LIPID SYNTHESIS, INTRACELLULAR TRANSPORT, AND STORAGE : III. Electron Microscopic Radioautographic Study of the Rat Heart Perfused with Tritiated Oleic Acid.

Rat hearts pulse-labeled by perfusion in vitro with 9,10-oleic acid-(3)H for 15 or 30 sec were shown to take up the fatty acid extensively. In hearts postperfused with unlabeled medium for 15 sec or more, 90% of the radioactivity was recovered in esterified lipids. The radioautographic reaction was localized initially over elements of the sarcoplasmic reticulum and mitochondria. After longer periods of postperfusion (2-20 min), there was concentration of silver grains over lipid droplets. In mitochondria and sarcoplasmic reticulum isolated from hearts postperfused for 1 min or more, most of the esterified lipid was in the form of triglyceride. The ratio of the specific activity of isolated sarcoplasmic reticulum triglyceride to mitochondrial triglyceride changed from a value of 3.2 to 1.3 during 5 min of postperfusion. Under conditions of hypothermia, considerable uptake of free fatty acid occurred. The radioactivity recovered in the heart was mostly in the form of free fatty acid, and the radioautographic reaction was seen over sarcoplasmic reticulum and mitochondria, but not over lipid droplets or myofibrils. The results are interpreted to show that intracellular transport of free fatty acid, which occurs also when esterification is repressed, proceeds through intracellular channels, i.e. the sarcoplasmic reticulum. Esterification of fatty acid into triglycerides occurs mostly in the sarcoplasmic reticulum, especially in the region of the dyad, in the vicinity of which lipid is stored in the form of droplets.

Lecithin synthesis, intracellular transport, and secretion in rat liver. IV. A radioautographic and biochemical study of choline-deficient rats injected with choline-3H.

Injection of choline-(3)H into choline-deficient rats resulted in an enhanced incorporation of the label into liver lecithin, as compared to the incorporation of label into liver lecithin of normal rats. The results obtained with the use of different lecithin precursors indicate that in the intact liver cell, both in vivo and in vitro, exchange of choline with phosphatidyl-choline is not significant. The synthesis and secretion of lecithins by the choline-deficient liver compare favorably with the liver of choline-supplemented rats, when both are presented with labeled choline or lysolecithin as lecithin precursors. Radioautography of the choline-deficient liver shows that 5 min after injection of choline-(3)H the newly synthesized lecithin is found in the endoplasmic reticulum (62%), mitochondria (13%), and at the "cell boundary" (20%). The ratio of the specific activity of microsomal and mitochondrial lecithin, labeled with choline, glycerol, or linoleate, was 1.53 at 5 min after injection, but the ratio of the specific activity of phosphatidyl ethanolamine (PE), labeled with ethanolamine, was 5.3. These results indicate that lecithin and PE are synthesized mainly in the endoplasmic reticulum, and are transferred into mitochondria at different rates. The site of a precursor pool of bile lecithin was studied in the intact rat and in the perfused liver. Following labeling with choline-(3)H, microsomal lecithin isolated from perfused liver had a specific activity lower than that of bile lecithin, but the specific activity of microsomal linoleyl lecithin was comparable to that of bile lecithin between 30 and 90 min of perfusion. It is proposed that the site of the bile lecithin pool is located in the endoplasmic reticulum and that the pool consists mostly of linoleyl lecithin.

Lipid synthesis, intracellular transport, storage, and secretion. I. Electron microscopic radioautographic study of liver after injection of tritiated palmitate or glycerol in fasted and ethanol-treated rats.

A time sequence study of intracellular movement of labeled lipid in the liver was carried out on fasted and ethanol-treated rats injected with either palmitate-(3)H or glycerol-(3)H by electron microscopic radioautography. The elimination of water-soluble lipid precursors during specimen preparation was checked and found to be complete. The labeled lipid product in the tissue was identified as mostly triglyceride. A dehydration procedure was adapted to minimize the loss of lipid during specimen preparation. At 2 min after injection, the earliest time interval studied, both precursors were found to have penetrated the liver cells, and the label was found over both rough and smooth elements of the endoplasmic reticulum, which is the site of glyceride esterification. From 5 min on, in fasted and especially in ethanol-treated rats, the label was seen also over lipid droplets 0.5-2.0 micro in diameter, which represent "storage lipid" (slowly turning over compartment). Mitochondria became labeled mostly at later time intervals after injection. From 10 min on, concentration of label was seen over the Golgi apparatus, containing small osmiophilic particles. Association of label with groups of particles in smooth-surfaced vesicles and vacuoles in and near the Golgi apparatus and in the vicinity of the sinusoidal border was seen, both after palmitate-(3)H and glycerol-(3)H. It is proposed that these particles represent lipoproteins which are formed in the endoplasmic reticulum, "processed" in the Golgi apparatus, and transported in vacuoles to the sinusoid surface to be discharged into the circulation.

