RPE Cells Internalize Low-Density Lipoprotein (LDL) and Oxidized LDL (oxLDL) in Large Quantities In Vitro and In Vivo

Abstract

purpose. To determine whether plasma low-density lipoprotein (LDL) could be internalized by the RPE and which receptors may be involved. A secondary objective was to determine whether ARPE19 cells could be used as a model to investigate cholesterol processing in the RPE.

methods. Commercially available human LDL was labeled with rhodamine or AlexaFluor 568. Immunofluorescence was performed using commercially available antibodies to LDL-R, CD36, and LOX-1. Cells and tissues were imaged using epifluorescence and confocal fluorescence microscopy. Immunoblot analysis and RT-PCR were performed using published techniques.

results. Intravenously injected rhodamine-labeled LDL (rhoLDL) was detected in the rat RPE by fluorescence confocal microscopy 24 hours after injection. The rhoLDL was present in some areas and absent in others. Cultured ARPE19 cells were also found to internalize LDL and oxidized LDL (oxLDL) readily. Using AlexaFluor 568–labeled LDL we determined that the average cultured RPE cell could internalize approximately 12 to 16 pg of LDL and oxLDL in 24 hours. Immunoblots readily detected the presence of CD36 and LDL-R in the cultured RPE cells but not LOX-1, whereas RT-PCR detected mRNA for all three receptors. Dual-labeling experiments using AlexaFluor 568–labeled LDL and AlexaFluor 488 for the immunolocalization of the receptors showed colocalization of LDL-R with the internalized LDL and CD36 with oxLDL particles.

conclusions. Plasma LDL readily enters the RPE through the choriocapillaris but is not found homogenously throughout the retina. This may suggest some form of regulation to the permeability of the fenestrated choroidal endothelial cell junctions. ARPE19 cells are a good model for studying the internalization mechanisms of LDL and oxLDL in vitro. LDL may be used as a vector to carry hydrophobic molecules into the RPE.

Age-related macular degeneration (AMD) is the leading cause of permanent vision loss in the elderly population of developed countries. ¹ Epidemiologic studies have shown a correlation between atherosclerosis and AMD. ² It is widely accepted that the clinical symptoms of AMD are precipitated by a gradual loss of function in the RPE that slowly leads to photoreceptor death and/or choroidal neovascularization. Recent scientific evidence suggests that there is an age-related accumulation of cholesterol in Bruch’s membrane under the macula in greater amounts than in peripheral retina. ³ This lipid/cholesterol accumulation in Bruch’s membrane also has been visualized recently in the form of 80- to 100-nm particles, by using a quick-freeze/deep-etch technique, ⁴ suggesting that aqueous fluid transport across Bruch’s membrane will be significantly impaired in older individuals. This accumulation may also lead to the oxidation of the cholesterol and damage to the RPE, either directly or through immune mediated processes. Apolipoprotein B (ApoB), the main backbone protein in the LDL molecule, was recently found in cholesterol-containing drusen and basal deposits in human eyes with AMD. ⁵ The origin of the cholesterol is unknown but these same authors ⁵ believe that the RPE has the potential to synthesize ApoB and possibly to assemble LDL-like molecules and thus suspect a retinal or RPE origin for the cholesterol. Alternatively, the high cholesterol requirements of the retina and the relatively low levels of de novo cholesterol synthesis, ^{6 7} coupled with the close proximity of Bruch’s membrane to the choriocapillaris, could also suggest a plasma origin for the LDL particles. Moreover, the RPE expresses the LDL receptor ⁸ and the scavenger receptor CD36 ⁹ which are essential for the internalization of LDL and oxidized LDL (oxLDL) in macrophages. ¹⁰ These two receptors work by fundamentally different mechanisms. The LDL receptor (LDL-R) recognizes the ApoB protein, and the CD36 receptor recognizes the oxidized phospholipids in the outer shell of the LDL particle. (For a comprehensive review of the structure of the LDL particle and its interacting receptors, please see Refs. ¹⁰ and ¹¹ , respectively.) The lectinlike oxidized LDL receptor (LOX-1) is a leukocyte adhesion molecule involved in mediating inflammatory responses. ¹² The LOX-1 receptor has not been reported in the RPE. The age-related accumulation of esterified and unesterified cholesterol in Bruch’s membrane and choriocapillaris especially under the macula ³ could lead to the formation of oxLDL and oxysterols. This in turn could directly impair RPE function if internalized through a variety of expressed receptors and could also attract scavenging macrophages, leading to an inflammatory response. OxLDL has been shown to inhibit photoreceptor outer segment phagocytosis in cultured RPE cells. ¹³ Much is known about the effects of oxLDL on macrophages and the events that lead to foam cell formation and cell death. However, one of the more interesting effects of oxLDL on macrophage function is the induction of VEGF, ¹⁴ which could explain the observed choroidal neovascularization in the exudative cases of AMD. 

