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

J. Biol. Chem., Vol. 275, Issue 39, 30458-30464, September 29, 2000

This Article Abstract Full Text (PDF) All Versions of this Article: 275/39/30458    most recent M002583200v1 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 Jeon, H. Articles by Shipley, G. G. Search for Related Content PubMed PubMed Citation Articles by Jeon, H. Articles by Shipley, G. G. Social Bookmarking                      What's this?

Vesicle-reconstituted Low Density Lipoprotein Receptor

VISUALIZATION BY CRYOELECTRON MICROSCOPY^*

Hyesung Jeon and G. Graham Shipley^§

From the Departments of Biophysics and Biochemistry, Center for Advanced Biomedical Research, Boston University School of Medicine, Boston, Massachusetts 02118

Received for publication, March 27, 2000, and in revised form, June 30, 2000

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The low density lipoprotein (LDL) receptor is a key protein for maintaining cellular cholesterol homeostasis by binding cholesterol-rich lipoproteins through their apoB and apoE apoproteins. The LDL receptor is a transmembrane glycoprotein of M_r ~115 kDa; based on its primary sequence, five distinct structural domains have been identified (Yamamoto, T., Davis, C. G., Brown, M. S., Schneider, W. J., Casey, M. L., Goldstein, J. L., and Russell, D. W. (1984) Cell 39, 27-38). As a first step toward providing a structural description of the intact LDL receptor, the receptor has been purified from bovine adrenal cortices, reconstituted into unilamellar egg yolk phosphatidylcholine vesicles, and imaged using cryoelectron microscopy (cryoEM). CryoEM has the advantage of providing images of the reconstituted LDL receptor in its frozen, fully hydrated state. LDL receptor molecules were visualized as elongated, stick-like projections from the vesicle surface with maximum dimensions ~120-Å length by ~45-Å width. In some of the images, a short arm (or arms) was visible at the distal end of the stick-like projections. The LDL receptor was labeled via accessible free cysteine residues, probably including that corresponding to Cys-431 of the known full-length sequence of the human LDL receptor. The accessible cysteine was demonstrated using a maleimide-biotin·streptavidin conjugate and confirmed by labeling with monomaleimido-Nanogold. Images obtained by cryoEM showed that the extracellular stick-like domain of the reconstituted LDL receptor was labeled by Nanogold. This combined cryoEM-Nanogold labeling study has provided the first low resolution structural images of the reconstituted, full-length bovine LDL receptor.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Low density lipoprotein (LDL)¹ is the most abundant cholesterol-containing lipoprotein in human plasma (1). Since an elevated level of blood cholesterol is a major risk factor in developing atherosclerosis, the LDL receptor, which is responsible for the uptake of LDL into cells, has been studied extensively, notably by Brown and Goldstein (2). The LDL receptor is located in coated pit regions on the cell surface; it binds cholesterol-rich lipoproteins (LDL and -VLDL) via receptor binding sub-domains of their apoB and apoE proteins, and the receptor-ligand complex is internalized by endocytosis. Acidification of the internal compartment of the endocytic vesicle leads to dissociation of the LDL receptor-LDL complex; the LDL receptor is recycled to the cell surface, and LDL is delivered to the lysosome. Release of cholesterol triggers a variety of intracellular metabolic and regulatory events, notably the regulation of cholesterol synthesis and expression of the LDL receptor and other genes (1-4). It should be stated that these studies of the LDL receptor were critical not only in providing a rationale for cholesterol homeostasis but also for laying the foundations for our understanding of intracellular trafficking mechanisms.

