Science 292 (5520): 1310, 2001
What determines the amount of LDL in plasma, and why do so many people have enough LDL to cause heart attacks? In medicine, common problems are often solved by studies of uncommon genetic diseases. In the case of LDL, answers have emerged from unraveling the aberrant genes underlying four disorders that elevate plasma LDL and cause premature heart attacks. The final two molecular defects of this quartet have been described within the last 6 months (3-5), and one of them is reported on page 1394 of this week's Science (5). Remarkably, all four defects raise the amount of plasma LDL by impairing the activity of hepatic LDL receptors (LDLRs), which normally clear LDL from plasma.
LDLs are composed of a collection of spherical particles with an average diameter of 22 nanometers. The average LDL particle contains a hydrophobic core of 1500 molecules of cholesteryl ester surrounded by a polar coat composed primarily of phospholipids and a 513-kilodalton protein called apolipoprotein B-100 (apoB-100) (6). LDLs are secreted from the liver as larger precursor particles (average diameter, 55 nanometers) called very low-density lipoproteins (VLDLs), whose cores contain triglycerides as well as cholesteryl esters (see the figure). VLDL-triglycerides are removed in the capillaries of muscle and adipose tissue, and the particles then undergo exchange reactions with other lipoproteins. The net effect is to reduce the size of VLDLs, restricting the core lipids to cholesteryl esters, and removing all proteins except apoB-100, thereby producing LDLs (6).
|Not all in the diet. The quartet of hypercholesterolemias. In these four monogenic diseases, the inability of defective LDLRs to remove cholesterol-carrying LDLs from plasma causes an increase in plasma LDL and the deposition of atherosclerotic fatty plaques in arteries, leading to heart disease. The mutant gene products of the cholesterol quartet are shown in red; also depicted are the points where their normal counterparts act in the cholesterol pathway. [Not shown are the intermediate density lipoproteins (IDL), which are highly atherogenic intermediates in the conversion of VLDL to LDL (6, 7).]|
LDLs circulate in human plasma with a mean life-span of 2.5 days. The particles are removed from plasma when apoB-100 binds to LDLRs on the surface of liver cells (7). The bound LDL is internalized by receptor-mediated endocytosis in coated pits, the internalized LDLs are degraded in lysosomes, and the liberated cholesterol enters the cellular cholesterol pool (see the figure). The number of LDLRs expressed by liver cells is controlled by negative-feedback regulation (7). When the concentration of cholesterol in hepatocytes rises, transcription of the LDLR gene is suppressed, and LDL is retained in plasma. In contrast, when hepatic cholesterol levels fall, LDLR gene transcription is induced, LDL is taken up more rapidly, and the amount of LDL in plasma falls. This mechanism explains most of the LDL-lowering action of statins, which deplete hepatic cholesterol by blocking an enzyme in the cholesterol synthetic pathway (7).
Hepatic pools of cholesterol are influenced by many variables, including absorption of dietary cholesterol from the intestine and excretion of hepatic cholesterol into bile. Recent genetic studies (3, 4) have shown that both absorption and excretion of cholesterol are controlled by a pair of adenosine triphosphate-binding cassette (ABC) transporters, called ABCG5 and ABCG8, that are believed to act in concert to pump cholesterol out of cells (see the figure). In the intestine, ABCG5 and ABCG8 re-excrete cholesterol that has entered gut epithelial cells from the gut lumen, thereby limiting cholesterol absorption. In hepatocytes, ABCG5 and ABCG8 secrete cholesterol into bile (3, 4).
The quartet of monogenic disorders that cause LDL to accumulate in plasma include familial hypercholesterolemia (FH), first described in 1938 (see the table) (8). The primary defect in this disorder, a deficit of LDLRs, was discovered in 1973 (7). Recently, more than 600 mutations in the LDLR gene have been identified in patients with FH (7). FH patients who are heterozygous for the LDLR defect produce one-half the normal number of LDLRs, and on average they have a 2.5-fold elevation in the number of LDL particles in plasma. The incidence of heterozygous FH is at least 1 in 500 people in all populations so far studied, making this disorder one of the most common monogenic diseases. It is the most frequent cause of premature coronary heart disease resulting from a single gene defect, and accounts for 5% of heart attacks in patients 60 years of age or less. The rare FH homozygotes (1 in 1 million) have LDL levels that are 6- to 10-fold above normal. These individuals have virulent coronary atherosclerosis, often dying from heart attacks in childhood (7).
The second of these genetic disorders, familial ligand-defective apoB-100 (FDB), was distinguished from FH in 1986 (9). This disease is caused by mutations in the gene encoding apoB-100, which reduce the protein's ability to bind to LDLRs, thereby retarding plasma clearance of LDLs (6). Heterozygous FDB is common in Europeans (1 per 1000). The syndrome is similar to heterozygous FH, although not as severe. The rare FDB homozygotes have higher levels of LDL than the FDB heterozygotes (see the table).
