High-Density Lipoprotein-Binding Protein (HBP)/Vigilin Is Expressed in Human Atherosclerotic Lesions and Colocalizes With Apolipoprotein E
Abstract Accumulation of cholesteryl esters within cells of the arterial intima is a hallmark of atherosclerosis. A small number of proteins have been shown in vitro to be upregulated by cellular cholesterol loading, including apolipoprotein E (apoE) and the recently cloned HDL-binding protein (HBP), but only apoE has been shown to be upregulated in cholesterol-loaded cells in atherosclerosis. To determine whether HBP (also called vigilin) might be expressed in human atherosclerosis, immunohistochemistry and in situ hybridization were performed on coronary arteries of 18 patients. HBP/vigilin was detected on all endothelial cells. HBP/vigilin mRNA and protein also were detected on a subset of macrophages and occasionally on smooth muscle cells (SMC) in atherosclerotic plaques but were not detected on these cell types in nondiseased coronary intima. The majority of HBP/vigilin-expressing macrophages were foam cells, but HBP/vigilin expression also was detected rarely in nonfoam cell macrophages. Foam cell macrophage HBP/vigilin expression was present in 100% of atherosclerotic quadrants, and nonfoam cell macrophage HBP/vigilin expression was present in 6% of atherosclerotic quadrants. HBP/vigilin-expressing human plaque cells also expressed apoE. However, HBP/vigilin was detected in cardiac myocyte foam cells of an apoE-deficient mouse, demonstrating that HBP/vigilin expression can occur independently of apoE. These results suggest that in vivo HBP/vigilin expression is upregulated by intracellular cholesterol loading but also that other factors present in atherosclerotic plaques may upregulate HBP/vigilin. Although the exact function of HBP/vigilin is unknown, its expression in plaque macrophages suggests a role for this molecule in atherogenesis.
Division of Cardiology (D.S.C, K.D.O), Division of Metabolism, Endocrinology and Nutrition (J.F.O., R.C.L.), Department of Medicine; Department of Nutritional Science (R.C.L.); Department of Pathology (C.E.A.) University of Washington, Seattle, Wash.
Presented in part to the Western Section, American Federation for Clinical Research, Carmel, CA, February 11, 1995, and to the American Federation for Clinical Research, San Diego, CA, May 6, 1995.
- Received May 23, 1996.
- Accepted August 19, 1997.
A hallmark of atherosclerosis is the accumulation of cholesteryl esters within cells of the arterial intima. Most of these intracellular esters are formed when macrophages and SMC ingest lipoproteins by endocytotic or phagocytotic processes, degrade the lipoproteins in lysosomes, and re-esterify the liberated free cholesterol in the cytoplasm.1 A fraction of this ingested cholesterol likely is excreted from cells in response to cell surface interactions of high-density lipoproteins (HDL).2–4
Although cellular processes for delivery of cholesterol have been well characterized, the molecular properties of pathways involved in handling and excretion of excess cellular cholesterol are poorly understood. Only a small number of cellular proteins have been shown to undergo positive regulation in response to cholesterol loading. Of those in macrophages, apolipoprotein E (apoE) has been studied the most extensively.5 Synthesis and secretion of apoE increase when cultured macrophages are overloaded with cholesterol.6–9 The secreted apoE interacts with the cell surface of macrophages to stimulate efflux of cholesterol.10,11 ApoE is expressed in human atherosclerotic lesions and is localized to foam cell macrophages.12–15 Thus, apoE may act in an autocrine fashion to mobilize excess cholesterol from macrophages of atherosclerotic lesions.
