- Email: [email protected]
Apolipoprotein-B-containing plasma lipoproteins in health and in disease
Apolipoprotein-B-Containing Plasma in Health and in Disease Lipoproteins Robert L. Hamilton Coronary heart disease is declining slowly in many affluent nations, although it remains their major cause of death and disability. In sharp contrast, many societies in Eastern Europe are experiencing a substantial increase in atherosclerotic heart attack, possibly due largely to diets rich in saturated fat and cholesterol and to smoking. Most fats and cholesterol are transported in blood plasma in lipoproteins. Many studies implicate excessiveblood levels of particles containing apolipoprotein B* (apo B) in the atherogenic process although apo-B-containing particles (very low density lipoproteins, low-density lipoproteins, and chylomicrons) are essential for good health as shown by the genetic diseaseof abetalipoproteinemia (abeta) in which theseparticles are absent. Recent researchhas identified a probable defect in abeta, the apparent absenceof microsomal triglyceride transfer protein (MTP) that may be obligatory for core lipidation of apo B in the rough endoplasmic reticulum (RER). This discovery coincides with the articulation of a novel concept called the two-step hypothesis of triglyceride-rich particle assembly in hepatocytes and enterocytesforming very low density lipoproteins and chylomicrons, respectively. The first step is predicted to be dependent on MTP core lipidating full-length apo B that is firmly bound to the RER membrane. Corelipidation of apo B releasesa small apo-B-rich particle from the RER membrane into the RER lumen. A larger triglyceride-rich but apo-Bdeficient particle is formed independently in the smooth endoplasmic reticulum (SER). Usually, an apo-B-rich small particle formed in the RER coalesceswith an apo-B-deficient largerparticle from the SER as the second step of assembly of nascent triglyceuide-rich particles for secretion. In severalconditions, small apo-B-rich particles formed in the first step in the RER are secreteddirectly into the blood, bypassing the second step. These new concepts of the mechanisms of origin of apo-B-containing plasma lipoproteins may soon facilitate dietary and pharmacologic interventions that lower blood levels of apo B, reducing the incidence of heart attack and stroke. (Trends Cardiovasc Med 1994;4: 13 l-1 39)
Robert L. Hamilton is at the Cardiovascular Research Institute, University of California, School of Medicine, San Francisco,
and the Department of Anatomy, CA 94143-0130, USA.
‘Abbreviations: abeta, abetalipoproteinemia; ACAT, acyl CoA cholesterol acyltransferase; apo B, apolipoprotein B; CE, cholesteryl ester; CURL, compartment of uncoupling of receptor and ligand; DGAT, diacylglycerol acyltransfemse; EM, electron microscopy; ER, endoplasmic reticulum; HDL, high-density lipoprotein: LDL, low-density lipoprotein: LDL-R, lowdensity lipoprotein receptor: mRNA, messenger ribonucleic acid: MTP, microsomal triglyceride transfer protein: MVB, multivesicular body; PDI, protein disulfide isomerase; RER, rough endoplasmic reticulum: SER. smooth endoplasmic reticulum: TG, triglyceride; and VLDL, very low density lipoprotein.
TCM
Vol. 4, No. 3, 1994
0 1994, Elsevier
Science
Inc.,
1050-I 738/94/$7.00
Living cells are comprised mostly of water and water-soluble substances, but lipids or nonpolar molecules are also vital for life. Phospholipids, for example, are essential for cellular membranes, without which there would be no subcellular organelles such as the nucleus, endoplasmic reticulum, Golgi, endosomes, and mitochondria. There would also be no plasma membranes controlling selective ingress and egress of substances between the outside world and the intracellular milieu. One vital function of the plasma membrane, in controlling amounts and kinds of molecules entering and exiting cells, depends largely upon the presence of another lipid, unesterified cholesterol. Plasma membrane unesterified cholesterol modulates the fluidity of the phospholipid environment in which the cell surface protein macromolecules exist. This sterol appears to regulate stearic functions
of
macromolecules
in
the
plasma
membrane such as receptors, enzymes, transporters, and ion pumps (Yeagle 1990). About 70%80% of cellular cholesterol is localized in the plasma membranes of many cell types. In those cells that have an extraordinary need for cholesterol, however, such as hepatocytes and steroidogenic cells, large amounts of cholesterol may be stored in a precursor pool for bile acid and steroid hormone biosynthesis, respectively. In complex organisms, storage and transport of this sterol occur largely as cholesteryl esters* (CBS) in which a single long-chain fatty acid is condensed to the 3-hydroxyl position of cholesterol. CE occurs in virtually all living tissues, usually in association with another oil, the triacylglycerols (TGs). TGs (or fats) provide most cellular ATP following hydrolysis by various lipases, releasing longchain fatty acids for p oxidation within mitochondria. In multicellular organisms, storage and transport of CEs and TGs are tightly regulated complex processes requiring the expression and modulation of many different proteins and nucleotides.
l
Origins
of Apolipoprotein
B
A unique protein essential for oil transport between cells and tissue fluids is apolipoprotein B (apo B), a very large and hydrophobic protein. Of the mammalian bodies’ countless different types of cells, only three known epithelia express apo-B mWA: (a) hepatocytes, (b) absorptive enterocytes, and (c) yolk-sac endoderm. A
131
CHYLOMICRON PATHWAY
CELL
HDL
1
CHYLOMICRON
a Figure 1. The plasma chylomicron transport pathway (a) and the plasma VL.DL and LDL transport pathways (b) depicted here are described in the text. Blue represents triacylglycerols, red cholesteryl esters, and yellow the LDL or B-lo&E receptors. FFA, free fatty acid; and MG, monoglyceride. From Have1 (1982). fourth is unidentified placental cells (Demmer et al. 1986). These three epithelial cells synthesize and secrete TG and CE molecules contained in a tiny oil droplet (core lipids) surrounded with a beltlike band of apo B (Phillips and Schumaker 1989). There are two molecular forms of apo B, but only one gene. A unique biologic process, called RNA editing, alters the transcript encoded in the apo-B message, arresting translation at a specific nucleotide that is changed to a stop codon. This results in the premature termination of apo-B mRNA translation and the formation of apo B-48 (that is, about one-half of the total apo-B mFWA message). In the absence of the editing event, the fulllength message is translated, giving rise to apo B-100 (Kane and Have1 1994, Garcia et al. 1992). Apo B-48 has an apparent molecular mass of 264 kD whereas apo B-100 is about 550 kD (Kane and Have1 1994). In most mammals, including hu-
132
mans, this editing to produce apo B-48 occurs largely in small intestinal absorptive enterocytes during formation of chylomicrons, the triglyceride-rich particles that transport dietary TG and cholesterol derived from the absorption of foodstuffs, into the lymph and hence the bloodstream (Have1 and Kane 1989; Garcia et al. 1992). Apo B-100 is secreted by hepatocytes in very low density lipoprotein (VLDL) particles as a mechanism of exporting excess long-chain fatty acids that the liver takes up from blood, or synthesizes. In many species studied, including humans, hepatocytic VLDLs contain only apo B-100, but in rats nascent hepatocytic VLDL fractions contain both apo B-48 and apo B-100 (Hamilton et al. 1991). Dogs, horses, mice, and probably other species secrete both apo Bs from liver, based upon apo-B mRNA editing (Greeve et al. 1993). The significance of this is unknown. The rat visceral yolk-sac endoderm secretes TG-
01994
Elsevier
Science
Inc., 1050-1738/94/$7.00
rich particles containing B-100, although most of these nascent particles are of low-density lipoprotein (LDL) size and density (Plonne et al. 1992). The significance of placental expression of apo-B mRNA is unknown.
l
Plasma Metabolism of APO-B Particles (Figure la and b)
In the adult organism, virtually all apo-Bcontaining particles circulating in blood plasma are derived from hepatocytes and absorptive enterocytes. Even in the fasting state, the small intestine continues to secrete apo B in triglyceride-rich particles that are fewer and smaller than the chylomicrons formed during a meal (Jones and Ockner 1971). The number and size of chylomicrons (that is, the mass of absorbed fats) are determined largely by the amount of fat calories in a given meal. The metabolism of apo-B-48 particles from the gut and the metabolism of B- 100 particles from the liver are very similar initially, but diverge subsequently (Have1 1989). Each triglyceride-
TCM
Vol. 4, No.
3, 1994
VLDL-LDL
PATHWAYS
r\ j
.,:;.:,.:.:: :::::; .:.::z;: .+ bwotein .,.‘... k Lipose li ..:::I
8-loo:E Receptor
Receptor
-
b rich particle synthesized and secreted contains a single copy of either apo B-48 or apo B-100, which remains with its particle throughout its lifespan. Following secretion of nascent chylomicrons and nascent VLDLs into the interstitial fluids or blood plasma, apo A-l and phospholipids are rapidly unloaded, presumably onto high-density lipoproteins (HDLs), in exchange for C-I, C-II, and C-III apolipoproteins and apolipoprotein E (apo E). These exchangeable apolipoproteins are present usually in multiple copies on each triglyceride-rich particle (Have1 1989, Have1 and Kane 1989). Both chylomicrons and VLDLs undergo hydrolysis of their core triglycerides during passage through capillary beds that contain lipoprotein lipase dangling from the endothelial cells, bound to heparan sulfate (Have1 1989, Have1 and Kane 1989). Most lipoprotein lipase is in heart, skeletal muscle and, in the fed state, in adipose tissue. Although 70% 90% of chylomicron and VLDL content triglycerides are lost during this lipoprotein lipase hydrolysis, the resultant partiTCM
Vol. 4, No. 3, 1994
cles are only modestly smaller in diameter because they are spheres. The surface apolipoproteins C-I, C-II, and C-III, together with phospholipids and perhaps unesterified cholesterol, transfer back to HDLs during this lipolytic cascade. The resultant apo-B-containing particles are now called remnants of triglyceride-rich particles, and normally have a very short half-life (minutes to hours) in plasma (Have1 1989, Have1 and Kane 1989).
l
Subcellular Catabolism APO-B Particles (Figure
of 2)
The vast majority of remnants of chylomicrons and most VLDL remnants are removed from the blood by the liver by the process of receptor-mediated endocytosis and are fully degraded in secondary lysosomes within hepatocytes (Have1 and Hamilton 1988). At this point in the catabolic cascade of TG-rich particles, the pathway of a fraction of VLDL remnants diverges from that of chylomicron remnants (Have1 1989, Have1 and Kane 1989). Depending on the species
01994,
Elsevier
Science
Inc.,
1050-l
73&J/94/$7.00
studied and the technique used to make the measurements, an estimated 5% 50% of VLDL remnants containing apo B-100 are converted to LDLs (Have1 1989, Have1 and Kane 1989). The molecular mechanisms of forming LDL from VLDL remnants are, however, obscure. Unlike short-lived remnants, LDLs are cholesteryl ester-enriched long-lived particles (-3 days in humans) that can deliver cholesterol to many cells of the body by the LDL receptor (LDL-R) pathway by receptor-mediated endocytosis (Brown and Goldstein 1986, Goldstein and Brown 1989, Have1 and Kane 1989). The LDL-R binds to a short sequence of apo B-100 (absent from apo B-48), and clusters LDL particles in coated pits of the plasma membrane before pinching off to form coated vesicles. Those coated vesicles rapidly uncoat clathrin and adaptin proteins and fuse with other early endosomes and/or a stable endosomal compartment near the cell surface (McKanna et al. 1979, Brown and Goldstein 1986, Have1 and Hamilton 1988, Klumperman et al. 1993). This
133
0
Sinusoid
.