Lipid synthesis, intracellular transport, and secretion. II. Electron microscopic radioautographic study of the mouse lactating mammary gland.

In the mammary glands of lactating albino mice injected intravenously with 9, 10-oleic acid-(3)H or 9, 10-palmitic acid-(3)H, it has been shown that the labeled fatty acids are incorporated into mammary gland glycerides. The labeled lipid in the mammary gland 1 min after injection was in esterified form (> 95%), and the radioautographic reaction was seen over the rough endoplasmic reticulum and over lipid droplets, both intracellular and intraluminal. At 10-60 min after injection, the silver grains were concentrated predominantly over lipid droplets. There was no concentration of radioactivity over the granules in the Golgi apparatus, at any time interval studied. These findings were interpreted to indicate that after esterification of the fatty acid into glycerides in the rough endoplasmic reticulum an in situ aggregation of lipid occurs, with acquisition of droplet form. The release of the lipid into the lumen proceeds directly and not through the Golgi apparatus, in contradistinction to the mode of secretion of casein in the mammary gland or of lipoprotein in the liver. The presence of strands of endoplasmic reticulum attached to intraluminal lipid droplets provides a structural counterpart to the milk microsomes described in ruminant milk.

Colchicine-induced inhibition of lipoprotein and protein secretion into the serum and lack of interference with secretion of biliary phospholipids and cholesterol by rat liver in vivo.

Rats were injected with colchicine and the secretion of triglycerides into the serum was studied for 90 min after injection of [(14)C]palmitic acid and Triton WR 1339. The release of labeled and chemically determined triglyceride was reduced to about 20-30% of control values. The effect of colchicine on serum triglyceride levels was not dependent on the presence of Triton and was similar in males and females and in fed and fasted rats. The effect was dose dependent and was reversible 6-7 h after injection of 0.05 mg/100 g body weight. Colchicine inhibited also the release of labeled proteins into the serum but did not affect the amount of [(3)H]leucine incorporated into liver proteins. Within 4 h of colchicine treatment there was an 80% fall in serum very low density lipoproteins (VLDL), a 30% fall in serum high density lipoproteins (HDL), and no change in the d > 1.21 protein level, but reduction in the appearance of labeled proteins was encountered in all serum fractions. Colchicine had no effect on the rate of bile flow and on the secretion of phospholipids and cholesterol into the bile. In the hepatocyte there was accumulation of Golgi-derived secretory vesicles, containing nascent VLDL particles; these vesicles were seen also in the vicinity of the sinusoidal cell surface, but the space of Disse contained few or no VLDL particles. There was an apparent reduction in microtubules and some increase in microfilaments. It is suggested that microtubules affect the secretion of lipoproteins and proteins into the serum by maintaining the organization of the plasma membrane required for its fusion with secretory vesicles. The lack of effect of colchicine on biliary lipid secretion indicates that the latter is not dependent on vesicular transport.

Acylation of lysophosphatides by plasma membrane fractions of rat liver.

The plasma membrane fraction of rat liver was isolated and incubated with labeled lysophosphatides in the presence of cofactors; the acylation of lysolecithin to lecithin by the fraction was compared to that of the rough and smooth microsomes. The purity of the isolated fractions was ascertained by enzyme markers and electron microscopy, and the maximal contamination of the plasma membrane fraction by microsomes did not exceed 20%. Under conditions at which the reaction was proportional to the amount of enzyme used, the plasma membrane had a specific activity similar to that of the smooth and rough microsomes. With doubly labeled lysolecithin (containing palmitic acid-(14)C and choline-(3)H) it was shown that the lecithin formed retained the same ratio of the two labels, which indicated that lysolecithin was converted to lecithin through an acylation reaction. The newly formed lecithin was shown to be bound to the plasma membrane fraction; this suggested that it is incorporated into the structure of the membrane itself.