Another cellular protein shown to be regulated positively by cholesterol loading of macrophages is HDL-binding protein (HBP).16 HBP was cloned from a human cDNA expression library by using an antibody raised against a 110-kD cellular protein that binds HDL on ligand blots.16 The chicken homolog of HBP, called vigilin (based on the prevalence of valine-isoleucine-glycine repeats), was cloned independently from chondrocytes.17 HBP/vigilin is expressed primarily as a 150-kD membrane-bound protein localized in the cytoplasm of cells.16–18 This protein appears to undergo processing to form a 110-kD protein that binds HDL on ligand blots and that is localized, at least partially, to the plasma membrane.16
The specific function of HBP/vigilin has not been established. The protein is expressed in a wide variety of cell types and tissues,16–22 and its relative abundance is sensitive to the growth and differentiation states of cultured cells.16,18,19,21 Although it was cloned during an attempt to identify a cellular HDL receptor and although forms of HBP/vigilin appear to be associated with plasma membrane, this molecule lacks the classic signal peptide and the membrane-spanning domain that typify cell surface receptors. Thus, HBP/vigilin is unlikely to be an HDL receptor. Findings from several studies have suggested other possible functions for HBP/vigilin. The full-length protein comprises 14 imperfect tandem repeats of approximately 70 amino acids that contain KH domains, sequence motifs that are usually found in RNA-binding proteins.16,17,23 Some of these KH domains in HBP/vigilin have been shown to have weak nucleotide-binding activities,24 and HBP/vigilin has been found to be associated with cytoplasmic tRNA21 and mRNA.22 This protein also contains a nuclear localization sequence and has been detected in the nucleus of cells.25 These properties have led to the hypothesis that HBP/vigilin plays a role in modulating protein synthesis through its direct interaction with RNA.21,22,25
Several other lines of evidence suggest that HBP/vigilin may play a role in cellular sterol metabolism. First, a study in rats showed that the HBP/vigilin mRNA levels in the corpus luteum of the ovary increases with gonadotropin treatment,20 a finding suggesting that HBP/vigilin expression is associated with steriodogenesis. Second, estrogen and testosterone treatment of tissues alters HBP/vigilin expression.22 Third, cholesterol loading of cultured J774 macrophages increases HBP/vigilin mRNA and protein levels.16 This latter study raises the possibility that HBP/vigilin may be functionally important in the biology of foam cell macrophages in atherosclerotic lesions.
Therefore, this study was designed to examine the following questions: (1) Is HBP/vigilin present in human atherosclerotic lesions? (2) If so, in which cell types is it expressed? (3) Is HBP/vigilin expression associated with intracellular lipid accumulation? (4) How does the expression of HBP/vigilin compare to that of apoE, another molecule that is upregulated by cell cholesterol loading?
Human Arterial Tissue
Human coronary artery segments were obtained from the excised native hearts of 18 cardiac transplant recipients and snap-frozen in OCT or fixed in 10% neutral buffered formalin (NBF) within 2 hours of organ removal. Six-micrometer-thick frozen sections were used for immunohistochemistry, histologic staining with oil red O (ORO, Sigma), and combined immunohistochemistry and ORO staining. NBF-fixed, paraffin-embedded tissue was used for in situ hybridization and for combined in situ hybridization and immunohistochemistry.
Antisera and Antibodies
Rabbit Anti-HBP/Vigilin IgG
A rabbit polyclonal antiserum was generated by immunization of rabbits with a 21-amino-acid peptide (RGQVLREIAEEYGGVMVSFPRC) corresponding to amino acid residues 819 to 840 of human HBP/vigilin.16 To enhance immunization, the peptide was synthesized with an additional cysteine residue and coupled to rabbit serum albumin using a NHS-ester-maleimide cross-linking agent (Pierce, Rockford, Illinois). A protein A column was used to isolate the IgG fraction from this antiserum and the IgG fraction was used (titer, 1:500) to detect HBP/vigilin. Specificity of this IgG for HBP/vigilin was confirmed by ligand and Western blots of HBP/vigilin-transfected BHK cells, which demonstrated the presence of a major band at 150 kD and of a minor band at 110 kD, corresponding to the predicted molecular weights of, respectively, the primary HBP/vigilin gene product and its processed form.
Mouse Monoclonal Antibodies
To distinguish cell types in the arterial wall, the following cell specific mouse monoclonal antibodies were used: (1) HAM-56 (Dako; titer, 1:1000) or anti-CD68 (Dako; titer, 1:4000), which are specific for human macrophages, and (2) anti-smooth-muscle α-actin (Boehringer Mannheim; titer, 1:250), which is, in this context, specific for SMC. A mouse monoclonal antibody directed against apoE (Monosan; titer, 1:75) was used to detect apoE. Before immunohistochemistry, tissues were post-fixed in cold methanol (for anti-HBP/vigilin IgG, HAM-56 or anti-CD68), 10% neutral buffered formalin (for anti–α-actin), or acetone (for anti-apoE). Slides were counterstained with either hematoxylin or methyl green.
