Model System for Phenotypic Characterization of ... - CiteSeerX

Clinical Chemistry 52:8 1469 –1479 (2006)

Molecular Diagnostics and Genetics

Model System for Phenotypic Characterization of Sequence Variations in the LDL Receptor Gene Trine Ranheim,* Mari Ann Kulseth, Knut Erik Berge, and Trond Paul Leren Background: Sequence variations in the LDL receptor (LDLR) gene cause defects of LDLR protein production and function through different molecular mechanisms. Here we describe a cell model system for the phenotypic characterization of sequence variations in the LDLR gene. Well-known sequence variations belonging to LDLR classes 2 to 5 (p.G565V, p.I161D, p.Y828C, and p.V429M) were studied in CHO and HepG2 cells. Methods: Expression of LDLR protein on the cell surface was detected by use of fluorescence-conjugated antibodies against the LDLR and the LDLR activity was measured by incubating the cells with fluorescently labeled and radiolabeled LDL. The intracellular locations of the LDLR mutants and wild-type were also investigated. Results: The class 2A p.G565V sequence variant exhibited an intracellular distribution of LDLR with no active receptors on the cell surface. Both the class 3 p.I161D and class 4 p.Y828C sequence variants gave surface staining but had a reduced ability to bind or internalize LDL, respectively. By determining the intracellular locations of the receptors we were able to visualize the accumulation of the class 5 p.V429M sequence variant in endosomes by means of a specific marker, as well as confirming that the class 4 p.Y828C variant was not localized in clathrin-coated pits. Flow cytometry allowed us quantitatively to determine the amount and activity of receptors. To confirm the results of binding and cell association of fluorescently labeled LDL analyzed by flow cytometry, assays using 125I-labeled LDL were performed. In addition to a useful and valid alternative to radiolabeled LDL, the unique properties of fluorescently labeled LDL allowed a variety of detection technologies to be used.

Department of Medical Genetics, Rikshospitalet-Radiumhospitalet Medical Center, N-0027 Oslo, Norway. * Author for correspondence. Fax 47-23-075561; e-mail trine.ranheim@ rikshospitalet.no. Received February 14, 2006; accepted May 15, 2006. Previously published online at DOI: 10.1373/clinchem.2006.068627

Conclusions: This new approach enables phenotypic characterization of sequence variations in the LDLR gene. The assays developed may be valuable for confirming the pathogenicity of novel missense sequence variations found throughout the LDLR gene. © 2006 American Association for Clinical Chemistry

Familial hypercholesterolemia (FH)1 is an autosomal dominantly inherited disorder of lipoprotein metabolism caused by sequence variations in the LDL receptor (LDLR)2 gene (1 ). At present, more than 900 sequence variations in the LDLR gene have been identified (2, 3 ). The LDLR is a key regulator of cholesterol homeostasis. Newly transcribed LDLR mRNA reaches the endoplasmic reticulum (ER) for translation into a protein of Mr 120 000. During transport through the Golgi apparatus, the carbohydrate side chains are modified leading to an apparent increase in molecular mass of 40 000. The mature glycosylated protein (Mr 160 000) is transported to the cell surface where it clusters in clathrin-coated pits (4 ). LDL bound to the cell-surface LDLR is internalized by receptor-mediated endocytosis, discharged from its receptor, and degraded in the lysosomes; subsequently, the receptor is recycled to the cell surface (5 ). LDLRs contain 5 distinct domains: the ligand-binding domain, which consists of clusters of cysteine-rich repeats; epidermal growth factor-like repeats; 6 modules of ⬃50 amino acid residues, containing a YWTD motif that is proposed to form a ␤-propeller domain; a transmembrane domain; and a region with a tyrosine-containing motif (NPXY) in the cytoplasmic tail of the receptor (6, 7 ). Despite the hereditary nature of the disease, FH shows great variability in phenotypic expression. Plasma con-

1 Nonstandard abbreviations: FH, familial hypercholesterolemia; LDLR, LDL receptor; ER, endoplasmic reticulum; EYFP, enhanced yellow fluorescent protein; T-Rex, tetracycline-regulated expression; FBS, fetal bovine serum; DiI, 1,1⬘-dioctadecyl-3,3,3⬘,3⬘-tetramethylindocarbocyanine perchlorate; PBS, phosphate-buffered saline; BSA, bovine serum albumin; mAb, monoclonal antibody; EEA1, early endosome antigen 1; FITC, fluorescein isothiocyanate; and DiD, 1,1⬘-dioctadecyl-3,3,3⬘,3⬘-tetramethylindodicarbocyanine perchlorate. 2 Human gene: LDLR, LDL receptor.

1469

1470

Ranheim et al.: Phenotypic Classification of LDLR

centrations of LDL-cholesterol show large interindividual variation even in persons carrying the same LDLR gene sequence variation (8 ). However, in general, compared with persons who do not have FH, heterozygotes have 2to 3-fold higher plasma concentrations of LDL-cholesterol and homozygotes have 6- to 8-fold higher concentrations. Sequence variations in the LDLR gene have been categorized into 5 different classes based on biophysical and functional characteristics. The class of sequence variations and the resulting defective protein are suggested to impact on clinical phenotype (5, 9 ). Class 1 sequence variations produce no immunologically detectable protein, and the most frequent types of class 1 sequence variations are nonsense and deletion sequence variations (10 ). Class 2 sequence variations are the most common, and encode proteins that are blocked, either completely (class 2A) or partially (class 2B), in their transport from the ER to the Golgi apparatus (11 ). Class 3 sequence variations encode proteins that are synthesized and transported to the cell surface but fail to bind LDL normally (9 ). Sequence variations that encode receptors that move to the cell surface and bind LDL normally but are unable to cluster in clathrin-coated pits are denoted class 4 sequence variations (6, 12 ). Class 5 sequence variations produce receptors that bind ligands in coated pits but fail to release them at acidic pH in endosomes, and thus do not recycle to the cell surface (13 ). The most extensive classification of sequence variations in the LDLR gene is based on cultured fibroblasts of FH-homozygous persons (9 ). Homozygous FH is rare; therefore, to mimic the homozygous situation and to perform classification, cell lines are transfected with plasmids carrying the LDLR sequence variations. In this report we describe a cell model system for the phenotypic characterization of disease-causing sequence variations in the LDLR gene and use the previously characterized sequence variations p.G565V, p.I161D, p.Y828C, and p.V429M as model sequence variations for classes 2 to 5, respectively. Confocal laser microscopy and flow cytometry are systematically used to determine the cellular localization and biological activity of the mutant LDLR. The model can be used to identify the effects that the particular sequence variations will have on receptor function.