The metabolism of chylomicron cholesteryl ester in rat liver. A combined radioautographic-electron microscopic and biochemical study.

Chylomicrons containing labeled cholesterol, mainly (70%) present as cholesteryl ester, were injected intravenously into intact rats, and samples of liver were obtained 27-210 min later. Most (58-75%) of the injected label was recovered in the liver after 27-75 min. Hepatic uptake occurred without hydrolysis of the labeled cholesteryl ester. In separate experiments, in vitro perfusion of livers of similarly treated rats for 30-35 min washed out only 3-9% of the labeled sterol. Samples of liver and small intestine were prepared for electron microscopy with Aquon as the dehydrating agent. Good retention (70% or more) of labeled cholesterol and satisfactory preservation of ultrastructure were obtained. After 30 min, the radioautographic reaction was localized mainly over the region of the cell boundary of the parenchymal liver cells, with fewer grains being present over intracellular organelles. At later time intervals, when considerable hydrolysis of the labeled cholesteryl ester had occurred, the radioautographic reaction was more evenly distributed. Phagocytosed labeled lipid was seen in Kupffer cells after the larger lipid load; phagocytosis by parenchymal cells was not seen. In other experiments, cholesteryl ester hydrolase activity was found in all subcellular fractions, the microsome and plasma membrane fractions showing the highest activity per mg protein. The mechanism of cholesteryl ester transport into the liver cell may involve: (1) hydrolysis at the cell surface; or (2) slow entry of intact molecules followed by intracellular hydrolysis of the ester bond.

Phospholipases in arterial tissue. IV. The role of phosphatide acyl hydrolase, lysophosphatide acyl hydrolase, and sphingomyelin choline phosphohydrolase in the regulation of phospholipid composition in the normal human aorta with age.

The role of phospholipases in the regulation of the changing phospholipid composition of normal human aortae with age was studied. Portions of grossly and histologically lesion-free ascending aortae from 16 females and 29 males obtained at autopsy, were analyzed for deoxyribonucleic acid (DNA), phospholipid, and cholesterol content and phospholipid composition. Enzymic activity toward four substrates, lecithin (LE), phosphatidyl ethanolamine, lysolecithin, and sphingomyelin (SP), was determined on portions of the same homogenate. By regression analysis for correlation between all determinations and age the following results were obtained: (a) total phospholipids and choleserol increased linearly with age; (b) the increase in sphingomyelin accounted for about 70% of the phospholipid increment; (c) hydrolysis of lecithin and phosphatidyl ethanolamine increased markedly with age, that of lysolecithin only moderately; (d) hydrolysis of sphingomyelin decreased with age; and (e) an inverse relation between the SP/LE ratio and age and sphingomyelinase/lecithinase activity and age was obtained. These results were interpreted to indicate that a causal relation exists between the fall in sphingomyelinase activity, both absolute and relative to lecithinase activity, and the accumulation of sphingomyelin with age.

Defective metabolism of hypertriglyceridemic low density lipoprotein in cultured human skin fibroblasts. Normalization with bezafibrate therapy.

The metabolism of hypertriglyceridemic low density lipoprotein (HTG-LDL) was investigated in upregulated cultured human skin fibroblasts. Low density lipoprotein (LDL) was isolated by zonal centrifugation from the plasma of seven HTG subjects, before and 2 wk after the initiation of bezafibrate (BZ) therapy. HTG-LDL is a cholesterol-poor, triglyceride-rich lipoprotein of smaller diameter than BZ-LDL or normal LDL (N-LDL). Binding, cell association, and proteolytic degradation of HTG-LDL were compared with that of BZ-LDL and N-LDL and were found to be significantly lower by a paired t test analysis (P less than 0.001). After 6 h preincubation with unlabeled HTG-LDL, the incorporation of [14C]acetate to sterols was significantly higher than with BZ-LDL or N-LDL (577 +/- 43.7; 330 +/- 41.5; 262 +/- 47, mean +/- SE, picomoles sterols per milligram cell protein per 2 h, respectively; P less than 0.001 by paired t test). To determine the effectiveness of HTG-LDL and BZ-LDL on the down-regulation of LDL receptor activity, up-regulated cells were incubated for 48 h with HTG-LDL and BZ-LDL. LDL receptor activity was significantly higher after preincubation with HTG-LDL compared with BZ-LDL, and the rates of sterol synthesis were similarly increased. These results demonstrate that HTG-LDL does not down-regulate the LDL receptor activity as efficiently as BZ-LDL and that its cholesterol content is not enough to adequately suppress cellular sterol synthesis. Significant correlation between LDL composition and cholesterol synthesis by cultured cells was found with all LDL preparations over a wide range of cholesteryl ester to protein ratio (0.8-2.2). This correlation indicates that the compositional and structural abnormalities of HTG-LDL, and especially the low cholesterol content of the lipoprotein, alter LDL metabolism and cellular cholesterol formation.