The amino acid sequence of the LDL receptor has been reported for several species including human (5), rabbit (6), rat (7), mouse (8), frog (9), and hamster (10); a partial sequence (its C-terminal 25%) of the bovine LDL is also available (11). These sequences exhibit significant sequence homology and confirm the presence of five distinguishable domains originally described for the human LDL receptor (5): the N-terminal ligand binding domain, the epidermal growth factor (EGF) precursor homology domain, the O-linked sugar domain, the transmembrane domain, and the C-terminal cytoplasmic domain. The ligand binding domain consists of the N-terminal 292 residues and includes 7 homologous repeats of a cysteine-rich 40-residue sequence (5). Each 40-residue repeat has 6 internally disulfide-linked cysteine residues, and the ligands for the LDL receptor, LDL and -VLDL, appear to require different combinations of these repeats for binding (12). The EGF precursor homology domain has three cysteine-rich growth factor-like repeats at the N and C termini and a spacer sequence containing the YWTD sequence repeated approximately every 40-60 amino acid residues (13). The O-linked sugar domain of 58 amino acids contains 18 serine or threonine residues and is immediately external to the membrane-spanning domain of 22 hydrophobic amino acids (5). The C-terminal 50-residue cytoplasmic domain is highly conserved among species (14), and it is thought that this domain is responsible for the receptor oligomerization and receptor-clustering in coated pits (15, 16).

The cysteine-rich repeat of the ligand binding domain, the growth factor-like repeat and the YWTD spacer sequence of EGF precursor homology domain, and the NPXY sequence of the cytoplasmic domain are shared by all members of the LDL receptor gene family (2). In addition, some of these sequences are also found in proteins such as complement factors and developmental proteins (13, 17, 18). Therefore, establishing the relationship between structure and function of the LDL receptor domains could also be helpful in predicting the functions of the shared modules for the other related proteins.

Studies of the full-length LDL receptor using prediction algorithm and circular dichroism methods have provided information on its secondary structure (19, 20). Based on CD spectroscopy studies (20), the detergent-solubilized bovine LDL receptor is particularly rich in -structure (~40%) and coil (~40%), with lesser amounts of -helix (~20%). Lacking the complete sequence of the bovine LDL receptor, we analyzed using predictive algorithms the secondary structure of both the whole human LDL receptor and its five domains (20). For the complete receptor, again a high -structure probability (~65%; ~40% -turn, ~25% -sheet)) was predicted with lesser amounts of -helix (~20%); similar results were obtained by De Loof et al. (19). In addition, we have analyzed the secondary structure of the five individual domains of the human LDL receptor (20). Interestingly, the ligand binding domain is predicted to be rich in -turns (~50%) and -sheet (~20%), with less (~15%) -helix; this agrees quite well with the experimentally determined structures of cysteine-rich repeats, which all contain an N-terminal -hairpin followed by several -turns (Refs. 21-24, see below). Similarly, the EGF precursor homology domain is predicted to be >60% -structure, in good agreement with recent modeling studies by Springer (25) suggesting that this YWTD-rich domain of the LDL receptor (and other endocytic receptors) forms a compact -propeller structure built from six -sheets.

Recent studies using NMR and x-ray crystallography focus on the structures of the first, second, and fifth cysteine-rich 40-residue repeats of the ligand binding domain (21-24). Similar conformations consisting of a -hairpin followed by -turns motif are reported. Interestingly, the x-ray crystallographic study of repeat-5 shows a calcium binding loop that utilizes several of the conserved acidic residues (previously thought to contribute to interactions with basic residues of the apoB or apoE ligand), thus sequestering these acidic residues away from the surface (24). Recently, structural studies of two concatamers, repeat-1/repeat-2 (26) and repeat-5/repeat-6 (27), conclude that there is little or no inter-module interaction, i.e. the two modules are essentially structurally independent.

Our approach has been somewhat different in that we have focused on structural studies, albeit at lower resolution, of the full-length, intact functional LDL receptor. Thus, we have (i) purified the LDL receptor from bovine adrenal cortex, (ii) reconstituted the LDL receptor into lipid vesicles, and (iii) studied the reconstituted LDL receptor in the frozen, hydrated state by cryoelectron microscopy (cryoEM) and gold-labeling methods.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