Materials and Methods mutagenesis, cloning, and expression of LDLR We constructed the plasmid pcDNA4-LDLR by transferring LDLR cDNA from the plasmid pGEM 3zf⫹, which contains the open reading frame of LDLR cDNA (kindly provided by Dr. Niels Gregersen, Aarhus University Hospital, Aarhus, Denmark) to pcDNA4/TO (Invitrogen Corporation). The LDLR plasmid contained an enhanced yellow fluorescent protein (EYFP) sequence fused to the 3⬘ end of the LDLR coding sequence to facilitate sorting of transfected cells and to visualize LDLR expression in

living cells. pcDNA4-LDLR-EYFP was constructed as described in Sørensen et al. (14 ). We created the following sequence variations in the LDLR gene: class 2A, p.G565V; class 3, p.I161D; class 4, p.Y828C; and class 5, p.V429M. These sequence variations are also denoted in the FH sequence variation databases as G544V, I140D, Y807C, and V408M, respectively. Sequence variations were introduced by oligonucleotidedirected mutagenesis using the QuickChange XL Mutagenesis Kit (Stratagene) according to the manufacturer’s instructions. The following oligonucleotides were used: 5⬘-ATTCAGTGGCCCAATGTCATCACCCTAGAT-3⬘ (mutated base in bold) was used to change the codon GGC, which encodes G565, to GTC, which encodes V; 5⬘ - CAACAGCTCCACCTGCGACCCCCAGCTGTG - 3⬘ (mutated bases in bold) was used to change the codon ATC, which encodes I161, to GAC, which encodes D; 5⬘-TGACAACCCCGTCTGTCAGAAGACCACAGA-3⬘ (mutated base in bold) was used to change the codon TAT, which encodes Y828, to TGT, which encodes C; 5⬘-CAACCTGAGGAACATGGTCGCTCTGGACAC-3⬘ (mutated base in bold) was used to change the codon GTG, which encodes V429, to ATG, which encodes M. The region containing the entire cDNA coding region was confirmed by DNA sequencing. The tetracycline-regulated expression (T-Rex)-CHO cell line carries the plasmid pcDNA6/TR (cat. no. R71807; Invitrogen), which stably expresses the tetracycline (Tet) repressor. This system makes possible induction of gene expression by addition of tetracycline to the cell culture medium. The T-Rex-CHO cell line was maintained in Ham’s F-12 medium containing blasticidin (10 mg/L; Invitrogen) and supplemented with 100 m/L fetal bovine serum (FBS; Euroclone), 50 kIU/L penicillin, and 50 mg/L streptomycin. HepG2 cells (cat. no. 85011430) obtained from European Collection of Cell Cultures were grown in EMEM supplemented with 100 m/L FBS, 50 kIU/L penicillin, 50 mg/L streptomycin, 2 mmol/L l-glutamine and 10 ml/L nonessential amino acids (Invitrogen). HepG2 cells were grown in cell culture flasks coated with collagen 1 (BD Biosciences). Cell growth media containing the above-mentioned ingredients are hereafter designated “complete medium”. T-Rex-CHO cells were transfected with the pcDNA4/TO containing cDNA that encoded wild-type or mutant LDLR by lipid-mediated transfection (Lipofectamine 2000 Reagent; Invitrogen). The Zeocin resistance gene in pcDNA4/TO allows selection of a stable CHO cell line by use of Zeocin (100 mg/L; Invitrogen). Gene expression was induced with tetracycline (1 mg/L), and cells expressing the EYFP tag were sorted out in a FACS-Diva flow cytometer (BD Biosciences). The expression of the wild-type and mutant forms of LDLR in the stably transfected cells was similar, as evaluated by Northern blot analysis (Fig. 1).

Clinical Chemistry 52, No. 8, 2006

Fig. 1. Northern blot analysis of wild-type and mutant LDLR in CHO cells. (Right panel), Northern blot showing mRNA encoding LDLR after incubation of CHO cells stably transfected with wild-type (wt) or mutant LDLR with cell culture medium containing tetracycline. (Left panel), mRNA from cells incubated in medium lacking tetracycline did not show any expression of LDLR mRNA. Bands of internal controls (␤-actin) were included to demonstrate loading of wells.

HepG2 cells were plated at a density of 3–5 ⫻ 104 cells/cm2 in a 6-well dish or on a culture slide, and were transiently transfected with DNA 1–2 days after plating. Briefly, 6 ␮L of jet PEI Cationic Polymer Transfection Reagent (Qbiogene) was mixed with 3 ␮g of plasmid DNA, both diluted in 100 ␮L of NaCl (150 mmol/L) and incubated for 25 min at room temperature. The mixture was then added to the cell culture medium and incubated overnight at 37 °C. The cells were used in experiments 1 day after transfection. 25-Hydroxycholesterol (5 mg/L) was added to the cell culture medium to decrease the endogenous synthesis of LDLR. The transfection efficiency was evaluated in cells expressing LDLR coupled to the EYFP tag and analyzed by flow cytometry. The obtained transfection efficiency was 30%–35%.

northern blot analysis mRNA was isolated by use of oligo(dT)25 Dynabeads (Dynal Biotech), as described by the manufacturer. mRNA was eluted in 0.05 mol/L MOPS (pH 7.0) containing 1 mmol/L EDTA, 56 mL/L formaldehyde, and 400 mL/L formamide. Loading buffer (2 mmol/L sodium phosphate buffer, 10 mL/L Ficoll 400, and 0.25 g/L bromphenol blue) was added, and the mRNA was separated on 1% agarose gels containing 67 mL/L formaldehyde. RNA was blotted on a nylon membrane (Hybond N⫹; Amersham Biosciences), and hybridization was carried out according to Church and Gilbert (15 ). For detection of human LDLR and ␤-actin mRNA, the probes pSP 15 (16 ) and the mouse ␤-actin cDNA (cat. no. A9715; Sigma-Aldrich), respectively, were used.