High density lipoproteins reduce the uptake of low density lipoproteins by human endothelial cells in culture.

Endothelial cells, explanted from human umbilical veins and cultured, maintained morphological characteristics of vascular endothelium. When exposed to human serum lipoproteins, the cells bound and took up low density lipoproteins in preference to high density lipoproteins. High density lipoproteins reduced markedly the uptake of low density lipoproteins and affected surface binding to a lesser extent. These data suggest that the different levels of high density lipoprotein encountered in normal plasma of males and females could modulate differently the transendothelial transport of low density lipoproteins and provide a possible explanation for the lesser severity of atheromatosis in the aortic intima of premenopausal females.

Surface binding and interiorization of homologous and heterologous serum lipoproteins by rat aortic smooth muscle cells in culture.

Rat aortic smooth muscle cells in culture were incubated with rat or human iodinated low and high density lipoprotein at 5-50 mug/ml for 3 h. With the homologous lipoproteins, 25-49% of total cellular protein radioactivity was trypsin releasable and was considered as surface-bound radioactivity, while the balance represented cellular uptake. The ratio of surface-bound to cellular label was higher when the cells were incubated with human lipoproteins and was about 9 : 1 with human high density lipoprotein. Cellular uptake of rat low density lipoprotein was about twice that of rat high density lipoprotein, while degradation of labeled protein, which had presumably followed protein uptake, was similar and ranged from 20 to 25% of protein uptake in 3 h. Experiments designed to test the effect of cell density on lipoprotein uptake have shown that the uptake was related inversely to cell density. Thus, the lower lipoprotein uptake encountered in the rat smooth muscle cells, compared to that described for human fibroblasts (Goldstein, J.L. and Brown, M.S. (1974) J. Biol. Chem. 249, 5153-5162), could be due in part to the much lower cell density used in the latter studies, as well as to cell type and species difference.

Bovine aortic endothelial cells display macrophage-like properties towards acetylated 125I-labelled low density lipoprotein.

Bovine aortic endothelial cells in culture were shown to take up and degrade acetylated 125I-labelled low density lipoproteins (125I-acetylated LDL) in preference to 125I-labelled low density lipoproteins (LDL). The confluent cultures of endothelial cells had a higher rate of degradation of 125I-acetylated LDL than did subconfluent cells. The ratio of degradation of 125I-acetylated LDL to 125I-LDL was 3--9 in the case of the endothelial cells, 0.06--0.11 for aortic smooth muscle cells and 18 for mouse peritoneal macrophages. The uptake and degradation of acetylated LDL by the endothelial cells was accompanied also by an increase in cellular cholesterol. The present findings indicate that cultured endothelial cells display certain macrophage-like properties towards serum lipoproteins.

Cholesterol content and sterol synthesis in human skin fibroblasts and rat aortic smooth muscle cells exposed to lipoprotein-depleted serum and high density apolipoprotein/phospholipid mixtures.