LDL Receptor Purification-- The purification protocol for the LDL receptor is based on the original method reported by Schneider et al. (28) and subsequent modifications (20). Briefly, cortices from ~50 adrenal glands were homogenized, and a membrane pellet was isolated. After detergent solubilization with Triton X-100, ion exchange column chromatography was performed with a HiLoad 16/10 Q-Sepharose column using a fast protein liquid chromatography system. The column was washed with 75 mM NaCl in buffer (50 mM Tris-maleate, 2 mM CaCl₂, 40 mM -octylglucoside (OG), 1 mM phenylmethylsulfonyl fluoride, 0.01 mM Leupeptin, pH 6.0), then eluted with a linear NaCl gradient of 75 to 375 mM in buffer at a flow rate of 1 ml/min. Five-milliliter fractions were collected, and the Pierce protein assay based on the Bradford method (29) was performed for all aliquots to generate a protein profile (Fig. 1A). The dot-blot assay using the monoclonal IgG-C7 antibody against the N terminus of the receptor was also performed (Fig. 1B). Fractions containing the LDL receptor were applied to an IgG-C7-Sepharose affinity column, and the eluted fractions were analyzed by silver stain SDS/PAGE and Western blot analysis using both IgG-C7 and the C-terminal antibody IgG-4A4. Aliquots of affinity column chromatography fractions were assayed for protein amount, and the A(III) fraction was used for further study.

View larger version (31K): [in this window] [in a new window]   Fig. 1.   Ion exchange column chromatography using the Q-Sepharose/fast protein liquid chromatography system for purification of the LDL receptor. A, the protein profile was generated using the Pierce protein assay after elution by a linear NaCl gradient of 75 to 375 mM. B, dot-blot assay using monoclonal IgG-C7 antibody identified the fractions containing LDL receptor. Fractions 13-19 were collected for the next step.

Preparation of Unilamellar Vesicles-- Egg yolk phosphatidylcholine (EYPC) in chloroform was transferred into a 25-ml round-bottom flask, and the solvent removed on a rotary evaporator. Dried EYPC (~2 mg) was redissolved in 1 ml of chloroform:methanol (2:1) solution and rotary-evaporated twice to make a thin, even lipid film on the bottom of the flask. The EYPC film was lyophilized overnight at 10 microtorr and protected from light with aluminum foil. The lyophilized EYPC film was hydrated with 1 ml of hydration buffer (10 mM HEPES, 2 mM CaCl₂, pH 7.0). The EYPC dispersion was rapidly frozen in a liquid nitrogen bath for 30 s then thawed immediately in a water bath at ~50 °C and this freeze/thaw step repeated 4 times. The EYPC dispersion was passed through 0.1-µm pore membrane 10 times using an extruder (Lipex Biomembranes Inc.) to make large unilamellar lipid vesicles (30). The extruded EYPC vesicle solution was used for LDL receptor reconstitution.

Reconstitution of the LDL Receptor-- The purified LDL receptor was concentrated to 1.0-1.6 mg/ml using Centricon 30 concentrators (Amicon), and the final concentration of the receptor was determined (31) using bovine serum albumin solution as a standard. The -OG-solubilized LDL receptor, EYPC vesicles, detergent (-OG, 500 mM), and dialysis buffer (50 mM HEPES, 2 mM CaCl₂, pH 6.15) were mixed for a final concentration of the LDL receptor of 0.10 mg/ml, 1.0 mg/ml EYPC, and 19.7 mM -OG. The solution was placed into the wells of a microdialyzer and dialyzed against 1 liter of gradient detergent concentration 20 to 0 mM -OG in dialysis buffer. Finally, the sample was dialyzed against an additional 1 liter of dialysis buffer.

Cryoelectron Microscopy-- Specimens for cryoEM were prepared by rapid freezing the blotted sample on holey grids (32). Typically, 6 µl of sample containing 0.1 mg/ml reconstituted LDL receptor and 1 mg/ml lipid vesicle was used on each grid. Excess liquid was blotted from the back side of the grid, and the grid was immediately plunged into liquid ethane, cooled in a liquid nitrogen bath. The plunging process was carried out at room temperature in a humidified chamber. Samples were transferred to a Gatan cryo holder (Gatan, Inc., Pittsburgh) using a Gatan cryo transfer station and maintained at a temperature less than 160 °C. Samples were viewed using a Philips CM12 transmission electron microscope under low electron dose conditions to minimize irradiation and sample damage. Electron micrographs were recorded on Kodak Electron Image Film SO-163 at 75,000× magnification. Micrographs were recorded at 1.7 µm under-focus for the best visualization of protein and lipid vesicles. For the Nanogold-labeled proteins, electron micrographs were recorded at both 1.7 µm and 0.5 µm defocus to optimize visualizing protein and gold, respectively. Films were developed in concentrated D19 for 12 min. Electron micrographs recorded on films were digitized and scanned. For the individual particle selection and image manipulation, the SPIDER/WEB software package was used (33) after the conversion of file format from the scanned images.