confocal laser scanning microscopy We used confocal laser scanning microscopy to determine the cellular location of the LDLR-EYFP fusion protein and LDL internalization in transfected cells. Receptor proteins located on the cell surface and intracellularly were differentially visualized by immunohistochemical staining tech-

1471

niques. Briefly, T-Rex-CHO and HepG2 cells were plated on fibronectin- or collagen I-coated glass slides, respectively (BD Biosciences) and allowed to adhere overnight before incubations or transfections. For studies of binding and internalization of LDL, fluorescently labeled 1,1⬘-dioctadecyl-3,3,3⬘,3⬘-tetramethylindocarbocyanine perchlorate (DiI)– LDL (20 mg/L; Molecular Probes) was added to each well, and cells were incubated at 37 °C for 4 h. After the incubation, the medium was removed, and the cells washed with probe-free medium. For staining of cell-surface receptors, the cells were blocked in phosphate-buffered saline (PBS, 200 mg/L KCl, 200 mg/L KH2PO4, 8 g/L NaCl, 2.16 g/L Na2 HPO4-7H2O) supplemented 5 g/L bovine serum albumin (PBS-0.5% BSA) for 1 h at ambient temperature, and then incubated on ice for 5 min before incubation with monoclonal antibody (mAb) IgG-C7 (2.5 mg/L; Progen Biotechnik GmbH) for 1 h at 4 °C (17 ). Cells were then washed 3 times at 4 °C in PBS containing 1 g/L BSA (PBS-0.1% BSA) and fixed in a methanol-free solution containing 40 mL/L formaldehyde (Sigma-Aldrich) for 5 min on ice and for an additional 15 min at room temperature. The fixed cells were washed 3 times with PBS-0.1% BSA and then incubated with Alexa Fluor 555- or 568-conjugated goat anti-mouse IgG (1:500; Molecular Probes) for 45 min in the dark. For intracellular detection of the receptor, the cells were incubated with mAb IgG-C7 (2.5 mg/L) for 1 h at 4 °C, washed, and then incubated for 10 min at 37 °C. Cells were then fixed as described above and permeabilized by incubation in PBS containing 1 mL/L Triton X-100 and 10 mmol/L glycine for 10 min at room temperature. Cells were then washed 3 times in PBS-0.1% BSA and incubated with either concanavalin A tetramethylrhodamine conjugates (an ER marker; 1:400; Molecular Probes), anti-early endosome antigen 1 conjugated to fluorescein isothiocyanate (EEA1:FITC; 1:100; BD Biosciences) or anti-clathrin heavy chain:FITC (1:100; BD Biosciences) for 1 h at room temperature. After 3 washes, the cells were incubated with Alexa Fluor 555conjugated goat anti-mouse IgG (1:500) for 45 min in the dark. Cells were then washed 3 times in PBS-0.1% BSA. The slides were equilibrated and mounted with SlowFade Anti Fade Kit with 4⬘,6-diamidino-2-phenylindole (DAPI; Molecular Probes) and sealed with nail polish. Fluorescent images were obtained on a Leica laser-scanning confocal microscope with the 100⫻ oil objective and analyzed by Leica confocal software (Leica Microsystems Ltd.). Image processing was carried out with PhotoShop, Ver. 5.5 (Adobe Systems).

lipoprotein isolation From freshly prepared plasma collected from healthy persons who had fasted overnight before venipuncture, LDL was isolated by sequential ultracentrifugation in a Beckman Coulter Optima LE-80K ultracentrifuge (Beckman Coulter) in the density range 1.019 –1.063 kg/L in an ultrarotor 70TI (Beckman Coulter) (18 ). The final preparations were dialyzed extensively against PBS containing 0.2 mmol/L EDTA (pH 7.4), sterilized by passage through

1472

Ranheim et al.: Phenotypic Classification of LDLR

a 0.45 ␮m filter, and stored under sterile conditions at 4 °C. The lipoprotein concentration was expressed in terms of protein content, determined by BCA Protein Assay Kit.

flow cytometry LDLR expression. T-Rex-CHO cells (0.5 ⫻ 106 cells) were grown in cell culture medium supplemented with tetracycline (1 mg/L) in 6-well plates for 24 h. Cells were detached from the culture dish with Cell Dissociation Solution Non-enzymatic (Sigma-Aldrich) and resuspended in PBS containing 10 g/L BSA (PBS-1% BSA). A wash step with PBS-1% BSA was repeated once, and cells were incubated at 4 °C for 1 h with 250 ␮L of mAb IgG-C7 (2.5 mg/L) diluted in PBS-1% BSA. Subsequently, the samples were washed 3 times with PBS-1% BSA and incubated at 4 °C for 30 min in the dark with secondary antibody Alexa Fluor 488 or Alexa Fluor 647 goat antimouse IgG diluted in PBS-1% BSA (1:400; Molecular Probes). Cells were again washed 3 times with PBS-1% BSA and resuspended in CellFix (1⫻; BD Biosciences). Measurements were performed on a FACSCanto flow cytometer (BD Biosciences) equipped with a diode laser at 488 nm and a helium/neon laser at 633 nm. The cell population was selected in a forward- vs side-scatter window to exclude dead cells and debris. Dead cells (unfixed) were also confirmed by propidium iodide staining (final concentration, 1 g/L; Sigma-Aldrich). For each sample, fluorescence of 20 000 events was acquired for the data analysis. Fluorescence signals were recorded to produce a histogram of the gated cells vs relative fluorescence intensity after logarithmic amplification. The results were expressed as the geometric mean fluorescence of the gated cells. Fluorescence representing specific binding to LDL receptors in T-Rex-CHO cells was obtained by subtracting the fluorescence of a blank cell sample had been treated in the same way as our positive sample except that the cells had not been incubated with tetracycline; therefore, no LDLR expression was induced. LDLR function. LDL was labeled with 1,1⬘-dioctadecyl3,3,3⬘,3⬘-tetramethylindodicarbocyanine perchlorate (DiD) (Molecular Probes) according to the procedure of Pitas et al. (19 ). DiD is an analog of DiI with markedly red-shifted absorption and fluorescence emission spectra; this characteristic is useful for 2-color labeling. There is no spillover of signals between the detectors that measure DiD and EYFP. T-Rex-CHO cells (0.5 ⫻ 106 cells) were grown in cell culture medium supplemented with tetracycline (1 mg/L) for 24 h in 6-well plates. The effect of increasing concentrations of fluorescently labeled LDL (0, 2, 5, 10, or 20 mg/L), as well as the time course of uptake (0, 15, 30, 45, 60, 120, and 240 min) in CHO cells stably transfected with wild-type LDLR, showed that the increase in binding and