.
sis, recycling of receptors, downregulation of cell surface receptors, and degradation of endosomal contents by acid hydrolases in secondary lysosomes is a common metabolic pathway for countless different ligands in virtually all cells (McKanna et al. 1979, Brown and Goldstein 1986, Jost-Vu et al. 1986, Have1 and Hamilton 1988, Klumperman et al. 1993). Whereas it is widely accepted that many different cells utilize this LDL-R pathway to obtain cholesterol, the bulk (60% 70%) of LDL turnover in plasma in vivo occurs by receptor-mediated endocytosis back in hepatocytes (Brown and Goldstein 1986, Goldstein and Brown 1989, Have1 and Kane 1989), the same cells that started this entire metabolic process by synthesizing and secreting nascent VLDLs into plasma in the space of Disse (Have1 and Kane 1989, Hamilton et al. 1991). Thus, hepatocytes determine to a very large extent the concentration of apo-B-containing lipoproteins in plasma (Have1 and Hamilton, 1988).
.
l
Figure 2. This diagram depicts our current concepts of receptor-mediated endocytosis and intracellular processing of triglyceride-rich remnants of chylomicrons and VLDLs in hepatecytes.
The same or closelysimilar receptor-mediated processesoccur during catabolism of cholesteryl ester-rich LDLs both in hepatocytes and in cellsof peripheral tissuesas described in the text. L’, primary lysosome; L*, secondary lysosome; and RRC, receptor recycling compartment. From Have1&d Hamilton (1988). early endosomal compartment has been dubbed CURL for compartment of uncoupling of receptor and Zigand. A proton pump in CURL lowers the pH to -6.0, causing the dissociation of apo B-100 from the LDL-R. Most LDL-R in CURL becomes sequestered into a membranous appendage that is enriched in receptors that recycle many times between early endosomes and the cell surface (RRC; see Figure 2). In addition, those membrane-bound receptors that are destined to be downregulated are carried with surface membrane pinching off into the lumen of CURL, forming small lipid bilayer vesicles. Because these vesicles become mixed with the contents (McKanna et al. 1979), they will be degraded by acid hydrolases in secondary lysosomes. Thus, CURL gives rise to nascent multivesicular bodies (MVBs) growing by endosomal fusions near the cell surface and at the same time sequestering receptors into appendages that pinch off and return long-lived recycling receptors
134
to the plasma
membranes
and
internalizing some receptors for downregulation (Have1 and Hamilton 1988). Upon reaching a limited size of about 0.4-0.5 i.un, MVBs become rapidly transported across the cytoplasm on microtubules to the cell center, specifically to the transGolgi network, where they fuse with coated blebs that deliver lysosomal enzymes into the MVB contents (Have1 and Hamilton 1988, Klumperman et al. 1993). Lysosomal enzymes synthesized in the rough endoplasmic reticulum (RER) are concentrated in coated blebs of the truns-Golgi network, bound to mannose6-phosphase receptors also concentrated in the coated blebs (Klumperman et al. 1993). Thus, MVBs represent the last endosomal (prelyosomal) compartment (Have1 and Hamilton 1988) that transforms into secondary lysomes adjacent to the Golgi apparatus (Jost-Vu et al. 1986). The mannosed-phosphate receptor and perhaps other macromolecules are recycled from MVBs back to the Golgi (Klumperman et al. 1993). This process of receptor-mediated endocyto01994,
Else&x
Science
Inc.,
1050-1738/94/$7.00
Apo B in Atherogenesis
It is now widely societies
consuming
(Figure
recognized large
3)
that in
amounts
of
cholesterol and saturated fats, particularly of porcine and bovine origins, high levels of apo-B-containing plasma lipoproteins represented largely by LDL cholesterol concentrations and possibly delayed clearance of remnants of triglyceride-rich particles (Have1 1994, Schneeman et al. 1993) are both leading causal factors of atherogenesis leading to heart attack, a major cause of death (Brown and Goldstein 1986, Goldstein and Brown 1989, Have1 and Kane 1989). Evidence is mounting steadily that remnants of triglyceride-rich particles can be atherogenic (Zilversmit 1979, Have1 1990, Nordestgaard and ‘Qbjaerg-Hansen 1992, Have1 1994). We used to say that this debilitating disease of atherosclerosis was confined largely to Western and/or affluent societies, such as the USA and Western Europe, which could afford excessive fat calories. Now, in Eastern Europe, many of those societies are experiencing a substantial increased incidence of heart attack (Figure 3), more than twice that of our Western societies, in which the rate has been declining gradually for about a dozen years (Uemura and Pisa 1988, Marmot 1992). TCM
Vol. 4, No. 3, 1994
Finland England and Wales
Czech
Ireland
Hungary
Belgium
France
FP Germany
Poland
1
I MA” I -40
-30 Change
-20 in mortality
How can this be explained? Partly diet and partly life-style changes in opposite directions may account for much of this increased heart disease. Whereas greatly reduced cigarette smoking, prudent dietary habits, and fitness awareness are contributing to decreased incidence of heart attack in affluent Western societies, just the opposite trends are evident in Eastern European countries. To feed the populace, the cheapest sources of calories predominate in the marketplace: pork in many different forms (Miller 1993). At the same time, cigarettes now are cheap and accessible, and the vast majority of people smoke them. The death and morbidity rates, particularly among men in their most productive years of life, are leaving a devastating impact on the peoples of Eastern European countries (Uemura and Pisa 1988, Marmot 1992, Miller 1993).
l
Too Much Apo B
or Too Little
Plasma
Two contrasting and rare genetic diseases that occur in humans are caused by defects in one of two of the major macromolecular players of plasma lipoprotein oil transport. The first and best understood is familial hypercholesterolemia, in which the LDL-R is defective, plasma cholesterol levels are grossly elevated (-1000 mg/dL), severe atherosclerosis occurs at a very early age, and heart attacks often occur in childhood (Brown and Goldstein 1986, Goldstein and Brown 1989, Have1 and Kane 1989). A closely similar disorder occurs in the Watanabe heritable hyperlipidemic (WHHL) rabbit, because of a single deletion of 12 nucleotides in the LDL-R mRNA (Brown and Goldstein 1986, TCM
Vol. 4, No. 3, I994
0
-10 1%)
fbmania}
10
-20
0
20 Change
40 in mortality (%I
60
Bl
Figure 3. This figure contrasts changes in coronary heart disease mortality (%) that is gradually decreasing in Western European countries (as in the USA), whereas it is increasing substantially in men in Eastern European societies. Men and women, ages 30-69, between 1970 and 1985. From Marmot (1992), with permission from Oxford University Press.
Yamamoto et al. 1986, Goldstein and Brown 1989). Homozygous WHHL rabbits rapidly develop severe atherosclerosis caused by high plasma levels of LDLs and remnants, apo-B-containing particles. Thus, although in humans many other factors such as cigarette smoking, hypertension, diabetes mellitus, and physical inactivity may contribute to atherosclerosis of the coronary and cerebral arteries, leading to heart attack and stroke, excessive levels of LDL cholesterol and/or remnants can by themselves cause this disease process (Goldstein and Brown 1989, Have1 and Kane 1989, Nordestgaard and Tybjaerg-Hansen 1992, Have1 1994). A second rare and debilitating disease, called abetalipoproteinemia (abeta), is caused by a defect in the assembly of apo-B-containing triglyceride-rich particles in which no chylomicrons, VLDLs, or LDLs are present in plasma (Kane and Have1 1994, Wetter-au et al. 1992). Cholesterol and TG levels are low and are found only in HDL fractions (Kane and Have1 1994). Organ abnormalities abound, such as spinocerebellar degeneration and retinitis pigmentosa, and death often occurs early in affected individuals. For many years, it was believed that the defect resulted from the genetic absence of apo B, because there was virtually none detectable in plasma lipoprotein fractions. With the development of Western and Northern blotting techniques, it was found that intestinal biopsy material of abeta individuals contains apo B-48 of
01994.
Elsevier
Science
Inc.,
1050.1738/94/$7.00
correct molecular weight, and apo-B mRNA levels severalfold higher than in normal human intestines (Kane and Have1 1994). What then is the defect in abeta? Wetterau et al. (1992) have discovered that biopsy material of the small intestine from four abeta subjects appears to lack completely a protein called microsomal triglyceride transfer protein, abbreviated MTP. Wetterau and various collaborators discovered and characterized this novel protein isolated from bovine liver microsomal membranes (Wetterau and Zilversmit 1985). MTP is a heterodimer of protein disulfide isomerase (PDI) (-55 kD) and another protein of -88 kD (Wetterau et al. 1990 and 1992). In vitro, this complex transfers phospholipids between liposomal bilayers, similar to several other phospholipid transfer proteins; it is unique, however, because it moves the oils, TGs, and CEs (Wetterau and Zilversmit 1985). The terminal enzymes of triacylglycerol synthesis [diacylglycerol acyltransferase (DGAT)] and cholestetyl ester synthesis Cacyl CoA cholesterol acyltransferase (ACAT)] are membrane-bound enzymes with catalytic sites on the cytosolic face of the endoplasmic reticulum (ER). Thus, their product oils must be transported across the ER membrane to reach the ER lumen, from which secreted proteins are transported to the Golgi before reaching the cell surface. MTP is an appealing candidate to participate in nascent VLDL and nascent chylomicron assembly as a
135
mediator of transfer of oils from their sites of synthesis to the site of apo-B core lipidation. PDI is a luminal enzyme of the RER that facilitates proper folding of many nascent polypeptides as they enter the RER lumen from polyribosomes, realigning sulfhydryl groups to form specific disulfide bridges that determine proper protein folding (Yamamoto et al. 1986, Wetterau et al. 1990 and 1992). Perhaps as MTP transfers TG and CE molecules to apo B, the PDI component coordinately realigns the cysteine residues of apo B; there are 25 in apo B-100, 22 of which are in intrachain disulfide bridges (Kane and Have1 1990). Interestingly, the 88-kD component of MTP aggregates and becomes inactive when dissociated from PDI, suggesting that it is a highly hydrophobic protein maintained soluble in the aqueous environment of the RER by PDI (Wetterau et al. 1991). The discovery by Wetterau et al. (1992) may be a major breakthrough for two different reasons. First, profound insights may be revealed for understanding some of the mechanisms by which triglyceride-rich particle assembly occurs. Second, their observations on the intestinal biopsy material of abeta subjects probably reveals the basis for the metabolic defect in this genetic disorder. The Wetterau et al. (1992) paper contains two basic findings. First, they showed that soluble proteins from intestinal biopsy material from four abeta subjects were virtually unable to transfer radiolabeled TGs from donor to acceptor phospholipid liposomes in the test tube. This functional assay for MTP was linear for 60 min when soluble proteins from normal intestinal biopsy material were substituted in the same assay (Wetterau et al. 1992). Second, immunoblots with antisera against the bovine liver MTP identified the 88-kD subunit of correct molecular weight in soluble material from not only normal human intestinal biopsies, but also in the same material from subjects with two other human disorders of chylomicron formation, hypobetalipoproteinemia and Anderson’s disease. No protein of this molecular weight could be detected in immunoblots from intestinal material from four abeta subjects, even when 4-5 times more soluble proteins were applied to the SDS-PAGE, although the PDI monomer was detected (Wetterau et al. 1992).