Biotin-BMCC Maleimide Labeling-- The purified bovine LDL receptor was labeled using the biotin-BMCC (1-biotinamido-4-[4'-(maleimidomethyl)cyclohexanecarboxamido]butane) reagent (Pierce) to confirm that the bovine LDL receptor has free, accessible cysteine residues. Biotin-BMCC was suspended in Me₂SO to a concentration of 50 mM. Five micrograms of LDL receptor dialysis buffer (50 mM HEPES, 2 mM CaCl₂, pH 6.15) and 1 µl of 50 mM biotin-BMCC were mixed to give a final concentration of 0.05 µg/µl LDL receptor and 0.5 mM biotin-BMCC. Samples were incubated at room temperature for 1 h, then run on a 7.5% homogeneous polyacrylamide gel using the Laemmli buffer system (34). The gel was transferred electrophoretically to a nitrocellulose membrane, and the membrane was blocked with 5% milk in wash buffer (10 mM Tris-Cl, pH 8, 150 mM NaCl, 0.05% Tween 20) at 4 °C. The membrane was washed, then incubated with streptavidin-horseradish peroxidase for 30 min at room temperature. The signal was detected by using an enhanced chemiluminescence system and exposing the membrane to film after washing. The peroxidase enzyme was inactivated by incubating the membrane with 15% hydrogen peroxide in phosphate-buffered saline buffer for 30 min. After the same washing steps, the membrane was incubated with IgG-C7 and IgG-4A4 for an hour, and the Western blot method was followed.

Nanogold Labeling-- Monomaleimido-Nanogold (Nanoprobes, Stony Brook, NY) containing a 1.4-nm-diameter gold particle was used to label free cysteine residues of the LDL receptor. Immediately after solubilization (20 µl Me₂SO or isopropanol, then 180 µl of deionized water), five molar equivalents of monomaleimido-Nanogold solution were added to the reconstituted LDL receptor solution and incubated at 4 °C for approximately 18 h. Unbound gold particles were removed by dialysis against 2 liters of the same dialysis buffer used in the reconstitution of the LDL receptor for 48 h with a flow rate of 20-25 ml/hr. After SDS-PAGE, an LI silver stain, specifically reactive to gold particles, was used to detect gold-labeled protein bands and to confirm Nanogold-labeling of the LDL receptor. LI silver stain is a light-insensitive silver enhancement system for visualizing Nanogold particles. The deposition of dense silver particles on Nanogold is enhanced during the initial reaction period.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Purification of the Bovine LDL Receptor-- The detergent-solubilized proteins from the membrane pellet were purified by ion exchange column chromatography with a change of the detergent Triton X-100 to -octylglucoside (-OG). The eluted fractions from the affinity column were analyzed by silver stain and Western blot following gel electrophoresis. As expected for the LDL receptor (Ref. 35; see also Ref. 20), molecular weight differences were observed for fraction A(III) samples examined by SDS-PAGE under reducing (~160 kDa) and non-reducing (~130 kDa) conditions (Fig. 2A). Results from the Western blot of the A(III) fraction under reducing conditions (Fig. 2A, right; lane 3) demonstrated that -mercaptoethanol prevented antibody binding to the LDL receptor. The A(III) fraction contained most of the highly purified LDL receptor and was used for further studies.

View larger version (57K): [in this window] [in a new window]   Fig. 2.   A, silver stain and Western blot analysis of affinity column chromatography fractions. A(I), A(II), A(III), and A(IV) are the eluted fractions, and b-ME indicates -mercaptoethanol. Samples were run on homogeneous 7.5% gels. LDL-R, LDL receptor. B, antibody binding assay for the purified LDL receptor. N-terminal IgG-C7 and C-terminal IgG-4A4 monoclonal antibodies were used for Western blot analysis under non-reducing conditions.