internalization of LDL was linear with LDL concentration and time of incubation. Thus, LDL concentrations of 10 or 20 mg/L and incubation times of 2 or 4 h were chosen for further experiments. After the LDL incubation, the medium was removed, and the cells were washed with probe-free medium. The cells were detached from the culture dish and fixed, as described above. HepG2 cells were transfected and incubated as described above. Cells transfected with an empty pcDNA4/TO plasmid were used to correct for nonspecific binding of DiD-LDL in HepG2 cells. Transfection efficiencies were visualized by means of the EYFP tag fused to the LDLR, and any differences were allowed for in the calculations.

labeling of ldl with iodine-125 One aliquot of LDL was labeled with 125I-labeled tyramine cellobiose (20 ). More than 98% of the radioactivity was precipitated by 100 g/L trichloroacetic acid. The advantage of labeling a protein with radioiodinated tyramine cellobiose is that the radiolabel is trapped in the organelles where degradation of the protein takes place. Before use, the 125I-labeled lipoproteins were diluted with unlabeled LDL to a specific activity of 100 cpm/ng of protein. 125I-labeled LDL was stored in the presence of EDTA under argon at 4 °C and used within 2–3 weeks.

binding, internalization, and degradation studies of 125i-labeled ldl The assays were performed as described by Goldstein et al. (21 ). Briefly, cells were grown in cell culture medium supplemented with tetracycline (1 mg/L) for 24 h in 6-well plates, as described above. For binding experiments, cells were placed for 30 min at 4 °C. The growth medium (Ham’s F-12 medium containing 10 mg/L blasticidin and supplemented with 100 mL/L FBS, 50 kIU/L penicillin, 50 mg/L streptomycin, and 2 mmol/L lglutamine) was removed and replaced with ice-cold medium containing 125I-labeled LDL (10 mg/L LDL; 100 cpm/ng of protein). After the 2-h LDL incubation, the cells were washed extensively, and the homogenized cell fractions were precipitated by an equal volume of trichloroacetic acid (200 g/L), and counted on a gamma counter (Packard Cobra Gamma Counter). We determined the specific binding by subtracting the amount of 125I-labeled LDL bound in the presence of a 20-fold excess unlabeled LDL from that bound in its absence. For uptake and degradation experiments, cells were incubated at 37 °C for 2 h. Uptake and degradation of labeled LDL were calculated from the trichloroacetic acid–insoluble plus the soluble fraction and the soluble fraction, respectively. Nonspecific cell-associated or degraded LDL was always ⬍5% of the total in cells expressing the wild-type LDLR. The protein content of cells was measured by use of a BCA Protein Assay Kit.

Clinical Chemistry 52, No. 8, 2006

detection of ldlr by radiolabeled antibody The receptor was also detected by IgG-C7 radiolabeled with secondary 125I-labeled anti-mouse IgG antibody (0.2 ␮Ci/mL; Amersham Biosciences) to confirm the results obtained with fluorescently labeled antibody. After the incubation, the cells were washed extensively, and the homogenized cell fractions were precipitated by an equal volume of trichloroacetic acid (200 g/L) and counted on a gamma counter.

statistical analysis Values are reported as the mean (SD) of the indicated samples and number of experiments for all analyses.

Results confocal laser scanning microscopy analysis of wild-type and mutant ldlr in cho cells CHO cells stably transfected with wild-type or mutant LDLR were analyzed by confocal immunofluorescence microscopy to study the intracellular location of mutant receptors. To determine whether the cellular localization of LDLR was affected by introduction of an EYFP tag, we evaluated CHO cells expressing wild-type LDLR, with or without EYFP tag, by confocal microscopy (Fig. 2A). Both showed a membrane-associated expression pattern, and both could efficiently bind and internalize fluorescently labeled LDL. In addition, we have previously shown that the EYFP tag does not interfere with the maturation of wild-type LDLR from the Mr 120 000 precursor to the Mr 160 000 mature form irrespective of whether the receptor had an EYFP tag (14 ). In the following experiments, we used CHO cells stably transfected with LDLR constructs containing the EYFP tag. We studied the intracellular location of the mutant receptors and their ability to bind and internalize LDL (Fig. 2B). The class 2A p.G565V sequence variation exhibited an intracellular distribution of LDLRs with no active receptors on the cell surface as determined by the failure to internalize LDL. The class 3 p.I161D and class 4 p.Y828C sequence variations gave surface staining but had decreased ability to bind and internalize LDL. The class 5 p.V429M sequence variation exhibited an intracellular distribution similar to that of the class 2A p.G565V sequence variation. To further characterize the different classes of sequence variations, we compared subcellular localization of the mutant LDLRs with known markers of cellular compartments. Class 2A mutant LDLRs are completely retained in ER (22 ). We used rhodamine-conjugated concanavalin A to visualize ER in transfected CHO cells expressing either the wild-type or the class 2A p.G565V mutant LDLR. The EYFP tag was used to detect the LDLR. The class 2A p.G565V sequence variation produced a protein that was colocalized with the ER marker (Fig. 3A). As shown in Fig. 2B, the class 5 p.V429M mutant receptor exhibited intracellular distribution. The class 2 mutant receptors are