Single-label immunohistochemistry (IHC) was performed as described previously.26–28 Briefly, tissues were blocked with 3% H2O2 (Sigma Corp, St Louis, Missouri), washed with phosphate-buffered saline (PBS), incubated for 60 minutes with the primary IgG or monoclonal antibody, and then washed again with PBS. A biotin-labeled anti-mouse secondary antibody then was applied for 30 minutes, followed by an avidin-biotin-peroxidase conjugate (ABC Elite, Vector Laboratories, Burlingame, Calif) for 30 minutes. Standard peroxidase enzyme substrate, 3,3′-diaminobenzidine (Sigma), was added to yield a brown reaction product or 3,3′-diaminobenzidine with NiCl2 was added to yield a black reaction product.
Oil-Red O Staining
Intracellular lipid accumulation was determined using ORO, which stains neutral lipids such as esterified cholesterol and triglycerides. Frozen sections were brought to room temperature and covered with filtered, working ORO solution for 4 minutes. Slides then were rinsed four times in 60% isopropanol, washed in H2O, and counterstained with hematoxylin.
Combined ORO Staining and Immunohistochemistry
Staining with ORO was combined with immunohistochemical staining with either anti-CD68 or anti-α-actin to identify, respectively, macrophage or smooth muscle cell foam cells. Frozen sections were brought to room temperature and covered with filtered, working ORO solution for 4 minutes. Slides were rinsed four times in 60% isopropanol, then washed in 1X PBS. Immunohistochemistry then was performed as described previously27 using mouse monoclonal antibody anti-CD68 (titer, 1:4000) or mouse monoclonal antibody anti α-actin (titer, 1:250). Slides were counterstained with hematoxylin, and a coverslip was applied.
To confirm the expression of HBP/vigilin by cells of the artery wall, in situ hybridization was performed on NBF-fixed, paraffin-embedded tissue with 35S-labeled antisense riboprobes to detect HBP/vigilin and apoE mRNA. The antisense HBP/vigilin riboprobe was transcribed from a 1545-bp fragment of the human HBP/vigilin cDNA, corresponding to bases 2810 to 4354. Antisense riboprobes were transcribed from the cDNA using reagents from Promega, except 35S-UTP 1000 to 3000 Ci/mmol, which was obtained from New England Nuclear (Boston, Massachusetts). The sense (control) HBP/vigilin riboprobe was transcribed from a fragment containing the first 848 bp of the human HBP/vigilin cDNA.
The HBP/vigilin antisense or sense riboprobe transcription reaction mixtures contained 1 μg of DNA; 250 μCi of 35S-UTP; 500 μmol/L each of rATP, rCTP, and rGTP; 40 U RNasin (Promega); 10 mm of dithiothreitol; 40 mm of Tris; and either 10 U of SP6 polymerase (for antisense riboprobe transcription) or 10 U of T7 polymerase (for sense riboprobe transcription). ApoE antisense riboprobe was transcribed from a 1-kb human apoE cDNA using SP6 polymerase as described previously.15 After 75 minutes of incubation at 37°C, the cDNA was digested by adding 1 U of RQ1 DNAse (Promega), and incubation was continued for an additional 15 minutes at 37°C. Free nucleotides were separated by using a Sephadex G-50 column, and the riboprobes were used within 24 hours of synthesis.