The activity of the purified LDL receptor was confirmed by Western blots under non-reducing conditions using IgG-C7 (36) and IgG-4A4 (11) monoclonal antibodies against the N- and C terminus of the receptor, respectively (Fig. 2B). Western blots using both antibodies showed the LDL receptor band at the same location as in the silver-stained gel. IgG-4A4 showed a weaker band against the same amount of the receptors compared with IgG-C7. The yield of purified intact LDL receptor was ~600 µg/50 adrenal glands.

Reconstituted LDL Receptor-- The purified LDL receptor was reconstituted by adding the detergent-solubilized LDL receptor to the pre-formed vesicles as described under "Experimental Procedures." The use of pre-formed 100-nm-diameter EYPC vesicles gave the most efficient reconstitution, and the reconstitution protocol was optimized by varying the lipid to protein ratio and the lipid and protein concentrations. The rate and period of the detergent dialysis step were also adjusted to obtain slow reconstitution conditions. A gradient detergent dialysis (20 to 0 mM -OG) step was developed that gave the most efficient, reproducible reconstitution of the LDL receptor. This method yielded uniform diameter (60-100 nm) unilamellar vesicles containing the LDL receptor.

CryoEM images of reconstituted vesicles containing the LDL receptor are shown at low magnification in Fig. 3. Most of the vesicles are unilamellar, and the vesicle diameter ranges from 60 to 100 nm. LDL receptors visualized by cryoelectron microscopy (examples are circled in Fig. 3) appear as elongated stem-like projections from the vesicle surface with a more electron-dense domain distal to the bilayer surface. The receptors appear to adopt an orientation primarily, but not exclusively, on the outward-facing membrane surface (for examples, see Fig. 3, open circles). An example of an inward-facing LDL receptor is also shown (Fig. 3, asterisked circle). In some cases, two (or more) receptors are viewed side-by-side (Fig. 3, circles labeled 2).

View larger version (153K): [in this window] [in a new window]   Fig. 3.   CryoEM images of the reconstituted LDL receptor preserved in vitreous ice. Individual receptors are circled (not all receptors are circled). 2 indicates a circle including two receptors. The asterisked circle indicates the reconstituted receptor oriented on the inside surface of the vesicle.

Electron micrographs recorded on film were digitized and scanned. For the individual particle selection and image manipulation, the SPIDER/WEB software package was used (33). A montage showing examples of the frozen, hydrated LDL receptor at the vesicle surface is shown in Fig. 4. Reconstituted LDL receptors visualized by cryoEM can be observed in three basic shapes; (i) stick-like receptors in which the whole extracellular length is the stem length (e.g. Fig. 4; A3, C1, C2, D3; image C1 is also shown at higher magnification in the upper right panel); (ii) bent stick-like receptors, which have one arm extending distally from the stem (e.g. Fig. 4; B1, B2, B3, C3, D1; image C3 is also shown at higher magnification in the lower right panel); and (iii) Y-shaped receptors, which have two arms branching from the stem (e.g. Fig. 4; A1, A2, D2). All three receptor shapes were observed for both outward- and inward-facing receptors. The three observed receptor images may result from different orientations of a unique conformation of the receptor with respect to the viewing plane. It is also possible that some conformational flexibility of the LDL receptor contributes to the variability of the observed images. Histogram measurements of the length and width of the receptor images (data not shown) indicate maximum dimensions of 120 Å and 45 Å, respectively; for bent-stick and Y-shaped receptors the arm length was ~70 Å. Assuming that the visualized images represent the extracellular domain of the LDL receptor, calculations of the volume based on these dimensions are in reasonably good agreement with the volume calculated from the amino acid sequence of the extracellular domain of the human LDL receptor.

View larger version (107K): [in this window] [in a new window]   Fig. 4.   Individual images of reconstituted LDL receptors visualized by cryoEM. Images C1 (upper) and C3 (lower) are shown at 50× magnification in the right panels. In each panel, the arrow points to the LDL receptor projecting from the bilayer surface. The combination of letter/number is used to identify specific panels cited under "Results."