In this study, we examined the internalization of LDL and oxLDL by human cultured RPE cells, the receptors involved and the amounts internalized. We also determined that the RPE is capable of internalizing LDL and accumulating LDL deposits in vivo. 

Methods

Materials

Human LDL cholesterol was purchased from Calbiochem (San Diego, CA); penicillin-streptomycin (10,000 IU/mL-10,000 μg/mL) from Media Teck, Inc. (Herndon, VA); mouse anti-CD36 antibody from Beckman Coulter Chemical Diagnostic Division (Brea, CA); rabbit anti-LOX-1 antibody from Santa Cruz Biotechnology Inc. (Santa Cruz, CA); a mouse monoclonal anti-LDL-R antibody from Oncogene Research Products (San Diego, CA). The antibodies were used at 1:1000 for immunoblots and at 1:100 or 1:50 for immunofluorescence. The secondary antibodies, anti-rabbit IgG biotin conjugate and anti-mouse IgG biotin conjugate were purchased from Sigma-Aldrich and used at 1:1000 and 1:600 for immunofluorescence, respectively. Mammalian protein solubilization reagent (M-PER) was purchased from Pierce (Rockford, IL). 

Preparation of Oxidized LDL

The LDL was oxidized using CuSO₄, as previously described. ¹⁵ LDL (1 mL) was dialyzed in 500 mL of 1× phosphate-buffered saline (PBS; pH 7.4) overnight. The LDL was then placed in another 500 mL of 1× PBS containing 20 μM CuSO₄ and allowed to oxidize at room temperature for 48 hours. The oxLDL is cytotoxic if given to ARPE19 cells for longer than 24 hours. ¹⁶  

Reverse Transcription–Polymerase Chain Reaction

RT-PCR was performed using cDNA synthesized on magnetic beads (Dynabeads; Dynal, Oslo, Norway), as previously described. ¹⁷ The oligonucleotides are listed in Table 1. The amplification was performed for 30 cycles using polymerase (Platinum Taq; Invitrogen, Carlsbad, CA) in standard conditions. 

Tissue Culture

ARPE19 cells were purchased from American Type Culture Collection (Manassas, VA). The cells were cultured in DMEM/F12 medium containing 10% fetal calf serum, 2 mM glutamine, 100 IU/mL penicillin, and 100 μg/mL streptomycin. Cells were grown in 24-well plates to 80% to 90% confluence (0.25 million cells) and placed in serum-free medium before the addition of LDL. 

Fluorescent Labeling of LDL

For in vivo experiments LDL was labeled with rhodamine (FluoReporter Rhodamine Red-X Protein Labeling Kit; Molecular Probes, Eugene, OR), according to the manufacturer’s instructions. For in vitro experiments, LDL was labeled with AlexaFluor 568 (AlexaFluor 568 Protein Labeling Kit; Molecular Probes) and purified according to the manufacturer’s instructions. 

SDS-PAGE Electrophoresis

Protein samples were mixed in equal portions with 2× Laemmli sample buffer containing 5% mercaptoethanol and heated at 100°C for 5 minutes. Approximately 20 μg of protein were loaded per well in 10% to 20% Tris-glycine-SDS gels and run in 1× Tris-glycine-SDS at room temperature for 1.5 hours at 150 V. The protein electrophoresis reagents and equipment were purchased from Invitrogen/Novex (Carlsbad, CA). 