1473

retained in the ER, whereas class 5 mutant receptors are transported to the cell surface, internalized, and trapped in the endosomes. We stained permeabilized CHO cells expressing wild-type or mutant LDLR with an early endosome antigen (EEA1) to study whether the class 5 p.V429M mutant receptor assembled in this compartment because of a recycling defect (Fig. 3B). Early endosomes are cellular compartments that receive endocytosed material and sort them for vesicular transport to late endosomes and lysosomes or for recycling to the plasma membrane (23 ). We used cells that were stably transfected with wild-type LDLR or class 5 p.V429M mutant LDLR without the EYFP tag, which allowed us to use the EEA1 FITC-conjugated marker and thus avoid problems with overlapping fluorescence spectra between FITC and EYFP. LDLR was visualized by use of C7 mAb detected by an Alexa Fluor 555-conjugated IgG antibody. The C7 mAb does not recognize the endogenous hamster LDLR (17 ). As shown in Fig. 2B the class 5 p.V429M mutant LDLR accumulated in early endosome compartments, as opposed to the wild-type protein (Fig. 3B). Class 4 mutant LDLRs have been described to be unable to assemble in clathrin-coated pits (24 ). Clathrin is the major protein component involved in receptor-mediated endocytosis (25 ). We analyzed the colocalization of wild-type and class 4 p.Y828C mutant receptor with clathrin, and as indicated in Fig. 3C, the class 4 p.Y828C mutant receptor did not colocalize with clathrin, in contrast to the wild-type LDLR. In summary, by determining the subcellular location of the receptors in combination with intracellular markers, we were able to show that the class 2 p.G565V mutant receptor was localized in the ER, that the class 5 p.V429M mutant receptor accumulated in early endosomes, and that the class 4 p.Y828C mutant receptor was not localized in clathrin-coated pits. However, to analyze the phenotypic characteristics of LDLR mutant classes, flow cytometric analyses were included.

flow cytometric analysis of cell-surface ldlr and binding and uptake of ldl in cells expressing wild-type and mutant ldlr The immunofluorescence studies described showed that the different classes of sequence variations in the LDLR gene affected the amount of cell-surface receptor and function of the LDLRs. We measured cell-surface LDLR in nonpermeabilized CHO cells by use of the C7 mAb. The receptor activity was determined by DiD-LDL incubations. The class 2A p.G565V and class 5 p.V429M mutants expressed no surface LDLR proteins, whereas the class 3 p.I161D and class 4 p.Y828C mutants expressed receptors that were present at the cell surface equally to wild-type receptors (Fig. 4A). To confirm the results obtained by fluorescently labeled antibody, we performed an assay with radiolabeled IgG-C7 (Fig. 4B). Neither of the

1474

Ranheim et al.: Phenotypic Classification of LDLR

Fig. 2. Confocal laser images of wild-type (WT) and mutant LDLR in CHO cells. (A), Cells stably transfected with either wild-type LDLR or wild-type LDLR with an EYFP tag were plated on collagen-coated glass slides, grown in medium containing tetracycline for 24 h, and then incubated for 2 h with DiI-LDL. For staining of cell-surface receptors of cells that did not have the fused EYFP tag, cells were incubated with mouse mAb IgG-C7, fixed, and counterstained with Alexa Fluor 488-conjugated goat anti-mouse IgG. Overlays are shown in the right panels with colocalization appearing as yellow. (B), Cells were plated on collagen-coated glass slides, grown in medium containing tetracycline for 24 h, and then incubated for 4 h with DiI-LDL. Overlays are shown in the right panels with colocalization appearing yellow. Similar results were obtained in 3 separate experiments.

Clinical Chemistry 52, No. 8, 2006

1475

Fig. 4. Measurements of cell-surface LDLR in CHO cells by flow cytometry. (A), for flow cytometric analysis of cell-surface LDLR, cells stably transfected with wild-type (WT) or mutant LDLR were cultured in cell culture medium containing tetracycline for 24 h. Cells were labeled with the IgG-C7 LDL receptor antibody and counterstained with anti-IgG Alexa Fluor 647. (B), for 125I-labeled monoclonal anti-LDLR IgG-C7 analysis, cells stably transfected with wild-type or mutant LDLR were grown in cell culture medium supplemented with tetracycline in 6-well plates for 24 h and then incubated with IgG-C7 radiolabeled with secondary 125 I-labeled anti-mouse IgG antibody. The results shown are the mean (SD; error bars) of triplicate incubations. Similar results were obtained in 2 separate experiments.

Fig. 3. Confocal laser images of wild-type (WT) and mutant LDLR in CHO cells. Intracellular detection of the receptor in combination with antibodies against an ER marker (concanavalin A; A), an endosome marker (EEA1; B), or a clathrin marker (clathrin heavy chain; C). Overlays are shown in the right panels with colocalization appearing yellow. Similar results were obtained in 2 separate experiments.

2 labeling procedures revealed any significant differences in the expression of surface LDLRs. We studied LDLR activity by incubating the CHO cells with DiD-LDL. Measurements revealed that all sequence variations impaired the ability of LDLR to internalize LDL (Fig. 5A). Cells expressing class 2 p.G565V and class 5 p.V429M mutant receptors had, as expected, almost no detectable residual activity. Cells expressing class 3 p.I161D and class 4 p.Y828C mutant receptors displayed 28% and 53% residual activity, respectively, compared

with cells expressing wild-type LDLR. The binding efficiency for LDL in cells expressing the class 4 p.Y828C mutant LDLR was similar to the efficiency for cells expressing wild-type LDLR (Fig. 5B). With respect to the class 3 p.I161D mutant, the binding of LDL was reduced by 80%. We included phenotypic characterization of the LDLR in transiently transfected HepG2 cells to determine whether cell type variation was present. As shown in Fig. 6, the uptake of DiD-LDL in HepG2 cells expressing class 2 p.G565V and class 3 p.I161D mutant receptors resembled that of CHO cells. However, the class 4 p.Y828C mutant receptor internalized 38% of labeled LDL, as opposed to 53% by the CHO cells, compared with wildtype cells. The class 5 p.V429M mutant receptor had a higher LDL uptake in HepG2 cells than in CHO cells. In summary, flow cytometry allowed us to determine, in a more quantitative way, the amounts of cell-surface LDLRs and binding and internalization of LDL, compared with confocal microscopy analysis. Phenotypic characterization of the LDLR in transfected CHO or HepG2 cells may demonstrate that a slight cell type variation is present.