In Situ Hybridization
Formalin-fixed, paraffin-embedded, 6-μm arterial sections were washed with 0.5X standard saline citrate (SSC) (1X SSC=150 mm NaCl, 15 mm Na citrate, pH 7.0) and digested with proteinase K (1 mg/mL) (Sigma) in RNAse A (Promega) buffer. After several 0.5X SSC washes, 50 μL of prehybridization buffer (0.3 mol/L NaCl, 20 mmol/L Tris pH 8.0, 5 mmol/L ethylenediaminetetraacetic acid, 1X Denhardt’s solution, 1X dextran sulfate, 10 mmol/L dithiothreitol) was applied for 2 hours at 50°C. For hybridizations, 35S-labeled anti-sense riboprobe (300 000 cpm in 50 μL of prehybridization buffer) was added, and hybridizations were allowed to proceed overnight at 50°C. After hybridization, sections were washed with 0.5X SSC, treated with RNAse A (20 μg/mL) for 30 minutes, and washed twice in 2X SSC, followed by three high-stringency washes in 0.1X SSC/Tween 20 (Sigma) at 37°C, followed by several 2X SSC washes. After air drying, the tissue was dipped in NTB2 nuclear emulsion (Kodak) and incubated in the dark for 10 to 35 days. After development, the sections were counterstained with hematoxylin.
Combined Immunohistochemistry and in Situ Hybridization
Immunohistochemistry with HAM-56 or anti-smooth-muscle α-actin was performed as described above, with the following exceptions: (1) RNasin (Promega) was added at a concentration of 180 U/mL to the primary and secondary antibodies as well as the avidin-biotin complex to inhibit endogenous RNases, (2) antibodies were used at twofold to fourfold higher concentrations, and (3) PBS was treated with diethyl pyrocarbonate (Sigma) to denature endogenous RNases. In situ hybridization with the apoE riboprobe then was performed as reported previously15 according to the above protocol.
HBP/Vigilin Expression in apoE (apoE−/−) Mice
Because HBP/vigilin protein has been demonstrated in murine macrophages16 and by Western blot analyses of murine tissues (Oram JF, unpublished observations, 1996), immunohistochemistry also was performed on 6-μm-thick frozen sections obtained at the level of the aortic sinuses from a 17.5-week old apoE−/− mouse that had been fed a high-fat, high-cholesterol diet containing 15% fat, 1.25% cholesterol, and 0.5% cholate for 8 weeks.29 This mouse had markedly elevated plasma lipids with total cholesterol of 6844 mg/dL and triglycerides of 343 mg/dL.
HBP/Vigilin Presence in All Endothelial Cells
Immunohistochemical staining of nonatherosclerotic segments with rabbit polyclonal HBP/vigilin antiserum localized HBP/vigilin to endothelial cells, both at the arterial lumen (Fig 1⇓) and in endothelium of the adventitial vasa vasorum. However, HBP/vigilin was not detected in nonendothelial cells of morphologically normal regions of the intima. Further, HBP/vigilin was not detected either in medial SMCor in nonlipid-laden resident macrophages of the adventitia.
B HBP/Vigilin Expression by Lipid-Laden Nonendothelial Cells of Atherosclerotic Plaques
In contrast, HBP/vigilin (Fig 2a⇓) frequently was detected in nonendothelial cells of atherosclerotic arteries, particularly those with lipid inclusions as detected by ORO staining (Fig 2b⇓). The vast majority of these lipid-laden nonendothelial cells were macrophages, as determined by immunohistochemistry of adjacent sections (Fig 2c⇓).
Further confirmation of the foam cell types that expressed HBP/vigilin was obtained by comparison of slides stained with HBP/vigilin to adjacent sections double-stained with ORO plus the macrophage marker, anti-CD68, or with ORO and the smooth muscle cell marker, anti-α-actin (Fig 3⇓). This technique allowed the unambiguous identification of either macrophage or smooth-muscle-cell foam cells. The vast majority of HBP/vigilin-positive foam cells (Fig 3b⇓) were macrophage-derived (Fig 3a⇓), while HBP/vigilin-positive SMC foam cells were detected only rarely (Fig 3c⇓).
Infrequent HBP/Vigilin Expression by Nonfoam Cell Macrophages
A second pattern of HBP/vigilin expression, specifically HBP/vigilin expression by ORO-negative (ie, nonfoam cell) nonendothelial cells, also was seen (Figs 4a⇓, 4b⇓). Immunohistochemical staining of adjacent sections with cell-specific antibodies indicated that these HBP/vigilin-positive nonfoam cells were macrophages.