Western Blot Analysis

The blot transfer was performed in 25 mM Tris, 192 mM glycine, and 10% methanol at 20 V overnight. Antibody detection was performed using a kit (amplified Opti-4CN Substrate Kit; Bio-Rad Laboratories, Inc., Hercules, CA), according to the manufacturer’s instructions. The membrane was equilibrated in PBS/0.1%Tween-20 (PBST) for 15 minutes, and blocked in PBST, 3% blocker for 2 hours. Incubations with primary antibodies were performed overnight at 4°C, followed by 1 to 2 hours of incubation with anti-rabbit or anti-mouse IgG horseradish peroxidase (HRP)–conjugated secondary antibody (Pierce) and development with the substrate (Opti-4CN; Bio-Rad). 

Immunofluorescence

ARPE19 cells were plated in eight-well chamber slides (Lab-Tek; Nalge Nunc International, Naperville, IL) and washed in PBS to remove traces of medium. The cells were then fixed for 20 minutes in fresh 4% paraformaldehyde-PBS, permeabilized, and blocked with normal goat serum (diluted 1:10) for 2 hours in PBST. The cells were washed three times in PBST and incubated with primary antibody overnight at room temperature. The slides were washed three times with PBS and incubated with secondary antibody (anti-mouse or anti-rabbit biotin conjugate) at room temperature for 2 hours. Cells were washed three times with PBS and incubated with AlexaFluor 488 FluoroNanogold Streptavidin (Molecular Probes) for 1 hour. The cell nuclei were counterstained with 4′,6′-diamino-2-phenylindole (DAPI; 1 μg/mL in PBS) for 10 minutes. The slides were mounted (GelMount; Biomeda Corp., Foster City, CA) and kept in the dark until viewing. 

Confocal Microscopy

Confocal microscopy was performed on a laser scanning confocal microscope (model SP2, with TCS software version 11.04; Leica Microsystems, Exton, PA) and a 40× objective. The DAPI staining was visualized by exciting with 351- and 364-nm laser beams and collecting emissions between 400 and 500 nm. Green fluorescence and AlexaFluor 488 staining was visualized by exciting with a 488-nm laser beam and collecting emissions between 500 and 552 nm. Rhodamine and AlexaFluor 568 were visualized by exciting with a 568-nm laser beam and collecting emissions between 580 and 650 nm. Magnifications varied, and scale bars are therefore digitally included in some of the pictures. 

Intravenous Injection of Rhodamine-Labeled Human LDL into Rats

Sprague-Dawley male rats (8-month-old retired breeders) were purchased from Charles River (Wilmington, MA). The rats were anesthetized with 40 to 80 mg of ketamine and 8 mg xylazine per kilogram of body weight. Intravenous injections of 100 μL of rhodamine LDL (5 μg/μL, 500 μg total) were given in the penile vein. The rats were euthanatized at 24 hours after injection by CO₂ asphyxiation, and the eyes and a small piece of liver tissue collected and fixed in 4% paraformaldehyde in PBS for 2 hours. The tissue was then frozen and sectioned for confocal microscopy. The animals were treated in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. 

Results

Intravenous Injection of Rhodamine-Labeled LDL in Albino Rats

The retina is capable of de novo cholesterol synthesis ^{6 7} but the presence of receptors capable of internalizing LDL in the RPE/choroid and Müller cells ⁸ suggests that circulating cholesterol may also be used. To demonstrate that circulating LDL could be internalized by the RPE, rhodamine-labeled nonoxidized LDL (rhoLDL) was injected intravenously (500 μg in 0.1 mL PBS) into two rats and their livers and eyes removed after 24 hours. The liver control clearly showed uptake of the LDL along the temporal artery (data not shown). The RPE also readily internalized the LDL (Fig. 1) . Two different control rats were used: One received PBS only (data not shown) and another received the neutralized rhodamine reagent in PBS. The sections were photographed under identical conditions. Under these conditions no distinguishable fluorescence was detected in either control rat when the tissues were excited with the 488-nm laser (for green) or the 568-nm laser (for red) (Figs. 1A 1C , respectively). The rats that received the rhoLDL showed red fluorescence in the RPE similar to that in the liver and cultured RPE cells. In addition, the RPE cells showed significantly increased green fluorescence over the control. This fluorescence was also observed in the liver but to a lesser extent. This suggests that the processing of the rhoLDL may be responsible for the green fluorescence. The rhoLDL was not detected homogenously in the RPE but only in certain areas, generally toward the center of the eye. No further efforts were made to identify these areas. We were able to detect large rhoLDL deposits in Bruch’s membrane in different locations in the rat eye (Fig. 2) . 