1476

Ranheim et al.: Phenotypic Classification of LDLR

Fig. 5. Flow cytometric measurements of LDL binding and uptake in CHO cells. For flow cytometric analysis of LDL uptake (A) and binding (B), cells stably transfected with wild-type (WT) or mutant LDLR were cultured in cell culture medium containing tetracycline for 24 h and then incubated for 4 h with DiD-LDL at 37 °C (A) and 4 °C (B). The results shown are the mean (SD; error bars) of triplicate incubations. Similar results were obtained in 3 separate experiments.

binding, internalization, and degradation of 125 i-labeled ldl in cho cells expressing wild-type and mutant forms of ldlr To confirm the results of binding and cell association of fluorescently labeled LDL, we performed 125I-LDL assays, which are acknowledged as the gold standard for LDL metabolism, in transfected CHO cells (Fig. 7). LDL binding (Fig. 7A), carried out at 4 °C for 2 h, was absent for the

Fig. 6. Flow cytometric measurements of LDL uptake in HepG2 cells. For flow cytometric analysis of LDL uptake, HepG2 cells transiently transfected with wild-type (WT) or mutant LDLR were cultured in cell culture medium containing 25-hydroxycholesterol for 24 h and then incubated for 4 h with DiD-LDL. The results shown are the mean (SD; error bars) of triplicate incubations. Similar results were obtained in 3 separate experiments.

Fig. 7. Binding, uptake, and degradation of

125

I-LDL in CHO cells.

125

For analysis of I-LDL binding (A), uptake (B), and degradation (C), cells stably transfected with wild-type (WT) or mutant LDLR were cultured in cell culture medium containing tetracycline for 24 h and then incubated for 2 h with 125I-LDL at 4 °C (A) or 37 °C (B and C). Similar results were obtained in 2 separate experiments with different preparations of LDL. The results shown are the mean (SD; error bars) of triplicate incubations.

class 2 p.G565V and class 5 p.V429M mutants and was present in only 15% for the class 3 p.I161D mutants, compared with wild-type. Wild-type and class 4 p.Y828C mutant cell lines demonstrated similar binding of labeled LDL. In accordance with the minimal LDL binding, the amount of 125I-LDL that was contained within the cells (Fig. 7B) and degraded (Fig. 7C) was decreased by ⬃75% in the cells expressing the class 3 p.I161D mutant receptors compared with wild-type. With respect to class 2 p.G565V and class 5 p.V429M mutant receptors, the internalization and degradation of 125I-LDL were minimal (Fig. 7, panels B and C, respectively).

1477

Clinical Chemistry 52, No. 8, 2006

Table 1. Summary of methods to distinguish different classes of sequence variations in the LDLR gene. LDLR sequence variation

Confocal microscopy

Class 2A; p.G565V Class 3; p.I161D Class 4; p.Y828C

ER localization

Class 5; p.V429M

Endosome accumulation

No clathrin colocalization

Internalization of radiolabeled LDL was higher in cells expressing the class 4 p.Y828C mutant receptor than in these cells incubated with fluorescently labeled LDL, although the binding of LDL was similar. At 10 mg LDL protein/L, the class 4 p.Y828C cells internalized a total of 150 ng of 125I-LDL per mg of cell protein. [Total internalization is the sum of the amount of LDL that had been degraded plus the amount of LDL that was still contained within the cell (6 ).] Approximately 30 ng/mg was bound to the surface. Thus, the cells had internalized ⬃5 times as much LDL as was bound to the surface after 2 h of incubation. When the cells were transfected with the wild-type, these cells had internalized 8 times as much LDL as was bound to the surface. In spite of the relatively high uptake of 125I-LDL for this sequence variation, it indicated that the class 4 p.Y828C mutant had 60% decreased proteolytic degradation of LDL in lysosomes. On degradation of the radiolabeled protein, the label is not itself degraded, but left behind in the cells as a cumulative marker of the amount of degradation that occurs in those cells (20 ). Although it is easy to use the versatile labeling of LDL with fluorescence, it is not possible to quantify the degradation products of LDL (26 ). In summary, the results from the 125I-LDL assay confirmed the results obtained by flow cytometric analysis of DiD-LDL binding and cell association. Thus, flow cytometry appears to be a valid replacement for the 125I-LDL assay.

Discussion In this report we have classified known sequence variations in the LDLR gene by use of a cell model system in combination with confocal laser microscopy and flow cytometry. CHO and HepG2 cells were transfected with an LDLR–EYFP fusion protein. After having verified that the EYFP tag did not affect intracellular traffic or receptor function, we used the fusion protein to develop methods for characterizing LDLR sequence variations in transfected cells. Sequence variations in the LDLR gene were introduced by oligonucleotide-directed mutagenesis to create the different classes of sequence variations. This permits the functional analysis of allelic variants found in heterozygous FH individuals who inherit 1 mutant LDLR allele. The presence of the wild-type allele in FH heterozygotes makes direct analysis difficult. Using confocal laser microscopy and flow cytometry methods, we were able to perform phenotypic characterization of 4 mutant LDLRs.

Flow cytometry

No surface LDLR or LDL internalization Normal surface LDLR; minimal LDL binding Normal surface LDLR; normal LDL binding; decreased receptor internalization

The results were confirmed by radiolabeled LDL. In accordance with Schmitz et al. (27 ), we documented that the binding characteristics of DiI-LDL were similar to those of iodide-labeled LDL. The class 2A mutant receptor p.G565V has previously been shown to be completely retained in the ER (14, 28 ). Our confocal images showed an intracellular distribution of p.G565V mutant LDLR colocalized with an ER marker and no active receptors on the cell surface. As shown in Fig. 5B and Fig. 7A, the binding of fluorescently labeled and radiolabeled LDL by the class 3 p.I161D mutant was reduced by 80% and 85%, respectively, although this is not as dramatic as the reductions in binding reported by Russell et al. (29 ). They reported a 98% decrease in 125I-LDL binding in COS cells transfected with plasmids harboring the class 3 mutant receptor. The residual activity determined for the class 4 p.Y828C mutant LDLR is higher than shown in other systems (Figs. 5, 6, and 7). It has been reported that only 5% of the receptor is internalized in skin fibroblasts from a homozygote with the same sequence variation (24 ) and 30% in a mutant line of CHO cells (ldlA-7) (6 ). A critical point in LDLR activity quantification is the question of which cell type provides the most reliable results. The T-Rex-CHO cell line, which stably expresses the tetracycline repressor, makes it possible to induce high amounts of LDLRs after adding tetracycline to the cell culture medium. Absolute measurements of internalization are difficult in a system in which the receptor cycles rapidly into and out of cells (24 ). The difference in cell association might also be attributable to the high induction of the receptor in our cell system. Davis et al. (6 ) indicated 2 possibilities that might account for the discrepancy between primary human cells and CHO cells: (a) hamster cells may have a more rapid rate of nonspecific incorporation of the LDLR into clathrin-coated pits compared with human fibroblasts; and (b), the transfected cells may internalize receptors through noncoated pit pathways more rapidly than do human fibroblasts. However, in terms of nonspecific incorporation into clathrin-coated pits, the indicated fluorescence in our experiments represents specific binding of LDLRs or internalization of LDL in CHO cells. This is obtained by subtracting the fluorescence from cells that are not induced with tetracycline, and thus do not express the receptor, but are otherwise treated identically as cells induced by the drug. We found that the class 4 p.Y828C sequence variation did not colocalize with a clathrin marker in CHO cells, as opposed