136
These elegant studies strongly support the conclusion that the genetic defect in abeta is the apparent absence of correctly folded 88-kD protein subunits of MTP. It is probable that the biogenesis of genetically defective 88-kD subunits during their translation may cause improper folding of this subunit of the MTP heterodimer, which may prevent association with correctly formed PDI. In other words, the genetic defect is probably not the absence of the 88-kD subunits, but rather their inability to undergo the correct folding to form the functional heterodimer.
l
The “Two-Step” Hypothesis of APO-B Particle Assembly (Figure 4a and b)
A new and intellectually stimulating hypothesis is emerging to describe molecular events that may occur during the assembly of triglyceride-rich particles in the endoplasmic reticulum (ER). The new hypothesis is termed the two-step model of apo-B core lipidation in the ER. The hypothesis evolved from two widely different observations. The first observation is that immunolabeling at the electron-microscopic (EM) level showed heavy staining of rat hepatocytic RER for apo B in the absence of an apparent VLDL-sized particle, whereas VLDLsized particles in the smooth ER (SER) lacked apo-B immunostaining (Alexander et al. 1976). This finding is consistent with VLDL assembly occurring by two distinct steps, although we did not use those words. Second, HepG2 cells secrete most apo B in particles of HDL and LDL size and density, rather than VLDLs, and these transformed cells contain little or no detectable SER compartment by EM examination (Thrift et al. 1986). Hence, HepG2 cells secrete apo-Bcontaining particles formed by the first step in the RER because they lack the second step that makes triglyceride-rich particles that may require the SER. In studying the relationship between apo-B length and particle size (that is, core volume) in isolated lipoprotein particles from HepG2 cells, Spring and colleagues (1992) concluded that small apo-Bcontaining nascent lipoproteins must acquire additional core lipid in normal hepatocytes to form a nascent VLDL particle provided by a “second stage” of assembly.
01994,
Elsevier
Science
Inc.,
1050-1738/94/$7.00
The concept of a two-step VLDL assembly process suggests that there may be circumstances in nature in which the two stages may be dissociated. This appears to be a plausible explanation for our earlier observations of apo-B secretion in isolated perfused rat livers. When rat livers are perfused in a recirculating system for prolonged periods (5-6 h) in the absence of exogenous FFA as substrate to drive fi oxidation and VLDL secretion, a significant amount of apo B is recovered in particles in the HDL density fraction (Fainaru et al. 1977). Furthermore, when the same techniques of perfusion are carried out with livers of cholestatic rats (bile duct ligation for 24-48 h), twice as much apo B is recovered in the HDL fraction, and most of the rest of the per&sate apo B is in particles of LDL density at the expense of apo-B mass in nascent VLDL (Felker et al. 1982). Cholestatic rat liver lobes produced by bile duct obstruction have only one-tenth as much TG as normal rat liver lobes (Felker et al. 1982). This suggests that in the absence of sufficient fatty acid substrate, the larger triglyceriderich particles that are normally formed in the SER must be much reduced in number in cholestasis, and particles of LDL and HDL size that are formed in the RER in the first step are secreted. More direct biochemical support for the twostep hypothesis has been obtained recently by Elovson et al. (1992) and by the author (Hamilton et al. unpublished work). By developing a novel procedure to isolate large amounts (- 100 mg protein/ 10 g liver) of structurally intact RER membranes very rapidly from rat livers, we have been able to recover nascent precursors of VLDLs. Two groups of particles are highly enriched in apo B, are denser than VLDLs (in the HDL-toLDL density range), are smaller than VLDLs (125-250 A diameter), and contain small amounts of TGs and CEs. These two particles band at different densities in sucrose gradients, depending on the apo-B-TG ratio. The second particle is larger, the same size as nascent Golgi VLDLs (Hamilton et al. 1991) (-450 A diameter), and contains the same TG and CE mass, but completely lacks apo B by Western immunoblotting. This particle does, however, contain proteins that are unidentified. These particles appear to represent the ones that were predicted to exist in the SER TCM
Vol. 4, No. 3, 1994
and SER-RER junctions by immunostaining for apo B, which was absent from SER VLDL-sized particles (Alexander et al. 1976). Speculations Mechanisms
l
on Molecular of the Two-Step
SER LIpIds 1
I
Model
The following speculations are an attempt to bring together the two-step model and the observations by Wetterau et al. (1992) on the role(s) of MTP in the RER in core lipidation of apo B in triglyceride-rich particle assembly. Apo B is constitutively expressed by its mRNA in both adult liver and small intestine (Have1 and Kane 1989), and newly translated full-length apo B first becomes firmly associated with the RER membrane before it is found in an isolatable water-soluble lipoprotein particle in the ER lumen (Bijstrom et al. 1986, Thrift et al. 1992, Dixon and Ginsberg 1993). If sufficient substrate lipids are present in the cell, most of this RER-bound apo B becomes core lipidated, enabling its dissociation from the RER membrane as a small lipoprotein particle formed in the first step. When core lipid substrates (TGs and CEs) are not present in sufficient amounts to saturate this process, RER membranebound apo B undergoes local proteolytic degradation, a postulated regulatory process in apo-B secretion (Thrift et al. 1992, Dixon and Ginsberg 1993). The first step is predicted to be an MTP-dependent event whereby sufficient oils (TGs and CEs) are transferred from their cytosolic
Nascent VLDL
n
Apparatus
(
)
^^ n
Figure 4. The morphologic basis (a) and the biochemical basis (b) for the two-step hypothesis of nascent VLDL assembly are depicted. (a) Based on the absence of immunoperoxidase staining for apo B on VLDL-sized particles in the SER (solid black circles), whereas the tiny x’s represent locations of positive staining for apo B. (b) Based on characterization of small apo-B-rich particles and apo-Bdeficient particles of VLDL size isolated from the contents of a novel ribosomal enriched fraction from rat liver. The first step (I) depicts the initial core lipidation of apo B in the rough endoplasmic reticulum membrane, releasing apo B in a small lipoprotein particle into the lumen. The second step (2) occurs when a small apo-B-rich particle coalesces with an apo-B-deficient triglyceride-rich particle formed in the smooth endoplasmic reticulum completing nascent VLDL apo-B core lipidation. (a) From Alexander et al. (1976), with permission from Rockefeller University Press.
TCM
Vol. 4, No. 3, 1994
01994.
Elsevier
Science
Inc.,
1050-l 738/94/$7.00
137
sites of synthesis into hydrophobic domains of apo B residing within the RER membranes, perhaps by virtue of its binding to the RER phospholipids and/or to special apo-B-binding RER proteins. If the hypothesis is correct that a portion of RER-bound apo B is cytosolic (Thrift et al 1992) this proximity to the active sites of DGAT and ACAT (the enzymes that form TGs and CEs) may facilitate the initial core lipidation of apo B. When the MTPdependent process presents sufficient molecules of TGs and CEs that are more hydrophobic than the RER membranebinding sites, apo B may undergo structural reorganization to enable it to dissociate from its RER membrane-binding domains. When a critical mass of TGs and CEs has been transferred by MTP to interact with a sufficient number of hydrophobic domains of apo B, nucleation may produce a small core lipid-containing particle that is released from the RER membrane into the RER lumen as the completion of the first step (Figure 4b). This interpretation is consistent with recent molecular biologic experiments in which different apo-B constructs were inserted into both nonhepatic cells and hepatoma cells. Transfection of the Nterminal 53% of apo B into hepatoma cells led to the secretion of apo B-53 in lipoprotein particles. When the apo B-53 construct was transfected into Chinese hamster ovary cells (which lack MTP), no detectable apo B was secreted, because it could not be dissociated from RER membranes, where it was rapidly degraded by a protease (Thrift et al. 1992). Surprisingly, when an apo-B-15 construct was used in the same type of experiments, both cell types secreted apo B-15, but apparently in a lipid-free or lipid-poor form (Thrift et al. 1992). These results suggest the simple idea that apo B-15 is too short to become incorporated into RER membranes and therefore is translocated into the RER lumen like any soluble secretory protein, bypassing core lipidation events in those few cell types that express MTP. This predicts that some minimal length of apo B, about B-30 in humans (Linton et al. 1993), may be necessary to become firmly bound to the RER membrane and, as a consequence, to be core-lipidated in cells expressing MTP. If substrates are not adequate to supply sufficient amounts of oils to RER-bound apo B by those cells expressing MTP, or if cells do not express
138
MTP, RER-bound apo B undergoes proteolytic degradation. The two-step hypothesis of VLDL assembly may also explain some observations recently described in humans with a genetic disease designated hypobetalipoproteinemia. This disorder is characterized by truncated apo Bs of different lengths that are often found in HDL and LDL particles and associated with low plasma levels of apo B and cholesterol (Linton et al. 1993). The most severe phenotypes of this disorder may have clinical signs that are indistinguishable from abeta subjects requiring treatment, particularly dietary supplementation of lipid-soluble vitamins (Linton et al. 1993). One possible explanation for these truncated apo Bs in HDLs and LDLs is that some truncated forms of apo B may be long enough (> apo B-30) to form the small particles in the RER in the first step of VLDL assembly. Hence, some of these particles may be secreted directly because they may not be efficient in coalescing with apo-B deficient TG-rich particles of the SER. In abeta, the genetic absence of MTP would curtail the release of apo B from the RER membrane, completely preventing triglyceride-rich particle formation. The VLDL-sized particles lacking apo B that are formed independently of apo B in the SER may also require MTP to form the microemulsion particle, but this process would probably also require other unknown proteins and events to occur. Even if apo-B-deficient triglyceriderich particles were to be formed in the SER of hepatocytes or enterocytes, they most likely would not be released without apo B, the presumed protein signal enabling microemulsion oil droplets to transit the secretory pathway. The twostep hypothesis identifies two important areas for future research aimed at discovering mechanisms of triglyceriderich particle assembly by hepatocytes and absorptive enterocytes. The first is to unravel the molecular mechanisms by which MTP core lipidates apo B in the RER (note that no model is presented for this in Figure 4b). The second is to identify those special proteins and processes that form the microemulsion particles in the SER compartment in the absence of apo B. The fact that small apo-B-rich particles and larger apo-Bdeficient particles coexist in the ER lumen indicates that, to form a nascent
01994,
Elsevier science Inc.,
1050-1738/94/$7.00
VLDL particle, coalescence between these two particles must occur, presumably at the RER-SER junction. The two-step hypothesis of apo-B core lipidation and the discovery of the essential role of MTP in triglyceride-rich particle assembly have enormous potential clinical value. Together, these new insights may soon enable understanding of the molecular processes and therefore the regulation of VLDL and chylomicron secretion by hepatocytes and enterocytes, respectively. Once these processes are understood, pharmaceutical intervention, in addition to prudent diets, may reduce the first and/or the second steps such that apo-B-containing plasma lipoproteins are maintained at lower levels that greatly reduce the incidence of atherosclerotic disease, particularly heart attack and possibly stroke. For example, elevated concentrations of another class of apo-B-containing particles in human plasma, called lipoprotein(a), is another risk factor for atherosclerotic heart disease (Scanu 1991), but it may be even a greater risk factor for stroke (Nagayama et al. 1994, Jiirgens et al. 1994).