Quantification of LDL and oxLDL Internalization by ARPE19 Cells

Human LDL was labeled with AlexaFluor 568, as described earlier. The cells were grown in 24-well plates (0.25 million cells per well) and exposed to 10 to 100 μg/mL of LDL and oxLDL for 24 hours in serum-free medium. Cells were washed twice with PBS, solubilized with mammalian protein extraction reagent (M-PER) and the fluorescence measured directly. The fluorescence was normalized to a no-cell blank to control for any LDL adhering to the plastic wells. The results indicate that both LDL and oxLDL are readily internalized by the ARPE19 cells (Fig. 3) , reaching concentrations of 10 to 12 pg/cell for LDL and 14 to 16 pg/cell for oxLDL. At the higher concentrations, some cytotoxicity was observed with oxLDL, and longer exposure led to considerable cytotoxicity. The cytotoxicity of LDL and its accompanying oxysterols are the subject of a separate study (Ref. ¹⁷ , companion paper). 

Receptors Responsible for the Internalization of LDL and oxLDL in Cultured Human RPE Cells

The lipoprotein receptors LDL-R, ^{8 18} CD36, ^{8 9 19} SR-BI, and SR-BII ²⁰ have been reported in primary RPE cells. To determine whether the cultured ARPE19 cells used these receptors to internalize LDL, we analyzed them for the presence of LDL-R, CD36, and LOX-1. Using an RT-PCR technique, we could readily detect the mRNA for LDL-R and CD36, but LOX-1, although detectable, was present in very low levels (Fig. 4) . Immunoblots also detected the presence of CD36 and LDL-R but not LOX-1 (data not shown). The presence of the CD36 receptor in ARPE19 cells has been reported. ¹⁹  

To demonstrate the involvement of these receptors, we performed dual-labeling immunofluorescence localization on cultured RPE cells before and after internalization of LDL and oxLDL. The LDL was labeled with AlexaFluor 568 (A568LDL) and given to the cells before and after oxidation with Cu⁺². Using streptavidin AlexaFluor 488 to detect the receptors we looked for colocalization with the internalized A568LDL particles. As expected, CD36 localized to the plasma membrane (Fig. 5A) and did not colocalize with A568LDL (Fig. 5B) , but a significant amount of the CD36 delocalized from the plasma membrane and colocalized with the A568oxLDL (Fig. 5C) . Control results with no primary antibody, with and without LDL, are shown in Figures 5D and 5E , respectively. A similar experiment was performed with anti-LDL-R antibodies (Fig. 6) . As expected, a significant amount of colocalization was observed between A568LDL and LDL-R, but much less when the cells were treated with A568oxLDL (Figs. 6C 6D) . This demonstrates that ARPE19 cells internalize a significant amount of LDL and oxLDL through LDL-R and CD36, respectively. 

Discussion

In our study, circulating plasma human LDL was readily internalized by the rat RPE in vivo (Fig. 1) . However, the LDL uptake did not seem to be homogeneous throughout the RPE. LDL seemed to be present in some areas but not others. We also observed large LDL deposits in Bruch’s membrane at multiple locations (Fig. 2) . The LDL seemed to enter the RPE through fenestrated junctions in the choriocapillaris endothelium, crossing Bruch’s membrane to reach the RPE. However, there may be some size exclusion or some other mechanism that can exclude LDL from some locations while allowing passage in others. The rat retina has been shown to be capable of synthesizing cholesterol de novo. ⁶ Treatment of rats with U18666A, a potent cholesterol synthesis inhibitor, causes the accumulation of desmosterol, but the retinas are otherwise unaffected. ⁷ Based on this previous work ^{6 7} and the present results, we hypothesize that substantial amounts of cholesterol enter the retina through the RPE in the form of plasma lipoproteins such as LDL. The RPE breaks down these lipoproteins and possibly reassembles them into RPE-derived lipoproteins for export to the photoreceptors and other areas of the retina. A recent report indicating that ApoB may be synthesized by the RPE ⁵ suggests that this is certainly a possibility that needs further investigation. A previous in vivo study ²² has also demonstrated that circulating LDL can deliver omega-3 polyunsaturated fatty acids to the RPE, enhancing the acid lipase activity in monkeys. Our study also demonstrates that plasma LDL can get into the RPE very efficiently while carrying other molecules. Thus, the LDL molecule may be used as a way of introducing hydrophobic molecules into the RPE in vitro and in vivo. ²² The use of the rhoLDL in the in vivo experiments may have been a fortunate choice, because the succinimidyl ester used to react with the primary amines in proteins seemed to partition into the lipid moiety of the LDL molecule much more readily than the AlexaFluor 568 reagent. This not only makes for a significantly brighter labeling, but it may be responsible for the green fluorescence observed in the rat RPE after internalization of the rhoLDL. Thus, the rhodamine labeling may serve to follow the cholesterol esters after the ApoB has been cleaved and degraded. We did not detect any fluorescent labeling in the neural retina, perhaps due to a variety of reasons, but our opinion is that there is less cholesterol trafficking in the neural retina than in the RPE/choroid. Thus, to detect the plasma LDL in the neural retina unequivocally, the timing and dosage of the experiment may have to be optimized. A more comprehensive study involving more animals and LDL labeled with multiple dyes and other substances is under way to better understand this process. 

