The “best” of cholesterols, the “worst” of cholesterols: A tale of two receptors

Cardiovascular disease is the number one killer in the U.S., and atherosclerosis is the major cause of heart disease and stroke (1). It is widely appreciated that cholesterol plays an important role in atherogenesis. Normally, most cholesterol serves as a structural element in the walls of cells, whereas much of the rest is in transit through the blood or functions as the starting material for the synthesis of bile acids in the liver, steroid hormones in endocrine cells (e.g., adrenal gland, ovary, testes), and vitamin D in skin. The transport of cholesterol and other lipids through the circulatory system is facilitated by their packaging into lipoprotein carriers. These spherical particles comprise protein and phospholipid shells surrounding a core of neutral lipid, including unesterified (“free”) or esterified cholesterol and triglycerides. Risk for atherosclerosis increases with increasing concentrations of low density lipoprotein (LDL) cholesterol whereas risk is inversely proportional to the levels of high density lipoprotein (HDL) cholesterol (2, 3). The receptor-mediated control of plasma LDL levels has been well-defined (4, 5), and very recent studies have now provided new insights into HDL metabolism (6–11).

In 1974, Michael Brown, Joseph Goldstein, and colleagues began publishing a classic series of papers that described the receptor-mediated cellular metabolism of LDL (4, 12). Their work defined how the LDL receptor influences LDL metabolism in the body and helps to determine blood LDL levels. Fig. 1 summarizes in a simplified form the role of LDL in cholesterol transport. In brief, the liver synthesizes a precursor lipoprotein (very low density lipoprotein, VLDL) that is converted during circulation to intermediate density lipoprotein (IDL) and then to LDL (13). The majority of the LDL receptors expressed in the body are on the surfaces of liver cells, although virtually all other tissues (“peripheral tissues”) express some LDL receptors. LDL receptors, located in specialized indentations in the cell membrane called coated pits, specifically and tightly bind LDL. After binding, the receptor–lipoprotein complex is internalized by the cells via coated pits and vesicles, and the entire LDL particle is delivered to lysosomes, wherein it is disassembled by enzymatic hydrolysis, releasing cholesterol for subsequent cellular metabolism. This whole-particle uptake pathway is called “receptor-mediated endocytosis” (14). Cholesterol-mediated feedback regulation of both the levels of LDL receptors and cellular cholesterol biosynthesis help ensure cellular cholesterol homeostasis. Genetic defects in the LDL receptor in humans result in familial hypercholesterolemia, a disease characterized by elevated plasma LDL cholesterol and premature atherosclerosis and heart attacks (5). One attractive hypothesis for the deleterious effects of excess plasma LDL cholesterol is that the LDL enters the artery wall, is chemically modified, and then is recognized by a special class of receptors, called macrophage scavenger receptors, that mediate the cellular accumulation of the LDL cholesterol in the artery, eventually leading to the formation of an atherosclerotic lesion (15, 16). A major breakthrough in the pharmacologic treatment of hypercholesterolemia has been the development of the “statin” class of 3-hydroxy-3-methylglutaryl-CoA reductase inhibitory drugs (17). 3-Hydroxy-3-methylglutaryl-CoA reductase is the rate controlling enzyme in cholesterol biosynthesis, and its inhibition in the liver stimulates LDL receptor expression. As a consequence, both plasma LDL cholesterol levels and the risk for atherosclerosis decrease. The discovery and analysis of the LDL receptor system has had a profound impact on cell biology, physiology, and medicine.

Figure 1.

LDL and the LDL receptor in cholesterol transport. The liver secretes the triglyceride-rich VLDL, which is converted to IDL and then to cholesterol-rich LDL. Plasma LDL cholesterol levels are controlled by the receptor-mediated endocytic clearance of LDL from the circulation by LDL receptors. These receptors are expressed most abundantly in the liver but also are found in many other tissues (see text for details).

What about the counterpart of LDL: HDL and its cellular metabolism? HDL has been the subject of intense study for decades and much has been learned, especially about its constituent parts and its dynamic remodeling in the plasma. In rodents, in which HDL transports most of the plasma cholesterol, HDL has been shown to be an important source of cholesterol for biliary excretion and steroidogenesis. About 15 years ago, reports from Pitman and coworkers (18) and subsequently others indicated that there exists a mechanism for the delivery of HDL cholesterol to cells that is fundamentally different from receptor-mediated endocytosis (refs. 19 and 20, reviewed in refs. 21 and 7). This mechanism is called “selective lipid uptake” because, after HDL binds to cells, only some of the components of the HDL particle enters the cells. In particular, HDL cholesterol (primarily in the form of cholesteryl esters) is transferred efficiently and the lipid-depleted HDL particles subsequently dissociate from the cells and re-enter the circulation. In vivo, the highest levels of selective uptake are seen in the liver and steroidogenic organs. Fig. 2 indicates how selective uptake of the cholesteryl esters of HDL might be linked to the proposed role of HDL in removing free cholesterol from peripheral tissues (including the arterial wall and potential sites of atherosclerotic lesions). The overall HDL-mediated movement of cholesterol from peripheral tissues to the liver is called “reverse cholesterol transport” (22). In some species (e.g., humans but not mice), the cholesteryl esters in HDL also can be transferred to other lipoproteins for further transport and metabolism (23). The significance of the pioneering studies that identified the selective uptake pathway (7, 18–21) was not initially as widely appreciated as that of endocytosis, perhaps in part because the mechanism did not fit into the paradigm of receptor-mediated endocytosis and because there was considerable controversy regarding the existence, as well as the functions and characteristics, of HDL receptors.