Prevalence of HBP/Vigilin Expression by Foam Cells and Nonfoam Cells
The relative prevalences of foam cell–associated HBP/vigilin expression and of nonfoam cell–associated HBP/vigilin expression were determined semiquantitatively. Each arterial segment was divided into four quadrants, and each quadrant then was scored for the presence or absence of HBP/vigilin-expressing foam cells and for the presence or absence of HBP/vigilin-expressing nonfoam cells. One hundred percent of plaque quadrants contained HBP/vigilin-positive foam cells. In contrast, only 6% of plaque quadrants contained HBP/vigilin-positive nonfoam cells. Both patterns of HBP/vigilin expression were specific to atherosclerosis, as neither pattern of HBP/vigilin was found in any of the nonatherosclerotic control quadrants examined (Fig 5⇓).
Association of HBP/Vigilin With apoE
Cell-associated apoE has been detected in atherosclerotic plaques, particularly in foam cell macrophages, a finding suggesting that intracellular lipid accumulation might upregulate the expression of that protein,13–15,30 as has been suggested for HBP/vigilin. Immunohistochemistry demonstrated that apoE protein was present on both foam cell macrophages (Fig 2d⇑) and nonfoam cell macrophages (Fig 4c⇑) that expressed HBP/vigilin (comparison with Figs 2a⇑, 2b⇑, and 2c⇑ and Figs 4a⇑ and 4b⇑) in atherosclerotic sections. Cell-associated HBP/vigilin and apoE were present in all plaque quadrants with a highly statistically significant correlation between HBP/vigilin-positive foam cells and apoE-positive foam cells (χ2 value: 33.601, P < .001, Table 1). ApoE was not detected in SMC within atherosclerotic plaques. Extracellular apoE was present in all atherosclerotic quadrants and was detected in 22 of 37 nonatherosclerotic control quadrants.
Detection of HBP/Vigilin mRNA in Plaque Macrophages by in Situ Hybridization
To confirm local production of HBP/vigilin production of HBP/vigilin in atherosclerotic plaques, in situ hybridization was performed with antisense and sense (ie, control) HBP/vigilin riboprobes. Concordant with the immunohistochemical location of HBP/vigilin protein, strong in situ hybridization signal was detected with the antisense HBP/vigilin riboprobe (Figs 6c⇓, 6d⇓) in cells identified by immunohistochemistry on adjacent sections as macrophage-derived foam cells (Figs 6a⇓, 6b⇓), thus confirming that macrophage-derived foam cells synthesize HBP/vigilin in atherosclerotic plaques. Specificity of the in situ hybridizations was confirmed by the absence of specific signal on sections to which the sense HBP/vigilin riboprobe had been hybridized (Figs 6e⇓, 6f⇓).
Comparison of HBP/Vigilin and apoE mRNA Expression in Plaques by in Situ Hybridization
To confirm the expression of HBP/vigilin by plaque cells, in situ hybridization was performed to detect HBP/vigilin mRNA using a 35S-labeled antisense riboprobe on formalin-fixed sections (Fig 7b⇓). On neighboring sections, combined in situ hybridization for apoE mRNA (Figs 7a⇓, 7c⇓) and immunohistochemistry for either macrophages (Fig 7a⇓) or SMCs (Fig 7c⇓) was performed. HBP/vigilin mRNA was detected in foam cells (Fig 7b⇓), which also expressed apoE mRNA (Figs 7a⇓, 7c⇓) and were identified by immunohistochemistry as macrophages (Fig 7a⇓) rather than SMCs (Fig 7c⇓). The presence of HBP/vigilin mRNA in apoE mRNA–expressing macrophages is consistent with the hypothesis that similar factors may upregulate the expression of both genes in these cells.
Immunohistochemical Detection of HBP/Vigilin in Cardiac Myocyte Foam Cells in an ApoE−/− Mouse
To further investigate the relationships of HBP/vigilin expression to both intracellular lipid accumulation and apoE expression, immunohistochemistry was performed by using the rabbit anti-HBP/vigilin IgG or nonimmune (control) rabbit IgG at a titer of 1:500 on frozen sections from the level of the cardiac aortic sinuses in an apoE−/− mouse (Fig 8⇓). This mouse, which had extraordinarily high plasma cholesterol owing to a combination of the genetic apoE deficiency and a high-fat diet, had cardiac myocyte foam cells, which also contained immunohistochemically detectable HBP/vigilin (Fig 8a⇓). Specificity of immunohistochemical staining for HBP/vigilin was confirmed by demonstration of the absence of immunohistochemical staining with the nonimmune rabbit IgG (Fig 8b⇓). The presence of HBP/vigilin in cardiac myocyte foam cells of apoE−/− mice further strengthens the association of HBP/vigilin expression with excessive cell cholesterol accumulation. In addition, the presence of HBP/vigilin in apoE−/− mice demonstrates that HBP/vigilin expression does not require expression of apoE.