l
Acknowledgments
The author’s research is supported by NHLBI Arteriosclerosis Specialized Center of Research HL-14237. He is a member of the senior staff of the Cardiovascular Research Institute and Professor of Anatomy at the University of California, School of Medicine, San Francisco, Cali-. fomia.
References Alexander CA, Hamilton RL, Have1 RJ: 1976. Subcellular localization of B apoprotein of plasma lipoproteins in rat liver. J Cell Biol 69:241-263. Biistrom K, Wettesten M, Boren J, et al.: 1986. Pulse-chase studies of the synthesis and intracellular transport of apolipoprotein B 100 (apo B 100) in Hep G2 cells. J Biol Chem 261:13,800-13,806. Brown MS, Goldstein JL: 1986. A receptormediated pathway for cholesterol homeostasis. Science 232:34-47. Demmer LA, Levin MS, Elovson J, et al.: 1986. Tissue-specific expression and developmental regulation of the rat apolipoprotein B gene. Proc Nat1 Acad Sci USA 83:8102-8106. Dixon JL, Ginsberg HN: 1993. Regulation of
TCM Vol. 4, No. 3, 1994
hepatic secretion of apolipoprotein containing lipoproteins: information tained from cultured liver cells. J Lipid 34167-179.
BobRes
Elovson J, Bell GT, Doolittle MH, Phillips M: 1992. A two-step model for very low density lipoprotein assembly: characterization of intermediate lipoprotein particles [abst 27531. Circulation 86(Suppl l):PI-692. Fainaru M, Felker TE, Hamilton RL, Have1 RJ: 1977. Evidence that a separate particle containing B-apoprotein is present in highdensity lipoproteins from perfused rat liver. Metabolism 26:999-1004. Felker TE, Hamilton RL, Vigne J-L, Have1 RJ: 1982. Properties of lipoproteins in blood plasma and liver perfusates of rats with cholestasis. Gastroenterology 83:652-663. Garcia ZC, Poksay KS, Bostrom K, et al.: 1992. Characterization of apolipoprotein B mRNA editing from rabbit intestine. Arterioscler Thromb 12:172-179. Goldstein JL, Brown MS: 1989. Part 7: lipoprotein and lipid metabolism disorders. In Striver CR, Beaudet AL, Sly WS, Valle D, eds. The Metabolic Basis of Inherited Disease. New York, McGraw-Hill, pp 121% 1250. Greeve J, Altkemper I, Dieterich J-H, Greten H, Windler E: 1993. Apolipoprotein B mRNA editing in 12 different mammalian species: hepatic expression is reflected in low concentrations of apo B-containing plasma lipoproteins. J Lipid Res 34:1367-1383. Hamilton RL, Moorehouse A, Isolation and properties of teins from highly purified Golgi fractions. J Lipid Res
Have1 RJ: 199 1. nascent lipoprorat hepatocytic 32:529-543.
Have1 RJ: 1982. Approach to the patient with hyperlipidemia: symposium on lipid disorders. Med Clin North Am 66:319-321.
Metabolic Basis York, McGraw-Hill,
of Inherited pp 1129-l
Disease. 164.
New
Jones AL, Ockner RK: 1971. An electron microscopic study of endogenous very low density lipoprotein production in the intestine of rat and man. J Lipid Res 12:580-589. Jost-Vu E, Hamilton RL, Homick CA, Belcher JD, Have1 RJ: 1986. Multivesicular bodies isolated from rat hepatocytes: cytochemical evidence for transformation of multivesicular bodies into secondary lysosomes by fusion with primary lysosomes. Histochemistry 85457-466. Jtirgens G, Taddei-Peters WC, Koltringer P, et al.: 1994. Lipoprotein(a) concentration is a better predictor for risk of cerebrovascular disease than carotid artery stenosis or other lipoprotein levels [abst]. Stroke 25:247. Kane JP, Have1 RJ: 1994. Disorders of the biogenesis and secretion of lipoproteins containing the B apo-lipoproteins. In Striver CR, Beaudet AL, Sly WS, Valle D, eds. The Metabolic Basis of Inherited Disease. New York, McGraw-Hill (in press). Klumperman J, Hille A, Veenendaal T, et al.: 1993. Differences in the endosomal distributions of the two mannose-6-phosphate receptors. J Cell Biol 121:997-1010. Linton MF, Farese RV, Young Familial hypobetalipoproteinemia. Res 34:521-541.
SG:
1993. J Lipid
Marmot M: 1992. Coronary heart disease: rise and fall of a modem epidemic. In Marmot M, Elliott P, eds. Coronary Heart Disease Epidemiology: From Aetiology to Public Health. Oxford, Oxford University Press, pp 3-19. McKanna JA, Haigler HT, Cohen SC: 1979. Hormone receptor topology and dynamics: morphological analysis using ferritinlabeled epidermal growth factor. Proc Nat1 Acad Sci USA 7615689-5693. Miller J: 1993. A change Republic. UCSF Mag
of heart 14:23-25.
for the Czech
Have1 RJ: 1989. Contrasts and similarities between the metabolism of intestinal and hepatic lipoproteins: the role of particle size. In Windler E, Greten H, eds. Intestinal Lipid and Lipoprotein Metabolism. Munich, W Zuckschwerdt, pp 168-173.
Nagayama M, Shinohara Y, Nagayama T: 1994. Lipoprotein(a) and ischemic cerebrovascular disease in young adults. Stroke 25174-78.
Have1 RJ: 1990. Role of lipoproteins in progression sis. Circulation 8 1:694-696.
triglyceride-rich of atherosclero-
Nordestgaard BG, Tybjaerg-Hansen A: 1992. IDL, VLDL, chylomicrons and atherosclerosis. Em J Epidemiol 8(Suppl 1):92-98.
Have1 RJ: 1994. Postprandial and remnant lipoproteins. 01 5:102-109.
hyperlipidemia Curr Opin Lipid-
Have1 RJ, lipoprotein protein 1704. Have1 and CR,
TCM
Hamilton receptors catabolism.
RL: 1988. Hepatocytic and intracellular lipoHepatology 8:1689-
RJ, Kane JP: 1989. Part 7: lipoprotein lipid metabolism disorders. In Striver Beaudet AL, Sly WS, Valle D, eds. The
Vol. 4, No.
3, I994
Phillips ML, Schumaker VN: 1989. Conformation of apolipoprotein B after lipid extraction of low density lipoprotein attached to an electron microscopic grid. J Lipid Res 30:4 1 S-422. Plonne D, Winkler L, Franke H, Dargel R: 1992. The visceral yolk sac: an important site of synthesis and secretion of apolipoprotein B containing lipoproteins in the feto-placental unit of the rat. Biochim Biophys Acta 1127:174-185.
01994,
Else&r
Science
Inc.,
1050-l 738/94/$7.00
Scanu AM: 1991. ovascular risk Med 1:294-299.
Lipoprotein(a) factor. Trends
as a cardiCardiovasc
Schneeman BO, Kotite L, Todd KM, Have1 RJ: 1993. Relationships between the responses of triglyceride-rich lipoproteins in blood plasma containing apolipoproteins B-48 and B-100 to a fat containing meal in normolipidemic humans. Proc Nat1 Acad Sci USA 90:2069-2073. Spring DJ, Chen-Liu LW, Chatterton JE, Elovson J, Schumaker VN: 1992. Lipoprotein assembly: apolipoprotein B size determines lipoprotein core circumference. J Biol Chem 267:14,839-14,845. Thrift RN, Forte TM, Cahoon BE, Shore VG: 1986. Characterization of lipoproteins produced by the human liver cell line, HeP G2, under defined conditions. J Lipid Res 27:236250. Thrift RN, Drisko J, Dueland S, Trawick JD, Davis RA: 1992. Translocation of apolipoprotein B across the endoplasmic reticulum is blocked in a nonhepatic cell line. Proc Nat1 Acad Sci USA 89:9161-9165. Uemura K, Pisa Z: 1988. Trends in cardiovascular disease mortality in industrialized countries since 1950. World Health Stat Q 41:155-177. Wetterau JR, Zilversmit DB: 1985. Purification and characterization of microsomal triglyceride and cholesteryl ester transfer protein from bovine liver microsomes. Chem Phys Lipids 38:202-222. Wetterau JR, Combs KA, Spinner SN, Joiner BJ: 1990. Protein disulfide isomerase is a component of the microsomal triglyceride transfer protein complex. J Biol Chem 265:9800-9807. Wetterau JR, Combs KA, McLean LR, Spinner SN, Aggerbeck LP: 1991. Protein disulfide isomerase appears necessary to maintain the catalytically active structure of the microsomal triglyceride transfer protein. Biochemistry 30:9728-9735. Wetter-au JR, Aggerbeck LP, Bouma M-E, et al.: 1992. Absence of microsomal triglyceride transfer protein in individuals with abetalipoproteinemia. Science 258:9991001. Yamamoto T, Bishop RW, Brown MS, Goldstein JL, Russell DW: 1986. Deletion in cysteine-rich region of LDL receptor impedes transport to cell surface in WHHL rabbit. Science 232:1230-1237. Yeagle PL: 1990. Role of Cholesterol in Cellular Functions: Advances in Cholesterol Research. Philadelphia, Telford, pp 11 l132. Zilversmit prandial 485.