|FOUR MONOGENIC DISEASES THAT ELEVATE PLASMA LDL AND CAUSE HEART ATTACKS|
|Human disease||Prevalence in population||Typical plasma LDL-cholesterol level*||Mutant gene product||Mechanism for decreased LDL receptor function|
|Familial hypercholesterolemia||LDL receptor||Nonfunctional receptors|
|Heterozygous||1 per 500||300|
|Homozygous||1 per million||650|
|Familial ligand defective apoB-100||apoB-100||Decreased binding of LDL to receptors|
|Heterozygous||1 per 1000||270|
|Homozygous||<1 per million||320|
|Autosomal recessive hypercholesterolemia||<1 per 10 million§||470||ARH||? Altered location of receptors in liver|
|Sitosterolemia||<1 per 10 million||100 to 600 depending on diet||ABCG5 and/or ABCG8||Suppression of receptor gene transcription|
|*Typical adult plasma LDL-cholesterol is 120 mg/dl in the United States (6). All populations. Primarily in individuals of European descent. §Primarily in individuals of Italian and Middle Eastern descent.|
The third member of the Cholesterol Quartet, the rare autosomal recessive disorder sitosterolemia, was delineated in 1973 (10). Affected individuals accumulate a unique form of LDL that contains abundant plant sterols (sitosterol, campesterol, and stigmasterol), in addition to cholesterol. Sterol accumulation results from two abnormalities: the increased absorption of dietary cholesterol and plant sterols, and the reduced excretion of these sterols into bile (11). Together, these defects lead to a buildup of cholesterol in the liver, which suppresses transcription of the LDLR gene, causing LDL to accumulate in plasma. Recently, two groups led by Hobbs (3) and Patel (4) have used positional cloning to trace the sitosterolemia defect to loss-of-function mutations in genes encoding the two ABC transporters, ABCG5 and ABCG8. The authors postulate that these two transporters may pair up as heterodimers and that a deficiency of either partner may abolish ABC transporter activity.
The Hobbs group (5) now reports the molecular defect in the fourth member of the Cholesterol Quartet, autosomal recessive hypercholesterolemia (ARH). This syndrome, whose severity approaches that of homozygous FH, is distinguished from FH on genetic grounds: Obligate heterozygous parents of ARH patients have normal plasma LDL, unlike the heterozygous parents of FH homozygotes, who have 2.5-fold elevations in plasma LDL (12).
Young adults and children with ARH exhibit severe hypercholesterolemia, premature coronary heart disease, and massive deposits of LDL-derived cholesterol in the skin. Isotopic tracer studies show that these individuals, like FH homozygotes, have a severe defect in the removal of LDL from plasma (13). Yet, LDLR activity in cultured fibroblasts from ARH patients was almost normal, and no mutations were found in the LDLR gene of these patients.
Hobbs and colleagues have now traced the molecular defect to the gene encoding a previously undescribed cytosolic protein, ARH, which contains a phosphotyrosine-binding (PTB) domain (5). Other PTB domains bind to the cytoplasmic tails of cell surface receptors that contain an NPXY motif (14). This tetrapeptide was originally identified in the LDLR, where it is essential for incorporation of the receptor into clathrin-coated pits during endocytosis (15). Although firm biochemical data are not yet available, it seems reasonable to speculate that ARH binds to the NPXY motif of LDLRs. This binding might facilitate the entry of receptors into coated pits, or it could participate in receptor cycling from the cell surface to endosomes and back.
In patients with ARH, why is LDLR activity disrupted in the liver in vivo but not in fibroblasts in vitro? There are two possible answers. First, in fibroblasts ARH activity might be replaced by another PTB-containing protein. Second, ARH may be required only in polarized epithelial cells, such as hepatocytes, in which the receptor must cycle selectively between the cytoplasm and only one surface of the cell.
The defect in each member of the Cholesterol Quartet elevates LDL by decreasing LDLR activity in the liver, directly or indirectly. This is consistent with data in experimental animals showing that it is impossible to raise plasma LDL levels markedly unless LDLR activity has been disrupted by genetic defects (16) or by down-regulation through ingestion of a high-cholesterol diet (17). All four diseases lead to premature heart attacks at an age that is roughly inversely proportional to the amount of LDL in plasma. This correlation establishes a direct causal link between plasma LDL and coronary atherosclerosis.
How does elevated plasma LDL produce the complex lesions of atherosclerosis (18), with their hallmark features of inflammation, necrosis, cellular proliferation, and lipid deposition? The answer may lie in the unsaturated fatty acids of the cholesteryl esters and phospholipids of which LDL is composed. One of these, arachidonic acid, is the precursor of inflammatory prostaglandins. Within the artery wall other unsaturated fatty acids of LDL can undergo oxidation to generate toxic aldehydes and epoxides that induce the production of inflammatory cytokines (19). Thus, although the principal job of LDL is to transport cholesterol, and although its metabolism is regulated in response to cellular demands for cholesterol, the pathological consequence of its accumulation may be traced to the localized deposition of its fatty acids at sites of damage in artery walls.
Most of the hypercholesterolemia in the population is attributable not to single-gene defects, but rather to high-fat diets compounded by poorly defined susceptibility genes. Findings from the quartet of monogenic cholesterol diseases emphasize the importance of LDLRs and suggest that certain susceptibility genes may dictate the degree of suppression of LDLRs in response to different amounts of dietary cholesterol. Two types of transcription factors have been shown to mediate this suppression. One is a family of membrane-bound transcription factors, sterol regulatory element binding proteins (SREBPs), that are liberated from cell membranes in response to cholesterol depletion. After their release, they enter the nucleus and activate genes involved in the synthesis of cholesterol and fatty acids and in the uptake of cholesterol by LDLRs (20). The second transcription factor is the liver X receptor (LXR), a nuclear receptor that is activated by certain oxygenated derivatives of cholesterol (21). Upon activation LXR increases production of ABCG5 and ABCG8 (3), and also of one of the SREBPs (SREBP-1c) (22, 23).
Recent progress in understanding the Cholesterol Quartet of monogenic diseases and exposure of the mechanisms underlying cholesterol regulation create a sense of optimism that new and more powerful ways may soon be found to raise LDLR activity and to lower plasma LDL, with the goal of preventing atherosclerosis and coronary artery disease.
The authors are in the Department of Molecular Genetics, University of Texas Southwestern Medical Center, Dallas, TX 75390-9046, USA. E-mail: email@example.com; firstname.lastname@example.org