Figure 2.

HDL and the HDL receptor SR-BI in cholesterol transport. HDL is thought to remove unesterified, or “free” cholesterol (FC) from peripheral tissues, after which most of the cholesterol is converted to cholesteryl ester (CE) by enzymes in the plasma. Subsequently, HDL cholesterol is efficiently delivered directly to the liver and steroidogenic tissues via a selective uptake pathway and SR-BI or, in some species, transferred to other lipoproteins for additional transport and metabolism.

By analogy with the LDL system, a key missing element in the study of HDL metabolism was a well defined HDL receptor, which could give a molecular and cellular handle on the system. Approximately 2 years ago, the class B type I scavenger receptor (SR-BI), was unexpectedly shown to be the first molecularly well characterized HDL receptor (6). When expressed on the surfaces of cultured mammalian cells, this protein bound HDL (apparently via its main protein component, apoA-I, ref. 24) and mediated selective uptake of HDL lipids. Furthermore, SR-BI was found to be expressed in mice (6) [and subsequently in rats (25) and humans (26, 27)] at high levels in precisely those tissues that previously had been shown to exhibit the bulk of selective uptake of HDL cholesterol in vivo (ref. 18, Fig. 2). The temporal and spatial expression of SR-BI during murine embryogenesis also was consistent with a role of SR-BI in delivering cholesterol to the developing fetus (28). Additional correlative evidence for a role for SR-BI in HDL cholesterol metabolism came from studies of the effects on SR-BI expression of hormones, which induce or suppress steroid hormone synthesis. In vivo studies with intact mice and rats, as well as murine and human cultured cell lines, established that SR-BI expression was regulated coordinately with steroidogenesis in the adrenal gland, testes, and ovary (29–33, 47, 48) and that, within these tissues, SR-BI specifically was expressed in the steroidogenic cells (29, 30, 47, 48). Additional in vitro evidence suggesting involvement of SR-BI in hepatic selective uptake has appeared (34).

A series of in vitro and in vivo studies last year (8–10), along with the paper by Varban et al. (11) in this issue of the Proceedings, have focused on more directly determining whether SR-BI is a physiologically relevant HDL receptor for selective lipid uptake. Temel et al. (10) showed that a SR-BI-specific blocking antibody could inhibit selective uptake of HDL cholesteryl esters and conversion of HDL-derived cholesterol to steroid hormones in cultured adrenocortical cells. This finding provided the first evidence that SR-BI was involved directly in mediating selective uptake in a physiologic system. In the first study, which directly manipulated SR-BI levels in vivo, Kozarzky et al. (8) used an adenovirus to overexpress SR-BI on both sinusoidal and canalicular surfaces of hepatocytes in mice. This resulted in the virtual disappearance of plasma HDL and a doubling of biliary cholesterol. This work strongly suggested that SR-BI may play key roles in hepatic HDL metabolism, in determining plasma HDL concentrations, and possibly in mediating cholesterol efflux from cells. Definitive evidence for the physiological role of SR-BI in HDL metabolism came with the report by Rigotti et al. (9) of the first targeted disruption (null mutation, no protein product) of the SR-BI gene in mice. Relative to wild-type controls, heterozygous and homozygous mutants had substantially increased plasma cholesterol concentrations (30–40% and ≈2.2-fold, respectively). There was a slight increase in HDL size in the heterozygous mutants and a substantial increase in the size of HDL and its heterogeneity in the homozygous mutants. These results established that the gene encoding SR-BI can play a key role in determining the levels of plasma HDL cholesterol in mice, almost certainly because reduced expression of SR-BI resulted in decreased selective cholesterol uptake in the liver. In addition, there was a striking gene dose-dependent decrease in adrenal gland cholesterol content in the null mutants (42 and 72% reductions, respectively), establishing that SR-BI plays a key role in providing cholesterol for accumulation of cholesterol stores in steroidogenic tissue in vivo. Based on the in vitro activity and tissue distribution of the human SR-BI (26, 27, 44), it is reasonable to suggest that SR-BI may play a similar role in controlling plasma HDL in humans. Thus, activity of SR-BI may influence the development and progression of atherosclerosis, and SR-BI is an attractive candidate for therapeutic intervention in this disease (8, 9).