DB: 1979. phenomenon.
Atherogenesis: Circulation
a post60:473TCM
139
Robert L. Hamilton is at the Cardiovascular Research Institute, University of California, School of Medicine, San Francisco,
and the Department of Anatomy, CA 94143-0130, USA.
‘Abbreviations: abeta, abetalipoproteinemia; ACAT, acyl CoA cholesterol acyltransferase; apo B, apolipoprotein B; CE, cholesteryl ester; CURL, compartment of uncoupling of receptor and ligand; DGAT, diacylglycerol acyltransfemse; EM, electron microscopy; ER, endoplasmic reticulum; HDL, high-density lipoprotein: LDL, low-density lipoprotein: LDL-R, lowdensity lipoprotein receptor: mRNA, messenger ribonucleic acid: MTP, microsomal triglyceride transfer protein: MVB, multivesicular body; PDI, protein disulfide isomerase; RER, rough endoplasmic reticulum: SER. smooth endoplasmic reticulum: TG, triglyceride; and VLDL, very low density lipoprotein.
TCM
Vol. 4, No. 3, 1994
0 1994, Elsevier
Science
Inc.,
1050-I 738/94/$7.00
Living cells are comprised mostly of water and water-soluble substances, but lipids or nonpolar molecules are also vital for life. Phospholipids, for example, are essential for cellular membranes, without which there would be no subcellular organelles such as the nucleus, endoplasmic reticulum, Golgi, endosomes, and mitochondria. There would also be no plasma membranes controlling selective ingress and egress of substances between the outside world and the intracellular milieu. One vital function of the plasma membrane, in controlling amounts and kinds of molecules entering and exiting cells, depends largely upon the presence of another lipid, unesterified cholesterol. Plasma membrane unesterified cholesterol modulates the fluidity of the phospholipid environment in which the cell surface protein macromolecules exist. This sterol appears to regulate stearic functions
of
macromolecules
in
the
plasma
membrane such as receptors, enzymes, transporters, and ion pumps (Yeagle 1990). About 70%80% of cellular cholesterol is localized in the plasma membranes of many cell types. In those cells that have an extraordinary need for cholesterol, however, such as hepatocytes and steroidogenic cells, large amounts of cholesterol may be stored in a precursor pool for bile acid and steroid hormone biosynthesis, respectively. In complex organisms, storage and transport of this sterol occur largely as cholesteryl esters* (CBS) in which a single long-chain fatty acid is condensed to the 3-hydroxyl position of cholesterol. CE occurs in virtually all living tissues, usually in association with another oil, the triacylglycerols (TGs). TGs (or fats) provide most cellular ATP following hydrolysis by various lipases, releasing longchain fatty acids for p oxidation within mitochondria. In multicellular organisms, storage and transport of CEs and TGs are tightly regulated complex processes requiring the expression and modulation of many different proteins and nucleotides.
l
Origins
of Apolipoprotein
B
A unique protein essential for oil transport between cells and tissue fluids is apolipoprotein B (apo B), a very large and hydrophobic protein. Of the mammalian bodies’ countless different types of cells, only three known epithelia express apo-B mWA: (a) hepatocytes, (b) absorptive enterocytes, and (c) yolk-sac endoderm. A
131
CHYLOMICRON PATHWAY
CELL
HDL
1
CHYLOMICRON
a Figure 1. The plasma chylomicron transport pathway (a) and the plasma VL.DL and LDL transport pathways (b) depicted here are described in the text. Blue represents triacylglycerols, red cholesteryl esters, and yellow the LDL or B-lo&E receptors. FFA, free fatty acid; and MG, monoglyceride. From Have1 (1982). fourth is unidentified placental cells (Demmer et al. 1986). These three epithelial cells synthesize and secrete TG and CE molecules contained in a tiny oil droplet (core lipids) surrounded with a beltlike band of apo B (Phillips and Schumaker 1989). There are two molecular forms of apo B, but only one gene. A unique biologic process, called RNA editing, alters the transcript encoded in the apo-B message, arresting translation at a specific nucleotide that is changed to a stop codon. This results in the premature termination of apo-B mRNA translation and the formation of apo B-48 (that is, about one-half of the total apo-B mFWA message). In the absence of the editing event, the fulllength message is translated, giving rise to apo B-100 (Kane and Have1 1994, Garcia et al. 1992). Apo B-48 has an apparent molecular mass of 264 kD whereas apo B-100 is about 550 kD (Kane and Have1 1994). In most mammals, including hu-
132
mans, this editing to produce apo B-48 occurs largely in small intestinal absorptive enterocytes during formation of chylomicrons, the triglyceride-rich particles that transport dietary TG and cholesterol derived from the absorption of foodstuffs, into the lymph and hence the bloodstream (Have1 and Kane 1989; Garcia et al. 1992). Apo B-100 is secreted by hepatocytes in very low density lipoprotein (VLDL) particles as a mechanism of exporting excess long-chain fatty acids that the liver takes up from blood, or synthesizes. In many species studied, including humans, hepatocytic VLDLs contain only apo B-100, but in rats nascent hepatocytic VLDL fractions contain both apo B-48 and apo B-100 (Hamilton et al. 1991). Dogs, horses, mice, and probably other species secrete both apo Bs from liver, based upon apo-B mRNA editing (Greeve et al. 1993). The significance of this is unknown. The rat visceral yolk-sac endoderm secretes TG-
01994
Elsevier
Science
Inc., 1050-1738/94/$7.00
rich particles containing B-100, although most of these nascent particles are of low-density lipoprotein (LDL) size and density (Plonne et al. 1992). The significance of placental expression of apo-B mRNA is unknown.
l
Plasma Metabolism of APO-B Particles (Figure la and b)
In the adult organism, virtually all apo-Bcontaining particles circulating in blood plasma are derived from hepatocytes and absorptive enterocytes. Even in the fasting state, the small intestine continues to secrete apo B in triglyceride-rich particles that are fewer and smaller than the chylomicrons formed during a meal (Jones and Ockner 1971). The number and size of chylomicrons (that is, the mass of absorbed fats) are determined largely by the amount of fat calories in a given meal. The metabolism of apo-B-48 particles from the gut and the metabolism of B- 100 particles from the liver are very similar initially, but diverge subsequently (Have1 1989). Each triglyceride-
TCM
Vol. 4, No.
3, 1994
VLDL-LDL
PATHWAYS
r\ j
.,:;.:,.:.:: :::::; .:.::z;: .+ bwotein .,.‘... k Lipose li ..:::I
8-loo:E Receptor
Receptor
-
b rich particle synthesized and secreted contains a single copy of either apo B-48 or apo B-100, which remains with its particle throughout its lifespan. Following secretion of nascent chylomicrons and nascent VLDLs into the interstitial fluids or blood plasma, apo A-l and phospholipids are rapidly unloaded, presumably onto high-density lipoproteins (HDLs), in exchange for C-I, C-II, and C-III apolipoproteins and apolipoprotein E (apo E). These exchangeable apolipoproteins are present usually in multiple copies on each triglyceride-rich particle (Have1 1989, Have1 and Kane 1989). Both chylomicrons and VLDLs undergo hydrolysis of their core triglycerides during passage through capillary beds that contain lipoprotein lipase dangling from the endothelial cells, bound to heparan sulfate (Have1 1989, Have1 and Kane 1989). Most lipoprotein lipase is in heart, skeletal muscle and, in the fed state, in adipose tissue. Although 70% 90% of chylomicron and VLDL content triglycerides are lost during this lipoprotein lipase hydrolysis, the resultant partiTCM
Vol. 4, No. 3, 1994
cles are only modestly smaller in diameter because they are spheres. The surface apolipoproteins C-I, C-II, and C-III, together with phospholipids and perhaps unesterified cholesterol, transfer back to HDLs during this lipolytic cascade. The resultant apo-B-containing particles are now called remnants of triglyceride-rich particles, and normally have a very short half-life (minutes to hours) in plasma (Have1 1989, Have1 and Kane 1989).
l
Subcellular Catabolism APO-B Particles (Figure
of 2)
The vast majority of remnants of chylomicrons and most VLDL remnants are removed from the blood by the liver by the process of receptor-mediated endocytosis and are fully degraded in secondary lysosomes within hepatocytes (Have1 and Hamilton 1988). At this point in the catabolic cascade of TG-rich particles, the pathway of a fraction of VLDL remnants diverges from that of chylomicron remnants (Have1 1989, Have1 and Kane 1989). Depending on the species
01994,
Elsevier
Science
Inc.,
1050-l
73&J/94/$7.00
studied and the technique used to make the measurements, an estimated 5% 50% of VLDL remnants containing apo B-100 are converted to LDLs (Have1 1989, Have1 and Kane 1989). The molecular mechanisms of forming LDL from VLDL remnants are, however, obscure. Unlike short-lived remnants, LDLs are cholesteryl ester-enriched long-lived particles (-3 days in humans) that can deliver cholesterol to many cells of the body by the LDL receptor (LDL-R) pathway by receptor-mediated endocytosis (Brown and Goldstein 1986, Goldstein and Brown 1989, Have1 and Kane 1989). The LDL-R binds to a short sequence of apo B-100 (absent from apo B-48), and clusters LDL particles in coated pits of the plasma membrane before pinching off to form coated vesicles. Those coated vesicles rapidly uncoat clathrin and adaptin proteins and fuse with other early endosomes and/or a stable endosomal compartment near the cell surface (McKanna et al. 1979, Brown and Goldstein 1986, Have1 and Hamilton 1988, Klumperman et al. 1993). This
133
0
Sinusoid
.