HBP/vigilin is a ubiquitous cellular protein that is upregulated by treatment of cells with cholesterol and steroid hormones.16–25 In this study, immunohistochemistry and in situ hybridization for HBP/vigilin were used to characterize the expression of HBP/vigilin in normal and atherosclerotic human coronary arteries. In both nondiseased and diseased specimens, HBP/vigilin was localized to endothelial cells. However, nonendothelial HBP/vigilin expression was found only in cells with significant intimal lipid accumulation, as demonstrated by ORO staining. HBP/vigilin was associated primarily with foam cell macrophages, although a small subset of nonfoam cell macrophages also expressed HBP/vigilin. Rare SMC foam cells also showed positive HBP/vigilin staining. Overall, 100% of plaque quadrants were found to contain HBP/vigilin-positive foam cells, while only 6% of plaque quadrants contained HBP/vigilin-positive nonfoam cells. The expression of HBP/vigilin by nonendothelial cells of atherosclerotic lesions but not by nonendothelial cells of normal intima suggests that this protein is important in atherosclerosis. Additionally, the observation that nonendothelial cell HBP/vigilin expression is restricted primarily to foam cells suggests a prominent role for intracellular lipid accumulation in the regulation of HBP/vigilin expression. This observation is consistent with previous in vitro studies demonstrating that HBP/vigilin mRNA, and protein levels are increased with cholesterol loading in SMCs and in the J774 mouse macrophage cell line.16
ApoE, a lipid transport protein that is hypothesized to be important in atherosclerosis, is synthesized and secreted by several peripheral tissues,5 including atherosclerotic plaque macrophages.12,15 ApoE expression is upregulated by increases in cellular cholesterol content6–9 as well as by macrophage-activating growth factors and cytokines.31–33 Interestingly, 100% of atherosclerotic plaque quadrants exhibited HBP/vigilin and apoE colocalization in foam cell macrophages. A much smaller proportion of atherosclerotic plaque quadrants also had HBP/vigilin and apoE colocalization in nonfoam cell macrophages. The demonstration of HBP/vigilin and apoE coexpression by foam cell macrophages is consistent with in vitro observations that cell cholesterol loading upregulates expression of both proteins.
The observation that a few nonfoam cell macrophages expressed both proteins suggests that these proteins also may be coordinately regulated by other factors specific to plaques, such as cytokines or growth factors. Though several in vitro studies have elucidated how various plaque-associated cytokines, such as TGF-β1,33 IFN-γ,31 and macrophage-colony-stimulating factor,32 affect cellular apoE expression, the effect of these cytokines on HBP/vigilin expression is, as yet, unknown. Alternatively, these nonfoam cell macrophages may represent cells that have recently unloaded most of their sterol but have not yet downregulated their expression of apoE and HBP/vigilin.
The function of HBP/vigilin is unknown. Most studies suggest that HBP may be a complex protein with multiple functions. This protein contains 14 repeated KH domains that are usually found in nucleic acid–binding proteins, including the FMR protein involved in fragile X mental retardation and the α-poly (C)-binding protein associated with the α-globulin mRNA ribonucleic protein complex.23 The KH domains in HBP/vigilin have been shown to interact directly with polynucleotides.24 HBP/vigilin has also been found to be associated with cytoplasmic t-RNA-protein complexes isolated from human liver cells21 and from Xenopus liver bound to vitellogenin mRNA.22 HBP/vigilin has also been reported to contain a functional nuclear localization sequence and to be present in the nucleus.25 On the basis of these findings, it has been postulated that HBP/vigilin has several functions related to RNA metabolism, including protecting tRNA,21 stabilizing mRNA,22 and transporting RNA between the nucleus and cytoplasm.25 Despite these RNA-binding properties and the lack of a classic membrane-spanning domain, HBP/vigilin is tightly associated with cellular membranes,16,18,25 partially localizes to the plasma membrane when overexpressed in cells,16 and was identified in a screen for adipocyte surface proteins (Scherer PE, Bichel PE, Kotler M, Lodish HF, unpublished observations, 1997). Moreover, overexpression of HBP/vigilin increases HDL binding to cultured cells,16 and antibodies for HBP/vigilin can inhibit cell surface binding of HDL.34 These results raise the possibility that HBP/vigilin may play some role in recognition of extracellular molecules, particularly HDL.