1478

Ranheim et al.: Phenotypic Classification of LDLR

to the wild-type, in an experiment using fluorescent probes detected by a confocal laser microscopy (Fig. 3C), indicating that the receptor was incorporated through pathways other than via clathrin-coated pits. According to the lack of colocalization of the receptor and the clathrin marker, we suggest that this sequence variation has an internalization defect and could be classified as such. In addition, the decreased degradation of radiolabeled LDL confirms the deficiency. It might be possible to demonstrate the internalization defect more clearly if the CHO cells are incubated at a lower temperature (20 °C) instead of 37 °C, at which noncoated pit internalization might be relatively less important (27 ). We transiently transfected HepG2 cells with mutant LDLRs and found that the class 4 p.Y828C mutant receptor internalized 38% of labeled LDL, as opposed to 53% by the CHO cells, when compared with wild-type. The demonstrated cell type variation may possibly be caused by the presence of PCSK9, a proprotein convertase, in the HepG2 cells. PCSK9 is highly expressed in HepG2 cells but expressed only in low amounts in CHO cells (30 ). Studies have suggested that PCSK9 acts through a posttranscriptional mechanism to negatively regulate LDLR protein concentrations (31 ). The class 5 p.V429M mutant LDLR encoded a protein that resided only intracellularly, as shown both by confocal microscopy and flow cytometric assays (Fig. 2B and Fig. 4). This is probably because the cells were incubated in medium containing bovine serum. As a result, the lipoproteins bind to the LDLRs, which are internalized but cannot recycle, because of the recycling defect in this mutant receptor (32 ). It has been reported that if recycling does not occur in fibroblasts, all of the LDLR would be consumed within 10 min after exposure to LDL (33 ). We have confirmed by a pulse-chase experiment for the class 5 mutant that a mature LDLR band appeared after a chase of 1 h, but thereafter decreased in intensity and vanished over time because of degradation (data not shown). To verify that the p.V429M LDLR mutant could be classified as a class 5 and not a class 2 sequence variation, we used an immunofluorescence staining technique to show that the LDLR protein produced by the class 5 sequence variation accumulated in early endosome compartments (Fig. 3B). This is in accordance with Chang et al. (34 ), who reported that sequence variations in the epidermal growth factor domain of LDLR give rise to an endosomal staining pattern. This is the first time that a cell model has been used to apply many different methods for systematically distinguishing the different classes of sequence variations. Others (35 ) have reported a rapid characterization of sequence variations in the LDLR gene, but have not included confocal laser microscopy to determine the subcellular location of the LDLR-EYFP fusion protein and LDL internalization in the transfected cells. We argue that this method is crucial for correct classification. In addition, we have included phenotypic characterization of the

LDLR in transiently transfected HepG2 cells to determine whether cell type variation was present. Lastly, our use of flow cytometry to analyze binding and cell association of fluorescently labeled LDL has been validated by 125I-LDL assays, which are acknowledged as the gold standard for LDL metabolism assays. In conclusion, we have developed and validated a novel approach for the phenotypic characterization of sequence variations in the LDLR gene. The assays we have developed will be valuable for confirming the pathogenicity of some of the new missense sequence variations that are found throughout the LDLR gene. Table 1 summarizes the methods chosen for distinguishing the different classes of sequence variations.

We thank Tove Skodje for excellent technical assistance.

References 1. Goldstein JL, Hobbs HH, Brown MS. Familial hypercholesterolemia. In: Scriver CR, Beaudet AL, Sly WS, Valle D, eds. The Metabolic Basis of Inherited Disease. New York: McGraw-Hill, 2001:2863–913. 2. Division of Cardiovascular Genetics. Department of Medicine, University College London Medical School. The low density lipoprotein receptor (LDLR) gene in familial hypercholesterolemia. http:// www.ucl.ac.uk/fh (accessed ???). 3. Laboratory of Human Genetics. Universal mutation database (UMD). http://www.umd.necker.fr (accessed ????) 4. Tolleshaug H, Goldstein JL, Schneider WJ, Brown MS. Posttranslational processing of the LDL receptor and its genetic disruption in familial hypercholesterolemia. Cell 1982;30:715–24. 5. Brown MS, Goldstein JL. A receptor-mediated pathway for cholesterol homeostasis. Science 1986;232:34 – 47. 6. Davis CG, Lehrman MA, Russell DW, Anderson RG, Brown MS, Goldstein JL. The J.D. mutation in familial hypercholesterolemia: amino acid substitution in cytoplasmic domain impedes internalization of LDL receptors. Cell 1986;45:15–24. 7. Hobbs HH, Brown MS, Goldstein JL. Molecular genetics of the LDL receptor gene in familial hypercholesterolemia. Hum Mutat 1992; 1:445– 66. 8. Bertolini S, Pisciotta L, Di SL, Langheim S, Bellocchio A, Masturzo P, et al. Genetic polymorphisms affecting the phenotypic expression of familial hypercholesterolemia. Atherosclerosis 2004;174: 57– 65. 9. Hobbs HH, Russell DW, Brown MS, Goldstein JL. The LDL receptor locus in familial hypercholesterolemia: mutational analysis of a membrane protein. Annu Rev Genet 1990;24:133–70. 10. Hobbs HH, Leitersdorf E, Goldstein JL, Brown MS, Russell DW. Multiple crm⫺ mutations in familial hypercholesterolemia. Evidence for 13 alleles, including four deletions. J Clin Invest 1988;81:909 –17. 11. Goldstein JL, Brown MS, Anderson RG, Russell DW, Schneider WJ. Receptor-mediated endocytosis: concepts emerging from the LDL receptor system. Annu Rev Cell Biol 1985;1:1–39. 12. Lehrman MA, Goldstein JL, Brown MS, Russell DW, Schneider WJ. Internalization-defective LDL receptors produced by genes with nonsense and frameshift mutations that truncate the cytoplasmic domain. Cell 1985;41:735– 43. 13. Beglova N, Jeon H, Fisher C, Blacklow SC. Cooperation between