Varban et al. (11) now report the second application of gene targeting technology to study SR-BI function. They generated mice with a large (≈13.5 kilobase) insertion in a putative promoter region ≈2-kilobase upstream of the first exon of the SR-BI gene (9, 27). This mutation appears to have no effects on the structure of the SR-BI protein but rather alters the levels of SR-BI expression. A noteworthy consequence of this insertion is that, in homozygous mutants, there was an ≈50% reduction in hepatic SR-BI expression, which was accompanied by a 51% increase in plasma HDL cholesterol. In these and other key properties (FPLC lipoprotein cholesterol profile and slight increase in HDL size), the homozygous insertional mutants remarkably closely resemble the previously described heterozygous null mutants (9). Thus, this study independently confirms that SR-BI can play a key role in determining murine plasma HDL levels. Analysis of plasma lipoproteins in the null mutants (9) in conjunction with the initial in vitro studies (6) and the blocking antibody work (10) provided strong indirect evidence that SR-BI mediates selective uptake in vivo. An important additional contribution by Varban et al. (11) is that they have measured directly selective HDL cholesteryl ester uptake in the insertional mutants and controls. They have obtained compelling additional evidence for the crucial role of SR-BI in hepatic selective uptake in vivo.

The mechanism by which the insertion described by Varban et al. (11) blunts SR-BI expression remains to be determined. It is of interest that this insertion substantially reduces basal levels of SR-BI expression in the adrenal gland but does not appear to interfere with the stress-mediated induction of adrenal SR-BI expression by ACTH via cAMP (30–32). Although little is known about the molecular mechanisms underlying the regulation of SR-BI expression (29–33, 46–48), Hobbs and colleagues (27) recently have identified consensus-binding site sequences for several transcription factors in the putative promoter of the human SR-BI gene. These include C/EBP (cAMP and gonadotrophin responsive in endocrine cells, as is SR-BI expression), SREBP-1 (mediates cholesterol-regulated gene expression), and SF-1 (activates expression of several genes involved with steroidogenesis). There is striking overlap, but not identity, in the tissue distribution of SF-1 and SR-BI expression. Cao et al. (27) showed that SF-1 is likely to play an important role regulating SR-BI expression in adrenocortical cells. Because the binding sequences for these transcription factors lie within the first kilobase upstream of the transcriptional start site of human SR-BI , they are presumably ≈1-kilobase downstream from the insertion in the murine gene generated by Varban et al., and therefore their activities might not be attenuated in the mutant mice.

All of the above cell biological, physiological, pharmacological, immunochemical, and genetic analyses of SR-BI now have provided definitive evidence for the function of SR-BI. It is a high affinity, cell-surface HDL receptor that mediates physiologically relevant selective cholesterol transport, and it plays a key role in controlling plasma HDL cholesterol concentration, HDL structure, and delivery of cholesterol to steriodogenic tissues (7, 9). However, many questions regarding the structure and function of SR-BI remain unanswered. The mechanism of SR-BI-mediated selective lipid uptake has yet to be defined. The clustering of SR-BI in caveolae on the surfaces of cultured cells (35) and in cannaliculi-like structures in the adrenal cortex (ref. 30; perhaps representing microvillar channels; ref. 36) raised the possibility that specialized membrane domains may play a role in the selective uptake process. Because of microscopic reversibility and observations in mice overexpressing SR-BI, the potential of SR-BI for mediating cholesterol efflux from cells was proposed (8) and subsequently demonstrated (37). Its physiological significance has not been established, although it could play a role in reverse cholesterol transport. An alternatively spliced form of SR-BI with an altered C-terminal cytoplasmic domain was discovered by Webb et al. (38). This isoform was initially called SR-BI.2 but perhaps should be designated SR-BII to conform to the standard scavenger receptor nomenclature (39). Little is known currently about the physiological consequences of the alternative splicing, but it seems likely that further exploration of this discovery should uncover interesting new insights.

In addition to HDL, SR-BI has been shown to bind to a diverse array of ligands. SR-BI along with another class B scavenger receptor, CD36, were the first anionic phospholipid receptors to be identified (40). This finding suggested that SR-BI might be involved in recognizing senescent or apoptotic cells. Such recognition has been observed in vitro (26, 41), but its in vivo significance, along with that of SR-BI’s binding chemically modified LDL (42), is unclear. Perhaps most relevant to fully describing normal lipoprotein metabolism, SR-BI binds native LDL with high affinity (42, 45), although LDL does not effectively compete with HDL binding (6). Because normal mice have very low steady–state plasma LDL concentrations, it apparently will be necessary to examine the potential physiological relevance of the LDL binding in other species (e.g., rabbit and primates) or in genetically altered murine models (43). It is possible that SR-BI may serve as a backup or alternative receptor to the critically important, classic LDL receptor. Finally, we do not know how HDL exerts its anti-atherosclerotic effects, and thus, it is uncertain whether increasing or decreasing hepatic SR-BI activity might help prevent atherosclerosis. Given the rapid pace of progress in this area in the past two years, it is reasonable to expect resolution of many of these issues in the near future.

ABBREVIATIONS

LDL

low density lipoprotein

HDL

high density lipoprotein

VLDL

very low density lipoprotein

IDL

intermediate density lipoprotein

SR-BI

class B type I scavenger receptor