Expression of HBP/vigilin appears to be regulated tightly and varies considerably between tissues and cell types.16–22 It has been suggested that expression of HBP/vigilin correlates with the degree of protein production among various tissues.21 This hypothesis, however, does not explain why cholesterol loading of cultured J774 macrophages would cause a greater than sixfold increase in HBP/vigilin mRNA levels16 under conditions in which a net increase in total protein synthesis would not be expected. Also, the degree of expression of HBP/vigilin among cells of atherosclerotic lesions appears to reflect variations in intracellular lipid content. Interestingly, hearts of apoE−/− mice also contained lipid-laden myocytes that express higher levels of HBP/vigilin than do surrounding cells. In addition to cholesterol, estrogen and gonadotropin have been shown to induce HBP/vigilin mRNA in Xenopus liver and rat ovaries.20,22 One possible explanation for these findings is that HBP/vigilin plays a role in coupling sterol metabolism to protein synthesis.
In summary, this study demonstrates the following novel findings regarding HBP/vigilin expression: (1) HBP/vigilin expression is constitutive in endothelium, (2) nonendothelial HBP/vigilin expression is specific for atherosclerosis, (3) HBP/vigilin expression is associated with lipid accumulation in macrophages but also is detected in a minor subset of macrophages without lipid accumulation, (4) HBP/vigilin expression in macrophages is associated with apoE expression in human atherosclerotic plaques, and (5) HBP/vigilin is expressed in lipid-laden cardiac myocytes of apoE−/− mice. Also, we demonstrate the utility of double labeling with ORO and cell-specific antibodies for identifying foam cell types within atherosclerotic lesions. The findings that HBP/vigilin is expressed in human atherosclerotic lesions, that it is coexpressed with apoE, and that is expressed in murine cardiac foam cells suggest a role for HBP/vigilin in cellular cholesterol metabolism. Further studies are required to elucidate the precise mechanisms by which it participates in these processes.
This study was supported in part by Grants-in-Aid 94-WA-518R and 96-WA-304 (KDO) from the American Heart Association, Washington Affiliate, Seattle, WA, and by grants DK02456 (KDO), HL02788 (KDO), HL18645 (JFO), HL50367 (RCL, JFO), and HL47151 (CEA) from the National Institutes of Health, Bethesda, MD. The authors gratefully acknowledge Lisa Anne Billings for assistance in manuscript preparation and Winnie Chiu, Marina Ferguson, Susan Rozell and Randy Small for expert technical support.
Fielding CJ, Fielding PE. Molecular physiology of reverse cholesterol transport. J Lipid Res. 1995;36:211–228.
Oram JF, Yokoyama S. Apoliprotein-mediated removal of cellular cholesterol and phospholipids. J Lipid Res. 1996;37:2473–2491.
Mahley RW. Apolipoprotein E: cholesterol transport protein with expanding role in cell biology. Science. 1988;240:622–630.
Basu SK, Brown MS, Ho YK, Havel RJ, Goldstein JL. Mouse macrophages synthesize and secrete a protein resembling apolipoprotein E. Proc Natl Acad Sci U. S. A.. 1981;78:7545–7549.
Mazzone T, Gump H, Diller P, Getz GS. Macrophage free cholesterol content regulates apolipoprotein E synthesis. J Biol Chem. 1987;262:11657–11662.
Mazzone T, Basheeruddin K, Poulos C. Regulation of macrophage apolipoprotein E gene expression by cholesterol. J Lipid Res. 1989;30:1055–1064.