The ARPE19 cells seem to be a good model to study the internalization of LDL and oxLDL in vitro. The ARPE19 cells can internalize LDL and oxLDL in amounts ranging from 10 to 12 pg/cell and 14 to 16 pg/cell, respectively (Fig. 3) , in 24 hours. The cytotoxicity of oxLDL precluded performing internalizations studies longer than 24 hours. The internalization of LDL is mainly directed by LDL-R (Fig. 6) , which recognizes the ApoB protein, although significant amounts of LDL did not appear to be associated with LDL-R (Fig. 6C) . Preliminary studies treating rhoLDL with trypsin did not prevent its internalization, suggesting the possibility of another receptor (data not shown). The internalization of oxLDL occurs through receptors that recognize the oxidized phospholipids on the surface of the oxLDL molecule. ¹⁸ The ARPE19 cells, like macrophages, ^{10 21} seem to rely mainly on the CD36 receptor to internalize oxLDL. Immunocytochemistry experiments found that CD36 colocalized with most of the internalized oxLDL (Fig. 5) . The CD36 receptor has been demonstrated to be involved in the phagocytosis of rod outer segments by the RPE. ^{9 19} OxLDL has also been shown to inhibit rod outer segment phagocytosis in cultured RPE cells. ¹² This could be explained by either CD36 receptor depletion or cytotoxicity caused by oxLDL internalization. In the RPE cells, the CD36 receptor appears to be expressed apically, ⁹ but the possibility that this receptor may be used by the RPE to internalize oxLDL in vivo cannot be ruled out and should be investigated further. We have seen evidence of polarity with the CD36 receptor in ARPE19 cells, as shown in Figure 5 . This polarity in ARPE19 cells has been reported. ²³ However, similar to macrophages, the RPE cells also expresses other scavenger receptors ^{20 21} that may contribute to the internalization of the oxLDL. Our in vitro results with the ARPE19 cells are in general agreement with previous published work by Noske et al., ¹⁸ who used fluorescently labeled LDL (Dil-LDL) and SV40-transformed RPE cells and demonstrated that RPE cells poorly regulate the LDL receptor, leading to large intracellular accumulation of LDL. The same investigators hypothesized that this lipid accumulation may play an important role in the pathogenesis of AMD. 

The presence and function of the LOX-1 receptor in the RPE remains elusive. We failed to detect the presence of the protein in ARPE19 cells, even after attempting its induction by giving the cells oxLDL (data not shown). We also failed to detect LOX-1 by immunocytochemistry in monkey and rat tissues using the anti-LOX-1 commercially available antibody (Santa Cruz Biotechnologies). A recent study demonstrated that LOX-1 plays an important role in endotoxin-induced inflammation. ²³ This study showed the induction of LOX-1 in the endothelial cells of the rat retina venules after injection of a lipopolysaccharide endotoxin, ²³ but no LOX-1 was detected in the untreated control animals. LOX-1 seems to be involved in leukocyte recruitment and thus in mediation of inflammatory responses. Inflammation has long been suspected in the pathogenesis of AMD, and the role of LOX-1 in this process should be considered. 