Nanogold-labeled Receptor-- The human LDL receptor has only three free cysteine residues out of a total of 63 cysteine residues, the rest being involved in the numerous disulfide bonds of the cysteine-rich repeats of the ligand binding domain and the EGF precursor homology domain (5). A comparison of the human receptor sequence and the partial bovine receptor sequence revealed that a free cysteine in the transmembrane domain of the human receptor is an alanine in the bovine receptor (37). Due to the high homology of the human LDL receptor with the bovine receptor, we examined whether free cysteine residues of the bovine LDL receptor could be identified. The presence of free, accessible cysteine residues on the bovine LDL receptor was experimentally demonstrated using a maleimide-biotin·streptavidin conjugate. The detergent-solubilized bovine LDL receptor was incubated with the maleimide-biotin conjugate, and the biotin conjugate bound to the LDL receptor was detected by streptavidin-linked horseradish peroxidase (Fig. 5). The LDL receptor band appeared at the same location as in Western blots visualized using the C-terminal (strong binding) and N-terminal (weaker binding) antibodies after deactivating the peroxidase. In addition, under these labeling conditions upper bands, possibly due to LDL receptor dimers, were observed in the Western blots.

View larger version (31K): [in this window] [in a new window]   Fig. 5.   Maleimide-labeling analysis of the LDL receptor. A, biotin-BMCC was incubated with the purified LDL receptor and detected using the biotin-streptavidin-horseradish peroxidase complex. B, streptavidin- horseradish peroxidase was inactivated, and Western blot analysis was performed using the IgG-C7 and IgG-4A4 antibodies.

Cysteine sulfhydryl groups are also reactive with the maleimide group of monomaleimido-Nanogold. LDL receptors reconstituted into EYPC vesicles by detergent dialysis were incubated with excess monomaleimido-Nanogold, and further dialysis was performed to remove unbound Nanogold particles. The Nanogold-labeling of the reconstituted receptor was investigated by SDS-PAGE and LI silver stain (Fig. 6). LI silver stain, specifically reactive with the gold particles, detected the labeled LDL receptor band at the same location as in the silver-stained gel (Fig. 6A). The silver-stained gel showed a stronger band for the Nanogold-labeled receptors compared with the band for the same amount of unlabeled receptor. The Western blots using both N- and C-terminal antibodies detected weaker bands after the Nanogold-labeling (Fig. 6B), presumably due to interference in antibody binding by the maleimide-bound gold particles.

View larger version (35K): [in this window] [in a new window]   Fig. 6.   Nanogold-labeling of the LDL receptor. A, LI silver stain and silver stain. The gold-specific LI silver stain detected the Nanogold-labeled LDL receptors (LDL-R) at the same location as in the silver-stained gel. B, Western blot analysis using both the N- and the C-terminal antibodies. The Nanogold-labeled receptors showed weaker bands in Western blots than unlabeled receptors. Lane C is unlabeled control LDL receptor.

CryoEM images of the Nanogold-labeled reconstituted LDL receptors were recorded at two different defocus values. The 1.7-µm and 0.5-µm defocus values were used for optimal visualization of proteins and gold particles, respectively, and a montage of paired images of the gold-labeled LDL receptor is shown in Fig. 7. At 1.7-µm defocus, similar images were observed for the Nanogold-labeled LDL receptor (Fig. 7) to those described above for the unlabeled LDL receptor (Fig. 4). The paired images optimized for visualization of gold (0.5 µm defocus) showed electron-dense, circular Nanogold particles bound primarily to the stem of the LDL receptor. Either one or two gold particles were observed bound to the extracellular domain of the LDL receptor. In general, the Nanogold particles were located at the middle of the extracellular domain, but sometimes adjacent to the membrane vesicle surface. The observed images confirmed the successful Nanogold-labeling of the reconstituted LDL receptor and showed the location of free cysteine residues.