Confluent cultures of human skin fibroblasts and rat aortic smooth muscle cells were shown to lose 15-27% of their cellular cholesterol upon replacement of the fetal calf serum with human high density lipoprotein (50 mug cholesterol/ml) or lipoprotein-depleted serum at a concentration equivalent to 40% whole serum. Addition to the latter medium of high density apoliproprotein/phospholipid mixtures resulted in further enhancement of cellular cholesterol loss which was evident by 12 h of incubation. Human skin fibroblasts that had been enriched in cholesterol by previous incubation with low density lipoprotein lost their cholesterol in the presence of a high density apolipoprotein/sphingomyelin mixture as readily as non-enriched cells. Concomitant with the marked cholesterol depletion there was a stimulation of sterol synthesis from acetate. The more pronounced loss of cellular cholesterol induced by the presence of phosphatidylcholine or sphingomyelin resulted in a greater incorporation of acetate into sterol in both smooth muscle cells and skin fibroblasts. The present findings indicate that peripheral cells, in spite of their capacity to synthesize cholesterol, depend on exogenous cholesterol for the maintenance of normal levels. It is suggested that the native cholesterol "acceptor" in the lipoprotein-depleted serum is an apolipoprotein which under the experimental conditions can form a complex with phospholipids and might also represent the physiological cholesterol "acceptor" in peripheral lymph.

Interference with the transport of heparin-releasable lipoprotein lipase in the perfused rat heart by colchicine and vinblastine.

The effect of pretreatment with colchicine or vinblastine on the lipoprotein lipase activity of rat heart was studied. Administration of colchicine or vinblastine 4 h prior to perfusion of the heart caused a very marked reduction in lipoprotein lipase activity released into the perfusate within 1 min of heparin perfusion. At the same time an increase in residual heart lipase occurred so that total lipoprotein lipase content of the heart (heparin releasable plus residual) did not change. The colchicine effect was dose and time dependent; no decrease in heparin-releasable enzyme activity occurred after only 30 min of pretreatment or upon addition of colchicine into the perfusate. These results indicate that colchicine did not impede enzyme synthesis or its release from the cell surface, but may have interfered with the transport of lipoprotein lipase from the site of its synthesis to the endothelial cell surface.

The role of lysophosphatidylcholine and apolipoprotein A in the cholesterol-removing capacity of lipoprotein-deficient serum in tissue culture.

Lipoprotein-deficient serum (d greater than 1.21 or 1.25 g/ml fraction) is commonly used to deplete cellular cholesterol from cultured cells and presently we have studied some of the potential promoters of this process. Although serum albumin is the main protein component of the fraction, its cholesterol-removing capacity was quite limited, even in the presence of lysophosphatidylcholine, which is the major phospholipid of the d greater than 1.25 g/ml infranatant of serum. On the other hand, apolipoprotein A1 especially when complexed with lysophosphatidylcholine promoted considerable release of cellular cholesterol. The cholesterol-removing capacity of lysophosphatidylcholine alone was related to the fatty acid chain length and was low when the fatty acid chain length was below C-14. The release of cellular cholesterol is not related to shedding of surface glycoproteins and depends on the presence of suitable acceptors in the medium. Such acceptors were presently found in an ultrafiltrate of serum prepared by membrane filtration. It is proposed that in human serum there are low molecular weight protein-phospholipid complexes (less than 100,000), which can cross the capillary endothelial barrier, in preference to lipoproteins, and promote cholesterol removal from peripheral cells.

Effect of TNF on triacylglycerol in cultured vascular smooth muscle cells.

Smooth muscle cells (SMC) isolated from bovine aorta or human saphenous vein were cultured and used to study the putative effect of recombinant human tumor necrosis factor (TNF) on lipid metabolism in vascular cells. Addition of TNF to the culture medium for 24-48 h resulted in an increase of [3H]oleic acid uptake and esterification into lipids. The effect could be seen already with 0.3 ng/ml and was maximal with 30 ng/ml. The effect of TNF was mainly on the incorporation of [3H]oleic acid into triacylglycerol which increased by 140% in the bovine cells. There was also a significant increase in [3H]cholesteryl ester. In the human SMC there was a 40% increase in [3H]oleic acid into total lipids, while the rise in [3H]triacylglycerol ranged between 60-90%. TNF did not modulate cellular triacyglycerol synthesis in cultured mouse peritoneal macrophages. Since TNF was shown to be synthesized and secreted not only by macrophages but also by smooth muscle cells, it could play an autocrine role in lipid metabolism during development of atherosclerotic lesions. The cellular population of the lesions, i.e., predominance of macrophages or smooth muscle cells, could determine the relative proportion of triacylglycerol accumulation.

Interaction of concanavalin A with membrane-bound and solubilized lipoprotein lipase of rat heart.