.
sis, recycling of receptors, downregulation of cell surface receptors, and degradation of endosomal contents by acid hydrolases in secondary lysosomes is a common metabolic pathway for countless different ligands in virtually all cells (McKanna et al. 1979, Brown and Goldstein 1986, Jost-Vu et al. 1986, Have1 and Hamilton 1988, Klumperman et al. 1993). Whereas it is widely accepted that many different cells utilize this LDL-R pathway to obtain cholesterol, the bulk (60% 70%) of LDL turnover in plasma in vivo occurs by receptor-mediated endocytosis back in hepatocytes (Brown and Goldstein 1986, Goldstein and Brown 1989, Have1 and Kane 1989), the same cells that started this entire metabolic process by synthesizing and secreting nascent VLDLs into plasma in the space of Disse (Have1 and Kane 1989, Hamilton et al. 1991). Thus, hepatocytes determine to a very large extent the concentration of apo-B-containing lipoproteins in plasma (Have1 and Hamilton, 1988).
.
l
Figure 2. This diagram depicts our current concepts of receptor-mediated endocytosis and intracellular processing of triglyceride-rich remnants of chylomicrons and VLDLs in hepatecytes.
The same or closelysimilar receptor-mediated processesoccur during catabolism of cholesteryl ester-rich LDLs both in hepatocytes and in cellsof peripheral tissuesas described in the text. L’, primary lysosome; L*, secondary lysosome; and RRC, receptor recycling compartment. From Have1&d Hamilton (1988). early endosomal compartment has been dubbed CURL for compartment of uncoupling of receptor and Zigand. A proton pump in CURL lowers the pH to -6.0, causing the dissociation of apo B-100 from the LDL-R. Most LDL-R in CURL becomes sequestered into a membranous appendage that is enriched in receptors that recycle many times between early endosomes and the cell surface (RRC; see Figure 2). In addition, those membrane-bound receptors that are destined to be downregulated are carried with surface membrane pinching off into the lumen of CURL, forming small lipid bilayer vesicles. Because these vesicles become mixed with the contents (McKanna et al. 1979), they will be degraded by acid hydrolases in secondary lysosomes. Thus, CURL gives rise to nascent multivesicular bodies (MVBs) growing by endosomal fusions near the cell surface and at the same time sequestering receptors into appendages that pinch off and return long-lived recycling receptors
134
to the plasma
membranes
and
internalizing some receptors for downregulation (Have1 and Hamilton 1988). Upon reaching a limited size of about 0.4-0.5 i.un, MVBs become rapidly transported across the cytoplasm on microtubules to the cell center, specifically to the transGolgi network, where they fuse with coated blebs that deliver lysosomal enzymes into the MVB contents (Have1 and Hamilton 1988, Klumperman et al. 1993). Lysosomal enzymes synthesized in the rough endoplasmic reticulum (RER) are concentrated in coated blebs of the truns-Golgi network, bound to mannose6-phosphase receptors also concentrated in the coated blebs (Klumperman et al. 1993). Thus, MVBs represent the last endosomal (prelyosomal) compartment (Have1 and Hamilton 1988) that transforms into secondary lysomes adjacent to the Golgi apparatus (Jost-Vu et al. 1986). The mannosed-phosphate receptor and perhaps other macromolecules are recycled from MVBs back to the Golgi (Klumperman et al. 1993). This process of receptor-mediated endocyto01994,
Else&x
Science
Inc.,
1050-1738/94/$7.00
Apo B in Atherogenesis
It is now widely societies
consuming
(Figure
recognized large
3)
that in
amounts
of
cholesterol and saturated fats, particularly of porcine and bovine origins, high levels of apo-B-containing plasma lipoproteins represented largely by LDL cholesterol concentrations and possibly delayed clearance of remnants of triglyceride-rich particles (Have1 1994, Schneeman et al. 1993) are both leading causal factors of atherogenesis leading to heart attack, a major cause of death (Brown and Goldstein 1986, Goldstein and Brown 1989, Have1 and Kane 1989). Evidence is mounting steadily that remnants of triglyceride-rich particles can be atherogenic (Zilversmit 1979, Have1 1990, Nordestgaard and ‘Qbjaerg-Hansen 1992, Have1 1994). We used to say that this debilitating disease of atherosclerosis was confined largely to Western and/or affluent societies, such as the USA and Western Europe, which could afford excessive fat calories. Now, in Eastern Europe, many of those societies are experiencing a substantial increased incidence of heart attack (Figure 3), more than twice that of our Western societies, in which the rate has been declining gradually for about a dozen years (Uemura and Pisa 1988, Marmot 1992). TCM
Vol. 4, No. 3, 1994
Finland England and Wales
Czech
Ireland
Hungary
Belgium
France
FP Germany
Poland
1
I MA” I -40
-30 Change
-20 in mortality
How can this be explained? Partly diet and partly life-style changes in opposite directions may account for much of this increased heart disease. Whereas greatly reduced cigarette smoking, prudent dietary habits, and fitness awareness are contributing to decreased incidence of heart attack in affluent Western societies, just the opposite trends are evident in Eastern European countries. To feed the populace, the cheapest sources of calories predominate in the marketplace: pork in many different forms (Miller 1993). At the same time, cigarettes now are cheap and accessible, and the vast majority of people smoke them. The death and morbidity rates, particularly among men in their most productive years of life, are leaving a devastating impact on the peoples of Eastern European countries (Uemura and Pisa 1988, Marmot 1992, Miller 1993).
l
Too Much Apo B
or Too Little
Plasma
Two contrasting and rare genetic diseases that occur in humans are caused by defects in one of two of the major macromolecular players of plasma lipoprotein oil transport. The first and best understood is familial hypercholesterolemia, in which the LDL-R is defective, plasma cholesterol levels are grossly elevated (-1000 mg/dL), severe atherosclerosis occurs at a very early age, and heart attacks often occur in childhood (Brown and Goldstein 1986, Goldstein and Brown 1989, Have1 and Kane 1989). A closely similar disorder occurs in the Watanabe heritable hyperlipidemic (WHHL) rabbit, because of a single deletion of 12 nucleotides in the LDL-R mRNA (Brown and Goldstein 1986, TCM
Vol. 4, No. 3, I994
0
-10 1%)
fbmania}
10
-20
0
20 Change
40 in mortality (%I
60
Bl
Figure 3. This figure contrasts changes in coronary heart disease mortality (%) that is gradually decreasing in Western European countries (as in the USA), whereas it is increasing substantially in men in Eastern European societies. Men and women, ages 30-69, between 1970 and 1985. From Marmot (1992), with permission from Oxford University Press.
Yamamoto et al. 1986, Goldstein and Brown 1989). Homozygous WHHL rabbits rapidly develop severe atherosclerosis caused by high plasma levels of LDLs and remnants, apo-B-containing particles. Thus, although in humans many other factors such as cigarette smoking, hypertension, diabetes mellitus, and physical inactivity may contribute to atherosclerosis of the coronary and cerebral arteries, leading to heart attack and stroke, excessive levels of LDL cholesterol and/or remnants can by themselves cause this disease process (Goldstein and Brown 1989, Have1 and Kane 1989, Nordestgaard and Tybjaerg-Hansen 1992, Have1 1994). A second rare and debilitating disease, called abetalipoproteinemia (abeta), is caused by a defect in the assembly of apo-B-containing triglyceride-rich particles in which no chylomicrons, VLDLs, or LDLs are present in plasma (Kane and Have1 1994, Wetter-au et al. 1992). Cholesterol and TG levels are low and are found only in HDL fractions (Kane and Have1 1994). Organ abnormalities abound, such as spinocerebellar degeneration and retinitis pigmentosa, and death often occurs early in affected individuals. For many years, it was believed that the defect resulted from the genetic absence of apo B, because there was virtually none detectable in plasma lipoprotein fractions. With the development of Western and Northern blotting techniques, it was found that intestinal biopsy material of abeta individuals contains apo B-48 of
01994.
Elsevier
Science
Inc.,
1050.1738/94/$7.00
correct molecular weight, and apo-B mRNA levels severalfold higher than in normal human intestines (Kane and Have1 1994). What then is the defect in abeta? Wetterau et al. (1992) have discovered that biopsy material of the small intestine from four abeta subjects appears to lack completely a protein called microsomal triglyceride transfer protein, abbreviated MTP. Wetterau and various collaborators discovered and characterized this novel protein isolated from bovine liver microsomal membranes (Wetterau and Zilversmit 1985). MTP is a heterodimer of protein disulfide isomerase (PDI) (-55 kD) and another protein of -88 kD (Wetterau et al. 1990 and 1992). In vitro, this complex transfers phospholipids between liposomal bilayers, similar to several other phospholipid transfer proteins; it is unique, however, because it moves the oils, TGs, and CEs (Wetterau and Zilversmit 1985). The terminal enzymes of triacylglycerol synthesis [diacylglycerol acyltransferase (DGAT)] and cholestetyl ester synthesis Cacyl CoA cholesterol acyltransferase (ACAT)] are membrane-bound enzymes with catalytic sites on the cytosolic face of the endoplasmic reticulum (ER). Thus, their product oils must be transported across the ER membrane to reach the ER lumen, from which secreted proteins are transported to the Golgi before reaching the cell surface. MTP is an appealing candidate to participate in nascent VLDL and nascent chylomicron assembly as a
135
mediator of transfer of oils from their sites of synthesis to the site of apo-B core lipidation. PDI is a luminal enzyme of the RER that facilitates proper folding of many nascent polypeptides as they enter the RER lumen from polyribosomes, realigning sulfhydryl groups to form specific disulfide bridges that determine proper protein folding (Yamamoto et al. 1986, Wetterau et al. 1990 and 1992). Perhaps as MTP transfers TG and CE molecules to apo B, the PDI component coordinately realigns the cysteine residues of apo B; there are 25 in apo B-100, 22 of which are in intrachain disulfide bridges (Kane and Have1 1990). Interestingly, the 88-kD component of MTP aggregates and becomes inactive when dissociated from PDI, suggesting that it is a highly hydrophobic protein maintained soluble in the aqueous environment of the RER by PDI (Wetterau et al. 1991). The discovery by Wetterau et al. (1992) may be a major breakthrough for two different reasons. First, profound insights may be revealed for understanding some of the mechanisms by which triglyceride-rich particle assembly occurs. Second, their observations on the intestinal biopsy material of abeta subjects probably reveals the basis for the metabolic defect in this genetic disorder. The Wetterau et al. (1992) paper contains two basic findings. First, they showed that soluble proteins from intestinal biopsy material from four abeta subjects were virtually unable to transfer radiolabeled TGs from donor to acceptor phospholipid liposomes in the test tube. This functional assay for MTP was linear for 60 min when soluble proteins from normal intestinal biopsy material were substituted in the same assay (Wetterau et al. 1992). Second, immunoblots with antisera against the bovine liver MTP identified the 88-kD subunit of correct molecular weight in soluble material from not only normal human intestinal biopsies, but also in the same material from subjects with two other human disorders of chylomicron formation, hypobetalipoproteinemia and Anderson’s disease. No protein of this molecular weight could be detected in immunoblots from intestinal material from four abeta subjects, even when 4-5 times more soluble proteins were applied to the SDS-PAGE, although the PDI monomer was detected (Wetterau et al. 1992).