View larger version (75K): [in this window] [in a new window]   Fig. 7.   CryoEM images of the Nanogold-labeled LDL receptors preserved in vitreous ice. Electron micrographs were recorded as defocus pairs, 1.7-µm defocus (1-6) for optimal visualization of protein and 0.5 µm defocus (1'-6') for optimal visualization of gold. For the panels 1-6 at 1.7-µm defocus, the arrow points to the Nanogold-labeled receptor.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The purified LDL receptor was successfully reconstituted into pre-formed large unilamellar EYPC vesicles by detergent dialysis (Fig. 3). Apparently, using pre-formed vesicles (38) and a slow gradient detergent dialysis method favors, to some degree at least, unidirectional (right-side out) reconstitution of membrane proteins. However, the reconstitution we have used, although highly efficient in terms of receptor incorporation, uniformity of vesicle size, retention of unilamellar vesicle structure, etc., does produce some inward-facing receptor orientations (see Figs. 3 and 4). The optimized lipid to protein ratio 10:1 (w/w) gives an average of 41 receptors per 100-nm diameter EYPC vesicle assuming that the average surface area per EYPC molecule is 0.7 nm² and the bilayer thickness is 4 nm (39). Although a smaller number of receptors in the vesicles was observed by cryoEM even under the best reconstitution conditions, it should be realized that only the LDL receptor molecules oriented at the edge of the vesicle images can be seen. Other vesicle-reconstituted receptors were probably obscured by the dense projection of the phospholipid bilayer.

Images of the LDL receptor presented on the vesicle surfaces were selected by two criteria, the continuity of the receptor density from the membrane and the clearly distinguished protein density compared with the background density. Reconstituted receptors were visualized as stick-like density originating at the membrane bilayer surface and presumably representing the LDL receptor extracellular domain. Calculations of the volume of the extracellular domain based on its experimentally determined length and width parameters (and assuming a cylindrical geometry) of all three observed shapes (stick-like, bent-stick, and Y-shaped) agree well with the volume calculated from the amino acid sequence of the human LDL receptor. This implies that all three observed shapes represent images of the extracellular domain of the LDL receptor. Although the stick-like and bent-stick images would represent logical projections of this single-chain receptor, the observation, albeit less frequently, of Y-shaped images is somewhat puzzling. At this stage, it is sensible to include this as an additional view of the extracellular domain with perhaps the stick-like and bent-stick-like projections being rotational side views of a Y-shaped receptor. The volume calculations described above indicate that the Y-shaped images represent monomeric structures of the LDL receptor. However, the LDL receptor has been reported to be able to dimerize on the cell surface (16), and the possibility remains that the Y-shaped images represent a dimeric form of the LDL receptor. Our structural studies of the covalent disulfide-linked ₂₂ heterodimeric insulin receptor do indeed reveal a Y-shaped receptor (Ref. 40; see also Ref. 41).

The receptor-labeling experiment was performed after analyzing the human LDL receptor sequence (5) and other LDL receptor sequences (6-10) to identify potentially free cysteine residues. The partial sequence (C-terminal 25%) of the bovine LDL receptor has been determined (11); this includes the cytoplasmic domain, the transmembrane domain, the O-linked sugar domain, and part of the EGF precursor homology domain. Over these regions, the sequences of the bovine and human LDL receptors show a very high homology, and we considered that it was likely that the free cysteine at residue 431 of the human LDL receptor was likely to be present in the bovine receptor. Our biotin-BMCC and maleimido-Nanogold labeling studies (Figs. 5-7) do confirm that a free cysteine (or cysteines) is accessible in the bovine LDL receptor. Since gold labeling was performed after receptor reconstitution, this will favor labeling of extravesicular cysteine residues. Although labeling of an outward-facing C-terminal cysteine (corresponding to Cys-818 in the human LDL receptor sequence) could occur, based on the gold-labeled images shown in Fig. 7 it is clear that an additional site (or possibly two sites) is available on the extracellular domain (presumably corresponding to Cys-431 in the middle of the EGF precursor homology domain of the human LDL receptor). Labeling by monomaleimido-Nanogold confirmed that the low contrast extravesicular density assigned to the extracellular domain of the LDL receptor was indeed correct.

In summary, we have provided a low resolution structural picture of the bovine LDL receptor based on cryoEM images of unlabeled and Nanogold-labeled vesicle-reconstituted receptor. The orientation of receptors at the edge of the vesicle facilitates the otherwise difficult identification of this low molecular weight receptor using the low contrast cryoEM method. In Fig. 8 we present an interpretation of the three views of the receptor. Presumably the distal arm (or arms) contains the ligand binding domain with the seven cysteine-rich repeats (and possibly the similar A and B cysteine-rich repeats of the EGF precursor homology domain). Logically, the membrane-connecting stem would comprise the rest of the EGF precursor homology domain and the O-linked sugar domain. In the accompanying paper (42), we describe experiments designed to label and visualize the N-terminal ligand binding domain. Eventually, with the availability of higher resolution images of the full-length LDL receptor, mapping the atomic resolution structures of receptor repeats (21-24) and, in the future, expressed LDL receptor sub-domains (43, 44) onto the cryoEM image of the LDL receptor will be possible.