Concanavalin A was used to study the configuration of lipoprotein lipase at the surface of capillary endothelium. Incubation of heart homogenates with increasing concentrations of concanavalin A for 5-60 min resulted in inhibition of up to 50% of enzyme activity. The inhibition was related to the concentration of lectin and the time of incubation and was fully reversible by postincubation with alpha-methyl-D-mannoside. Rat hearts were perfused for 5-60 min and lipoprotein lipase activity determined in postheparin perfusates and in the perfused heart. When the lectin was introduced into the perfusate a significant reduction of heparin-releasable enzyme was found after 30 min of perfusion. The missing enzyme could be recovered by postperfusion with alpha-methyl-D-mannoside, but not by addition of the sugar to the perfusate withdrawn from the apparatus. These results suggested binding of lectin to the surface-located enzyme and support for such a binding was obtained by the finding of release of labeled lectin into the perfusate by heparin. Perfusion of hearts with concanavalin A for 60 min resulted also in a fall in nonreleasable lipoprotein lipase. The mechanism of this fall is not due to impairment of enzyme synthesis, as leucine incorporation into protein was not reduced. Since neither perfusion nor postincubation with alpha-methyl-D-mannoside restored enzyme activity, the fall was most probably due to irreversible inhibition. It is concluded that mannose residues of lipoprotein lipase in heart homogenates and at the endothelial surface of heart capillaries are available to interact with a specific lectin. Such an interaction renders the enzyme less releasable by heparin during perfusion and causes a significant inhibition of enzyme activity in homogenates.

Rat heart in culture as a tool to elucidate the cellular origin of lipoprotein lipase.

Lipoprotein lipase was determined in 5-day old cell cultures derived from hearts of newborn rats. With the help of the preplating method the cells were subdivided into cultures containing predominantly cardiac myocytes and into those composed mainly of mesenchymal cells. Lipoprotein lipase activity, associated with the mesenchymal cells was ten times higher than the activity found in the cultures containing mainly the myogenic cells. It is suggested that the mesenchymal cells are the progenitors of lipoprotein lipase in rat heart.

Colchicine-induced inhibition of plasma lipoprotein lipase release in the intact rat.

The release of plasma lipoprotein lipase by heparin was studied in fed and food-deprived rats pretreated with colchicine and vinblastine. Four hours after the administration of either drug the lipoprotein lipase activity released by heparin was only half of that found in controls. Colchicine affected the release of both protamine-sensitive and protamine-resistant lipoprotein lipase. It is suggested that colchicine and vinblastine interfere with the transport of lipoprotein lipase from the site of its storage to the vascular cell surface.

Angiotensin II stimulates receptor-mediated uptake of LDL by bovine adrenal cortical cells in primary culture.

Bovine adrenal cells were isolated from the subcapsular region of the gland to obtain cultures enriched in cells of the zona glomerulosa. The cells kept in primary cultures were shown to respond to angiotensin II and adrenocorticorticotropin (ACTH) by a significant increase in aldosterone production. These primary adrenal cultures were used to study the effect of angiotensin II on LDL metabolism. Addition of angiotensin II for 48 h to the culture medium resulted in a 200-300% increase in LDL metabolism, and the lowest effective concentration was 10(-8) -10(-9) M. The angiotensin II effect became evident after 12-16 h of incubation. To compare the metabolism of the 125I-labeled protein moiety to that of cholesteryl ester of LDL, the lipoprotein was labeled also with cholesteryl linoleyl ether, a nonhydrolyzable analog of cholesteryl ester. Under basal conditions and in the presence of angiotensin II or ACTH the ratio of [3H]cholesteryl linoleyl ether to 125I indicate some preferential uptake of the cholesteryl ester moiety. Stimulation of specific LDL binding at 4 degrees C and LDL metabolism at 37 degrees C by 10(-7) M angiotensin II occurred at all concentrations of LDL studied. Linearization of the kinetic data showed that angiotensin II increased the LDL receptor number significantly but not the affinity of the LDL receptor for its ligand. The present findings indicate that in analogy to ACTH, angiotensin II can influence receptor-mediated uptake of LDL by adrenal cortical cells. It remains to be shown whether the angiotensin II effect on LDL metabolism is limited to adrenal cells or will affect other cells which express the angiotensin II receptor.