136
These elegant studies strongly support the conclusion that the genetic defect in abeta is the apparent absence of correctly folded 88-kD protein subunits of MTP. It is probable that the biogenesis of genetically defective 88-kD subunits during their translation may cause improper folding of this subunit of the MTP heterodimer, which may prevent association with correctly formed PDI. In other words, the genetic defect is probably not the absence of the 88-kD subunits, but rather their inability to undergo the correct folding to form the functional heterodimer.
l
The “Two-Step” Hypothesis of APO-B Particle Assembly (Figure 4a and b)
A new and intellectually stimulating hypothesis is emerging to describe molecular events that may occur during the assembly of triglyceride-rich particles in the endoplasmic reticulum (ER). The new hypothesis is termed the two-step model of apo-B core lipidation in the ER. The hypothesis evolved from two widely different observations. The first observation is that immunolabeling at the electron-microscopic (EM) level showed heavy staining of rat hepatocytic RER for apo B in the absence of an apparent VLDL-sized particle, whereas VLDLsized particles in the smooth ER (SER) lacked apo-B immunostaining (Alexander et al. 1976). This finding is consistent with VLDL assembly occurring by two distinct steps, although we did not use those words. Second, HepG2 cells secrete most apo B in particles of HDL and LDL size and density, rather than VLDLs, and these transformed cells contain little or no detectable SER compartment by EM examination (Thrift et al. 1986). Hence, HepG2 cells secrete apo-Bcontaining particles formed by the first step in the RER because they lack the second step that makes triglyceride-rich particles that may require the SER. In studying the relationship between apo-B length and particle size (that is, core volume) in isolated lipoprotein particles from HepG2 cells, Spring and colleagues (1992) concluded that small apo-Bcontaining nascent lipoproteins must acquire additional core lipid in normal hepatocytes to form a nascent VLDL particle provided by a “second stage” of assembly.
01994,
Elsevier
Science
Inc.,
1050-1738/94/$7.00
The concept of a two-step VLDL assembly process suggests that there may be circumstances in nature in which the two stages may be dissociated. This appears to be a plausible explanation for our earlier observations of apo-B secretion in isolated perfused rat livers. When rat livers are perfused in a recirculating system for prolonged periods (5-6 h) in the absence of exogenous FFA as substrate to drive fi oxidation and VLDL secretion, a significant amount of apo B is recovered in particles in the HDL density fraction (Fainaru et al. 1977). Furthermore, when the same techniques of perfusion are carried out with livers of cholestatic rats (bile duct ligation for 24-48 h), twice as much apo B is recovered in the HDL fraction, and most of the rest of the per&sate apo B is in particles of LDL density at the expense of apo-B mass in nascent VLDL (Felker et al. 1982). Cholestatic rat liver lobes produced by bile duct obstruction have only one-tenth as much TG as normal rat liver lobes (Felker et al. 1982). This suggests that in the absence of sufficient fatty acid substrate, the larger triglyceriderich particles that are normally formed in the SER must be much reduced in number in cholestasis, and particles of LDL and HDL size that are formed in the RER in the first step are secreted. More direct biochemical support for the twostep hypothesis has been obtained recently by Elovson et al. (1992) and by the author (Hamilton et al. unpublished work). By developing a novel procedure to isolate large amounts (- 100 mg protein/ 10 g liver) of structurally intact RER membranes very rapidly from rat livers, we have been able to recover nascent precursors of VLDLs. Two groups of particles are highly enriched in apo B, are denser than VLDLs (in the HDL-toLDL density range), are smaller than VLDLs (125-250 A diameter), and contain small amounts of TGs and CEs. These two particles band at different densities in sucrose gradients, depending on the apo-B-TG ratio. The second particle is larger, the same size as nascent Golgi VLDLs (Hamilton et al. 1991) (-450 A diameter), and contains the same TG and CE mass, but completely lacks apo B by Western immunoblotting. This particle does, however, contain proteins that are unidentified. These particles appear to represent the ones that were predicted to exist in the SER TCM
Vol. 4, No. 3, 1994
and SER-RER junctions by immunostaining for apo B, which was absent from SER VLDL-sized particles (Alexander et al. 1976). Speculations Mechanisms
l
on Molecular of the Two-Step
SER LIpIds 1
I
Model
The following speculations are an attempt to bring together the two-step model and the observations by Wetterau et al. (1992) on the role(s) of MTP in the RER in core lipidation of apo B in triglyceride-rich particle assembly. Apo B is constitutively expressed by its mRNA in both adult liver and small intestine (Have1 and Kane 1989), and newly translated full-length apo B first becomes firmly associated with the RER membrane before it is found in an isolatable water-soluble lipoprotein particle in the ER lumen (Bijstrom et al. 1986, Thrift et al. 1992, Dixon and Ginsberg 1993). If sufficient substrate lipids are present in the cell, most of this RER-bound apo B becomes core lipidated, enabling its dissociation from the RER membrane as a small lipoprotein particle formed in the first step. When core lipid substrates (TGs and CEs) are not present in sufficient amounts to saturate this process, RER membranebound apo B undergoes local proteolytic degradation, a postulated regulatory process in apo-B secretion (Thrift et al. 1992, Dixon and Ginsberg 1993). The first step is predicted to be an MTP-dependent event whereby sufficient oils (TGs and CEs) are transferred from their cytosolic
Nascent VLDL
n
Apparatus
(
)
^^ n
Figure 4. The morphologic basis (a) and the biochemical basis (b) for the two-step hypothesis of nascent VLDL assembly are depicted. (a) Based on the absence of immunoperoxidase staining for apo B on VLDL-sized particles in the SER (solid black circles), whereas the tiny x’s represent locations of positive staining for apo B. (b) Based on characterization of small apo-B-rich particles and apo-Bdeficient particles of VLDL size isolated from the contents of a novel ribosomal enriched fraction from rat liver. The first step (I) depicts the initial core lipidation of apo B in the rough endoplasmic reticulum membrane, releasing apo B in a small lipoprotein particle into the lumen. The second step (2) occurs when a small apo-B-rich particle coalesces with an apo-B-deficient triglyceride-rich particle formed in the smooth endoplasmic reticulum completing nascent VLDL apo-B core lipidation. (a) From Alexander et al. (1976), with permission from Rockefeller University Press.
TCM
Vol. 4, No. 3, 1994
01994.
Elsevier
Science
Inc.,
1050-l 738/94/$7.00
137
sites of synthesis into hydrophobic domains of apo B residing within the RER membranes, perhaps by virtue of its binding to the RER phospholipids and/or to special apo-B-binding RER proteins. If the hypothesis is correct that a portion of RER-bound apo B is cytosolic (Thrift et al 1992) this proximity to the active sites of DGAT and ACAT (the enzymes that form TGs and CEs) may facilitate the initial core lipidation of apo B. When the MTPdependent process presents sufficient molecules of TGs and CEs that are more hydrophobic than the RER membranebinding sites, apo B may undergo structural reorganization to enable it to dissociate from its RER membrane-binding domains. When a critical mass of TGs and CEs has been transferred by MTP to interact with a sufficient number of hydrophobic domains of apo B, nucleation may produce a small core lipid-containing particle that is released from the RER membrane into the RER lumen as the completion of the first step (Figure 4b). This interpretation is consistent with recent molecular biologic experiments in which different apo-B constructs were inserted into both nonhepatic cells and hepatoma cells. Transfection of the Nterminal 53% of apo B into hepatoma cells led to the secretion of apo B-53 in lipoprotein particles. When the apo B-53 construct was transfected into Chinese hamster ovary cells (which lack MTP), no detectable apo B was secreted, because it could not be dissociated from RER membranes, where it was rapidly degraded by a protease (Thrift et al. 1992). Surprisingly, when an apo-B-15 construct was used in the same type of experiments, both cell types secreted apo B-15, but apparently in a lipid-free or lipid-poor form (Thrift et al. 1992). These results suggest the simple idea that apo B-15 is too short to become incorporated into RER membranes and therefore is translocated into the RER lumen like any soluble secretory protein, bypassing core lipidation events in those few cell types that express MTP. This predicts that some minimal length of apo B, about B-30 in humans (Linton et al. 1993), may be necessary to become firmly bound to the RER membrane and, as a consequence, to be core-lipidated in cells expressing MTP. If substrates are not adequate to supply sufficient amounts of oils to RER-bound apo B by those cells expressing MTP, or if cells do not express
138
MTP, RER-bound apo B undergoes proteolytic degradation. The two-step hypothesis of VLDL assembly may also explain some observations recently described in humans with a genetic disease designated hypobetalipoproteinemia. This disorder is characterized by truncated apo Bs of different lengths that are often found in HDL and LDL particles and associated with low plasma levels of apo B and cholesterol (Linton et al. 1993). The most severe phenotypes of this disorder may have clinical signs that are indistinguishable from abeta subjects requiring treatment, particularly dietary supplementation of lipid-soluble vitamins (Linton et al. 1993). One possible explanation for these truncated apo Bs in HDLs and LDLs is that some truncated forms of apo B may be long enough (> apo B-30) to form the small particles in the RER in the first step of VLDL assembly. Hence, some of these particles may be secreted directly because they may not be efficient in coalescing with apo-B deficient TG-rich particles of the SER. In abeta, the genetic absence of MTP would curtail the release of apo B from the RER membrane, completely preventing triglyceride-rich particle formation. The VLDL-sized particles lacking apo B that are formed independently of apo B in the SER may also require MTP to form the microemulsion particle, but this process would probably also require other unknown proteins and events to occur. Even if apo-B-deficient triglyceriderich particles were to be formed in the SER of hepatocytes or enterocytes, they most likely would not be released without apo B, the presumed protein signal enabling microemulsion oil droplets to transit the secretory pathway. The twostep hypothesis identifies two important areas for future research aimed at discovering mechanisms of triglyceriderich particle assembly by hepatocytes and absorptive enterocytes. The first is to unravel the molecular mechanisms by which MTP core lipidates apo B in the RER (note that no model is presented for this in Figure 4b). The second is to identify those special proteins and processes that form the microemulsion particles in the SER compartment in the absence of apo B. The fact that small apo-B-rich particles and larger apo-Bdeficient particles coexist in the ER lumen indicates that, to form a nascent
01994,
Elsevier science Inc.,
1050-1738/94/$7.00
VLDL particle, coalescence between these two particles must occur, presumably at the RER-SER junction. The two-step hypothesis of apo-B core lipidation and the discovery of the essential role of MTP in triglyceride-rich particle assembly have enormous potential clinical value. Together, these new insights may soon enable understanding of the molecular processes and therefore the regulation of VLDL and chylomicron secretion by hepatocytes and enterocytes, respectively. Once these processes are understood, pharmaceutical intervention, in addition to prudent diets, may reduce the first and/or the second steps such that apo-B-containing plasma lipoproteins are maintained at lower levels that greatly reduce the incidence of atherosclerotic disease, particularly heart attack and possibly stroke. For example, elevated concentrations of another class of apo-B-containing particles in human plasma, called lipoprotein(a), is another risk factor for atherosclerotic heart disease (Scanu 1991), but it may be even a greater risk factor for stroke (Nagayama et al. 1994, Jiirgens et al. 1994).
l
Acknowledgments
The author’s research is supported by NHLBI Arteriosclerosis Specialized Center of Research HL-14237. He is a member of the senior staff of the Cardiovascular Research Institute and Professor of Anatomy at the University of California, School of Medicine, San Francisco, Cali-. fomia.