View larger version (20K): [in this window] [in a new window]   Fig. 8.   A, schematic representation of the reconstituted LDL receptor images visualized by cryoEM. B, LDL receptor domain organization (adapted from Ref. 1). Cysteine-rich repeats are indicated as blue squares and red circles.

	ACKNOWLEDGEMENTS

We thank Drs. David Atkinson, Esther Bullitt, Kumkum Saxena, and Christine Woldin for helpful advice. Technical assistance was provided by Ann Tercyak, Cheryl England, Michael Gigliotti, Cynthia Curry, and Donald Gantz. Also, we thank Drs. Michael S. Brown and Joseph L. Goldstein (University of Texas Southwestern Medical Center, Dallas, TX) for providing us with the monoclonal antibody IgG-4A4.

	FOOTNOTES

^* This work was supported by National Institutes of Health Research Grants HL 57405 and HL 26335.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.

Present address: Dept. of Pathology, Harvard Medical School, Boston, MA 02115.

^§ To whom correspondence should be addressed: Biophysics Dept., Center for Advanced Biomedical Research, Boston University School of Medicine, 715 Albany St., Boston, MA 02118. Tel.: 617-638-4009; Fax: 617-638-4041; E-mail: shipley@med-biophd.bu.edu.

Published, JBC Papers in Press, July 10, 2000, DOI 10.1074/jbc.M002583200

	ABBREVIATIONS

The abbreviations used are: LDL, low density lipoprotein; -VLDL, -migrating very low density lipoprotein; apoB, apolipoprotein B₁₀₀; apoE, apolipoprotein E; EGF, epidermal growth factor; CD, circular dichroism; EM, electron microscopy; cryoEM, cryoelectron microscopy; EYPC, egg yolk phosphatidylcholine; -OG, -octyl glucoside; biotin-BMCC, 1-biotinamido-4-[4'-(maleimidomethyl)cyclohexanecarboxamido]- butane; 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:

				A. T. Nguyen, T. Hirama, V. Chauhan, R. MacKenzie, and R. Milne Binding characteristics of a panel of monoclonal antibodies against the ligand binding domain of the human LDLr J. Lipid Res., July 1, 2006; 47(7): 1399 - 1405. [Abstract] [Full Text] [PDF]

				M.-E. Roehrich, V. Mooser, V. Lenain, J. Herz, J. Nimpf, S. Azhar, M. Bideau, A. Capponi, P. Nicod, J.-A. Haefliger, et al. Insulin-secreting beta -Cell Dysfunction Induced by Human Lipoproteins J. Biol. Chem., May 9, 2003; 278(20): 18368 - 18375. [Abstract] [Full Text] [PDF]

				M. Brabec, G. Baravalle, D. Blaas, and R. Fuchs Conformational Changes, Plasma Membrane Penetration, and Infection by Human Rhinovirus Type 2: Role of Receptors and Low pH J. Virol., May 1, 2003; 77(9): 5370 - 5377. [Abstract] [Full Text] [PDF]

				G. Rudenko, L. Henry, K. Henderson, K. Ichtchenko, M. S. Brown, J. L. Goldstein, and J. Deisenhofer Structure of the LDL Receptor Extracellular Domain at Endosomal pH Science, December 20, 2002; 298(5602): 2353 - 2358. [Abstract] [Full Text] [PDF]

				H. Jeon and G. G. Shipley Localization of the N-terminal Domain of the Low Density Lipoprotein Receptor J. Biol. Chem., September 22, 2000; 275(39): 30465 - 30470. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 275/39/30458    most recent M002583200v1 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 Jeon, H. Articles by Shipley, G. G. Search for Related Content PubMed PubMed Citation Articles by Jeon, H. Articles by Shipley, G. G. Social Bookmarking                      What's this?