Clinical Chemistry 52, No. 8, 2006

14.

15. 16.

17.

18.

19.

20.

21.

22.

23.

24.

fixed and low pH-inducible interfaces controls lipoprotein release by the LDL receptor. Mol Cell 2004;16:281–92. Sorensen S, Ranheim T, Bakken KS, Leren TP, Kulseth MA. Retention of mutant low density lipoprotein receptor in ER leads to ER stress. J Biol Chem 2006;281:468 –76. Church GM, Gilbert W. Genomic sequencing. Proc Natl Acad Sci U S A 1984;81:1991–5. Peacock SL, Bates MP, Russell DW, Brown MS, Goldstein JL. Human low density lipoprotein receptor expressed in Xenopus oocytes. Conserved signals for O-linked glycosylation and receptor-mediated endocytosis. J Biol Chem 1988;263:7838 – 45. Beisiegel U, Schneider WJ, Goldstein JL, Anderson RG, Brown MS. Monoclonal antibodies to the low density lipoprotein receptor as probes for study of receptor-mediated endocytosis and the genetics of familial hypercholesterolemia. J Biol Chem 1981;256: 11923–31. Havel RJ, Eder HA, Bragdon JH. The distribution and chemical composition of ultracentrifugally separated lipoproteins in human serum. J Clin Invest 1955;34:1345–53. Pitas RE, Innerarity TL, Weinstein JN, Mahley RW. Acetoacetylated lipoproteins used to distinguish fibroblasts from macrophages in vitro by fluorescence microscopy. Arteriosclerosis 1981;1:177– 85. Pittman RC, Carew TE, Glass CK, Green SR, Taylor CA Jr, Attie AD. A radioiodinated, intracellularly trapped ligand for determining the sites of plasma protein degradation in vivo. Biochem J 1983;212: 791– 800. Goldstein JL, Basu SK, Brown MS. Receptor-mediated endocytosis of low-density lipoprotein in cultured cells. Methods Enzymol 1983;98:241– 60. Gillian-Daniel DL, Bates PW, Tebon A, Attie AD. Endoplasmic reticulum localization of the low density lipoprotein receptor mediates presecretory degradation of apolipoprotein B. Proc Natl Acad Sci U S A 2002;99:4337– 42. Mu FT, Callaghan JM, Steele-Mortimer O, Stenmark H, Parton RG, Campbell PL, et al. EEA1, an early endosome-associated protein. EEA1 is a conserved ␣-helical peripheral membrane protein flanked by cysteine “fingers” and contains a calmodulin-binding IQ motif. J Biol Chem 1995;270:13503–11. Goldstein JL, Brown MS, Stone NJ. Genetics of the LDL receptor: evidence that the mutations affecting binding and internalization are allelic. Cell 1977;12:629 – 41.

1479

25. Robinson MS. The role of clathrin, adaptors and dynamin in endocytosis. Curr Opin Cell Biol 1994;6:538 – 44. 26. Stephan ZF, Yurachek EC. Rapid fluorometric assay of LDL receptor activity by DiI-labeled LDL. J Lipid Res 1993;34:325–30. 27. Schmitz G, Bruning T, Kovacs E, Barlage S. Fluorescence flow cytometry of human leukocytes in the detection of LDL receptor defects in the differential diagnosis of hypercholesterolemia. Arterioscler Thromb 1993;13:1053– 65. 28. Esser V, Russell DW. Transport-deficient mutations in the low density lipoprotein receptor. Alterations in the cysteine-rich and cysteine-poor regions of the protein block intracellular transport. J Biol Chem 1988;263:13276 – 81. 29. Russell DW, Brown MS, Goldstein JL. Different combinations of cysteine-rich repeats mediate binding of low density lipoprotein receptor to two different proteins. J Biol Chem 1989;264: 21682– 8. 30. Seidah NG, Benjannet S, Wickham L, Marcinkiewicz J, Jasmin SB, Stifani S, et al. The secretory proprotein convertase neural apoptosis-regulated convertase 1 (NARC-1): liver regeneration and neuronal differentiation. Proc Natl Acad Sci U S A 2003;100: 928 –33. 31. Park SW, Moon YA, Horton JD. Post-transcriptional regulation of low density lipoprotein receptor protein by proprotein convertase subtilisin/kexin type 9a in mouse liver. J Biol Chem 2004;279: 50630 – 8. 32. Davis CG, Goldstein JL, Sudhof TC, Anderson RG, Russell DW, Brown MS. Acid-dependent ligand dissociation and recycling of LDL receptor mediated by growth factor homology region. Nature 1987;326:760 –5. 33. Goldstein JL, Anderson RG, Brown MS. Coated pits, coated vesicles, and receptor-mediated endocytosis. Nature 1979;279: 679 – 85. 34. Chang JH, Pan JP, Tai DY, Huang AC, Li PH, Ho HL, et al. Identification and characterization of LDL receptor gene mutations in hyperlipidemic Chinese. J Lipid Res 2003;44:1850 – 8. 35. Jensen TG, Andresen BS, Jensen HK, Jensen LG, Heath F, Pedersen S, et al. Rapid characterization of disease-causing mutations in the low density lipoprotein receptor (LDL-R) gene by overexpression in COS cells. Z Gastroenterol 1996;34(Suppl 3):9 –11.