References Alexander CA, Hamilton RL, Have1 RJ: 1976. Subcellular localization of B apoprotein of plasma lipoproteins in rat liver. J Cell Biol 69:241-263. Biistrom K, Wettesten M, Boren J, et al.: 1986. Pulse-chase studies of the synthesis and intracellular transport of apolipoprotein B 100 (apo B 100) in Hep G2 cells. J Biol Chem 261:13,800-13,806. Brown MS, Goldstein JL: 1986. A receptormediated pathway for cholesterol homeostasis. Science 232:34-47. Demmer LA, Levin MS, Elovson J, et al.: 1986. Tissue-specific expression and developmental regulation of the rat apolipoprotein B gene. Proc Nat1 Acad Sci USA 83:8102-8106. Dixon JL, Ginsberg HN: 1993. Regulation of
TCM Vol. 4, No. 3, 1994
hepatic secretion of apolipoprotein containing lipoproteins: information tained from cultured liver cells. J Lipid 34167-179.
BobRes
Elovson J, Bell GT, Doolittle MH, Phillips M: 1992. A two-step model for very low density lipoprotein assembly: characterization of intermediate lipoprotein particles [abst 27531. Circulation 86(Suppl l):PI-692. Fainaru M, Felker TE, Hamilton RL, Have1 RJ: 1977. Evidence that a separate particle containing B-apoprotein is present in highdensity lipoproteins from perfused rat liver. Metabolism 26:999-1004. Felker TE, Hamilton RL, Vigne J-L, Have1 RJ: 1982. Properties of lipoproteins in blood plasma and liver perfusates of rats with cholestasis. Gastroenterology 83:652-663. Garcia ZC, Poksay KS, Bostrom K, et al.: 1992. Characterization of apolipoprotein B mRNA editing from rabbit intestine. Arterioscler Thromb 12:172-179. Goldstein JL, Brown MS: 1989. Part 7: lipoprotein and lipid metabolism disorders. In Striver CR, Beaudet AL, Sly WS, Valle D, eds. The Metabolic Basis of Inherited Disease. New York, McGraw-Hill, pp 121% 1250. Greeve J, Altkemper I, Dieterich J-H, Greten H, Windler E: 1993. Apolipoprotein B mRNA editing in 12 different mammalian species: hepatic expression is reflected in low concentrations of apo B-containing plasma lipoproteins. J Lipid Res 34:1367-1383. Hamilton RL, Moorehouse A, Isolation and properties of teins from highly purified Golgi fractions. J Lipid Res
Have1 RJ: 199 1. nascent lipoprorat hepatocytic 32:529-543.
Have1 RJ: 1982. Approach to the patient with hyperlipidemia: symposium on lipid disorders. Med Clin North Am 66:319-321.
Metabolic Basis York, McGraw-Hill,
of Inherited pp 1129-l
Disease. 164.
New
Jones AL, Ockner RK: 1971. An electron microscopic study of endogenous very low density lipoprotein production in the intestine of rat and man. J Lipid Res 12:580-589. Jost-Vu E, Hamilton RL, Homick CA, Belcher JD, Have1 RJ: 1986. Multivesicular bodies isolated from rat hepatocytes: cytochemical evidence for transformation of multivesicular bodies into secondary lysosomes by fusion with primary lysosomes. Histochemistry 85457-466. Jtirgens G, Taddei-Peters WC, Koltringer P, et al.: 1994. Lipoprotein(a) concentration is a better predictor for risk of cerebrovascular disease than carotid artery stenosis or other lipoprotein levels [abst]. Stroke 25:247. Kane JP, Have1 RJ: 1994. Disorders of the biogenesis and secretion of lipoproteins containing the B apo-lipoproteins. In Striver CR, Beaudet AL, Sly WS, Valle D, eds. The Metabolic Basis of Inherited Disease. New York, McGraw-Hill (in press). Klumperman J, Hille A, Veenendaal T, et al.: 1993. Differences in the endosomal distributions of the two mannose-6-phosphate receptors. J Cell Biol 121:997-1010. Linton MF, Farese RV, Young Familial hypobetalipoproteinemia. Res 34:521-541.
SG:
1993. J Lipid
Marmot M: 1992. Coronary heart disease: rise and fall of a modem epidemic. In Marmot M, Elliott P, eds. Coronary Heart Disease Epidemiology: From Aetiology to Public Health. Oxford, Oxford University Press, pp 3-19. McKanna JA, Haigler HT, Cohen SC: 1979. Hormone receptor topology and dynamics: morphological analysis using ferritinlabeled epidermal growth factor. Proc Nat1 Acad Sci USA 7615689-5693. Miller J: 1993. A change Republic. UCSF Mag
of heart 14:23-25.
for the Czech
Have1 RJ: 1989. Contrasts and similarities between the metabolism of intestinal and hepatic lipoproteins: the role of particle size. In Windler E, Greten H, eds. Intestinal Lipid and Lipoprotein Metabolism. Munich, W Zuckschwerdt, pp 168-173.
Nagayama M, Shinohara Y, Nagayama T: 1994. Lipoprotein(a) and ischemic cerebrovascular disease in young adults. Stroke 25174-78.
Have1 RJ: 1990. Role of lipoproteins in progression sis. Circulation 8 1:694-696.
triglyceride-rich of atherosclero-
Nordestgaard BG, Tybjaerg-Hansen A: 1992. IDL, VLDL, chylomicrons and atherosclerosis. Em J Epidemiol 8(Suppl 1):92-98.
Have1 RJ: 1994. Postprandial and remnant lipoproteins. 01 5:102-109.
hyperlipidemia Curr Opin Lipid-
Have1 RJ, lipoprotein protein 1704. Have1 and CR,
TCM
Hamilton receptors catabolism.
RL: 1988. Hepatocytic and intracellular lipoHepatology 8:1689-
RJ, Kane JP: 1989. Part 7: lipoprotein lipid metabolism disorders. In Striver Beaudet AL, Sly WS, Valle D, eds. The
Vol. 4, No.
3, I994
Phillips ML, Schumaker VN: 1989. Conformation of apolipoprotein B after lipid extraction of low density lipoprotein attached to an electron microscopic grid. J Lipid Res 30:4 1 S-422. Plonne D, Winkler L, Franke H, Dargel R: 1992. The visceral yolk sac: an important site of synthesis and secretion of apolipoprotein B containing lipoproteins in the feto-placental unit of the rat. Biochim Biophys Acta 1127:174-185.
01994,
Else&r
Science
Inc.,
1050-l 738/94/$7.00
Scanu AM: 1991. ovascular risk Med 1:294-299.
Lipoprotein(a) factor. Trends
as a cardiCardiovasc
Schneeman BO, Kotite L, Todd KM, Have1 RJ: 1993. Relationships between the responses of triglyceride-rich lipoproteins in blood plasma containing apolipoproteins B-48 and B-100 to a fat containing meal in normolipidemic humans. Proc Nat1 Acad Sci USA 90:2069-2073. Spring DJ, Chen-Liu LW, Chatterton JE, Elovson J, Schumaker VN: 1992. Lipoprotein assembly: apolipoprotein B size determines lipoprotein core circumference. J Biol Chem 267:14,839-14,845. Thrift RN, Forte TM, Cahoon BE, Shore VG: 1986. Characterization of lipoproteins produced by the human liver cell line, HeP G2, under defined conditions. J Lipid Res 27:236250. Thrift RN, Drisko J, Dueland S, Trawick JD, Davis RA: 1992. Translocation of apolipoprotein B across the endoplasmic reticulum is blocked in a nonhepatic cell line. Proc Nat1 Acad Sci USA 89:9161-9165. Uemura K, Pisa Z: 1988. Trends in cardiovascular disease mortality in industrialized countries since 1950. World Health Stat Q 41:155-177. Wetterau JR, Zilversmit DB: 1985. Purification and characterization of microsomal triglyceride and cholesteryl ester transfer protein from bovine liver microsomes. Chem Phys Lipids 38:202-222. Wetterau JR, Combs KA, Spinner SN, Joiner BJ: 1990. Protein disulfide isomerase is a component of the microsomal triglyceride transfer protein complex. J Biol Chem 265:9800-9807. Wetterau JR, Combs KA, McLean LR, Spinner SN, Aggerbeck LP: 1991. Protein disulfide isomerase appears necessary to maintain the catalytically active structure of the microsomal triglyceride transfer protein. Biochemistry 30:9728-9735. Wetter-au JR, Aggerbeck LP, Bouma M-E, et al.: 1992. Absence of microsomal triglyceride transfer protein in individuals with abetalipoproteinemia. Science 258:9991001. Yamamoto T, Bishop RW, Brown MS, Goldstein JL, Russell DW: 1986. Deletion in cysteine-rich region of LDL receptor impedes transport to cell surface in WHHL rabbit. Science 232:1230-1237. Yeagle PL: 1990. Role of Cholesterol in Cellular Functions: Advances in Cholesterol Research. Philadelphia, Telford, pp 11 l132. Zilversmit prandial 485.
DB: 1979. phenomenon.
Atherogenesis: Circulation
a post60:473TCM
139