Dory L. Synthesis and secretion of apoE in thioglycolate-elicited mouse peritoneal macrophages: effect of cholesterol efflux. J Lipid Res. 1989;30:809–816.
Zhang WY, Gaynor PM, Kruth HS. Apolipoprotein E produced by human monocyte-derived macrophages mediates cholesterol efflux that occurs in the absence of added cholesterol acceptors. J Biol Chem. 1996;271:28641–28646.
Smith JD, Miyata M, Ginsberg M, Grigaux C, Shmookler E, Plump AS. Cyclic AMP induces apolipoprotein E binding activity and promotes cholesterol efflux from a macrophage cell line to apolipoprotein acceptors. J Biol Chem. 1996;271:30647–30655.
Rosenfeld ME, Butler S, Ord VA, Lipton BA, Dyer CA, Curtiss LK, Palinski W, Witztum JL. Abundant expression of apoprotein E by macrophages in human and rabbit atherosclerotic lesions. Arterioscler Thromb. 1993;13:1382–1389.
Salomon RN, Underwood R, Doyle MV, Wang A, Libby P. Increased apolipoprotein E and c-fms gene expression without elevated interleukin 1 or 6 mRNA levels indicated selective activation of macrophage functions in advanced human atheroma. Proc Natl Acad Sci USA.. 1992;89:2814–2818.
McKnight GL, Reasoner J, Gilbert T, Sundquist KO, Hokland B, McKernan PA, Champagne J, Johnson CJ, Bailey MC, Holly R, O’Hara PJ, Oram JF. Cloning and expression of a cellular high density lipoprotein-binding protein that is up-regulated by cholesterol loading of cells. J Biol Chem. 1992;267:12131–12141.
Kruse C, Grunweller A, Notbohm H, Kugler S, Purschke WG, Muller PK. Evidence for a novel cytoplasmic tRNA-protein complex containing the KH-multidomain protein vigilin. Biochem J. 1996;320:247–252.
Dodson RE, Shapiro DJ. Vigilin, a ubiquitous protein with 14 K homology domains, is the estrogen-inducible vitellogenin mRNA 3′-untranslated region-binding protein. J Biol Chem. 1997;272:12249–12252.
Burd CG, Dreyfuss G. Conserved structures and diversity of functions of RNA-binding proteins. Science. 1994;265:615–621.
O’Brien KD, Chait A, Gown AM, Nagano Y, Kita T. Probucol treatment affects the cellular composition but not anti-oxidized low density lipoprotein immunoreactivity of atherosclerotic plaques in Watanabe heritable hyperlipidemic rabbits. Arterioscler Thromb. 1991;11:751–759.
O’Brien KD, Gordon D, Deeb SS, Ferguson M, Chait A. Lipoprotein lipase is synthesized by macrophage-derived foam cells in human coronary atherosclerotic plaques. J Clin Invest. 1992;89:1544–1550.
O’Brien KD, Allen MD, McDonald TO, Chait A, Harlan JM, Fishbein DP, McCarty J, Ferguson M, Hudkins K, Benjamin CD, Lobb R, Alpers CE. Vascular cell adhesion molecule-1 is expressed in human coronary atherosclerotic plaques: implications for the mode of progression of advanced coronary atherosclerosis. J Clin Invest. 1993;92:945–951.
Nishina PM, Verstuyft J, Paigen B. Synthetic low and high fat diets for the study of atherosclerosis in the mouse. J Lipid Res. 1990;31:859–869.
Basu SK, Goldstein JL, Brown MS. Independent pathways for secretion of cholesterol and apolipoprotein E by macrophages. Science. 1983;219:871–873.
Querfeld U, Ong JM, Prehn J, Carty J, Saffari B, Jordan SC, Kern PA. Effects of cytokines on the production of lipoprotein lipase in cultured human macrophages. J Lipid Res. 1990;31:1379–1386.
Nazih-Sanderson F, Lestavel S, Nion S, Rouy D, Denefle P, Fruchart JC, Clavey V, Delbart C. HDL3 binds to glycosylphosphatidylinositol-anchored proteins to activate signalling pathways. Biochim Biophys Acta. 1997. In Press.