katz, S. Wehrli, M. Zaiou, Y. Newhouse, K. H. Weisgraber, and M. C. Phillips Effects of polymorphism on the microenvironment of the LDL receptor-binding region of human apoE J. Lipid Res., June 1, 2001; 42(6): 894 - 901. [Abstract] [Full Text]