In a companion study, ¹⁷ we used the ARPE19 cells to study the mechanisms of oxidized cholesterol cytotoxicity. Although there have been many studies demonstrating the cytotoxicity of cholesterol oxides in cultured cells, ²⁴ including the ARPE19 cells, ²⁵ those involved adding the purified oxysterol directly to the cells. This is an unlikely in vivo mechanism, since oxysterols are highly insoluble in aqueous medium and would not be readily available to the cells directly. The most likely way RPE cells internalize oxidized forms of cholesterol is through lipoproteins like LDL. The cytotoxicity of oxLDL is believed to be a major factor in atherosclerosis. ^{10 26} We believe that a related mechanism occurs in the RPE-Bruch’s membrane area of the retina and that this contributes to the pathogenesis of AMD. 

 Supported by the National Eye Institute’s intramural research program.

 Submitted for publication January 26, 2004; revised March 3, 2004; accepted March 15, 2004.

 Disclosure: N. Gordiyenko, None; M. Campos, None; J.W. Lee, None; R.N. Fariss, None; J. Sztein, None; I.R. Rodriguez, None.

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Table 1.

 

Oligonucleotide Primer Sequences

Gene	Forward Primer	Reverse Primer	Product Size (bp)	GenBank Accession No.
CD36	GATATTTGCAGGTCAATCTATGCT	ACATCACCACACCAACACTGAGTA	559	NM_000072
LDL-R	GAACTCCCGCCAAGATCAAGAAAG	GTCCGGGCAGGCGCAGGTAAA	491	NM_000527
LOX-1	GAAAAAGTGCACGTGAAGAAACAA	AGTGGGTGGAAAGGAAATAGAAGC	587	BC022295
GAPDH	CCCATCACCATCTTCCAGGAG	CTGCACCACCAACTGCTTAGC	257	BC014085

Figure 1.

 

Localization of rhoLDL in rat RPE 24 hours after intravenous injection. Rats received either rhoLDL or neutralized rhodamine reagent (control) in PBS. Eyes were removed after 24 hours, fixed, frozen, and sectioned. (A, C) Control sections and (B, D) rhoLDL-injected sections viewed in the red and green channels, respectively. The nuclei were stained with DAPI (blue).

Figure 2.

 

RhoLDL deposits in Bruch’s membrane of a rat 24 hours after injection. Longitudinal section through the rhoLDL-injected rat RPE-choroid demonstrating the deposition of rhoLDL between the choroidal endothelial capillary cell and Bruch’s membrane (BM).

Figure 3.

 

Internalization of LDL and oxLDL into ARPE19 cells. A568LDL (A) and A568oxLDL (B) were given to ARPE19 cells (0.25 million cells per well) at 10- to 100-μg/mL concentrations in serum-free medium for 24 hours, to determine the amount of LDL internalized by these cells. The values represent the average of four separate measurements from two different experiments. The error bars represent the standard deviation for the four measurements. The fluorescence was measured in a fluorometer exciting at 575 nm and measuring emissions at 603 nm.

Figure 4.

 

Detection of LDL-R, CD36, and LOX-1 receptors by RT-PCR. RT-PCR was performed to detect the presence of mRNA of the different receptors. Lane 1: GAPDH control; lane 2: LOX-1; lane 3: LDL-R; and lane 4: CD36.

Figure 5.

 

Immunofluorescence localization of CD-36 receptor in ARPE19 cells before and after internalization of LDL and oxLDL. Cultured RPE cells were treated with A568 LDL and A568oxLDL for 24 hours and then fixed. Immunofluorescence was performed using anti-CD36 antibody. (A) Anti-CD36, no LDL; (B) anti-CD36, A568LDL; (C) anti-CD36, A568oxLDL; (D) no-primary, no-LDL control; and (E) A568LDL, no antibodies.

Figure 6.

 

Immunofluorescence localization of the LDL receptor in ARPE19 cells before and after internalization of LDL and oxLDL. Cultured RPE cells were treated with AlexaFluor 568–labeled LDL and oxLDL for 24 hours then fixed. Immunofluorescence was performed using an anti-LDL-R antibody; (A) A568LDL, no antibodies; (B) anti-LDL-R, no LDL; (C) anti-LDL-R with A568LDL; (D) anti-LDL-R with A568oxLDL. Yellow deposits in (C) indicate colocalization of LDL and LDL-R. No significant amount of colocalization was observed with oxLDL (D). Scale bar, 20 μm.