Characterization of triglyceride rich lipoproteins with very light density by ultracentrifugation and agarose gel electrophoresis using triglyceride- and cholesterol-staining

Hypertriglyceridemia is an independent risk factor for atherosclerosis. This risk is most likely due to accumulation of circulating triglyceride rich lipoproteins with heterogeneous particles. The identification and characterization of these triglyceride rich lipoproteins is important to detect abnormality of triglyceride metabolism. In the present study, we developed a new method that combines ultracentrifugation and agarose gel electrophoresis with triglyceride- and cholesterol-staining. We investigated 40 subjects with hypertriglyceridemia. Triglyceride rich lipoproteins with very light density were recovered in the aqueous fraction after ultracentrifugation (17,000 x g, 15 min). The lipoproteins recovered in the aqueous fraction contained chylomicrons, if present, their remnants, and light-VLDL (d <1.000 g/ml) containing apoB-100, but not normal VLDL (d <1.006 g/ml) and IDL. Triglyceride rich lipoproteins in the aqueous fraction were characterized by electrophoresis patterns of triglyceride- and cholesterol-staining. Forty patients with hypertriglyceridemia were separated into 8 groups according to their electrophoretic patterns. In lipoproteins recovered in the aqueous fraction from each group, the triglyceride level was correlated with the respective cholesterol level. In summary, a system using ultracentrifugation and agarose gel electrophoresis with triglyceride- and cholesterol-staining is useful for characterization of triglyceride rich lipoproteins and their remnants.     

The in vitro interactions between serum lipoproteins and proteoglycans of the neointima of rabbit aorta after a single balloon catheter injury.

The effect of injury-induced alterations in the aortic neointimal proteoglycans on their binding with homologous serum lipoproteins was examined. Proteoglycans of the aortic intimal-medial tissues of rabbits that had undergone denudation with a balloon catheter 12 weeks earlier were isolated after homogenization of the tissues in 0.33 M sucrose, ultracentrifugation and subsequently by gel-exclusion chromatography. Lipoproteins from the plasma of healthy donors were prepared by sequential, ultracentrifugal floatation after density adjustment with KBr. To study the interactions, aliquots of electrophoretically pure very low-density lipoproteins (VLDL, d less than 1.006 g/ml), low-density lipoproteins (LDL, d = 1.019-1.063 g/ml), or high-density lipoproteins (HDL, d = 1.210 g/ml) were incubated with proteoglycans in the presence of Ca++ and Mg++ at 4 C. The amount of cholesterol found in the resulting pellet was measured as a marker of the binding capacity of the proteoglycans. Among lipoprotein fractions both VLDL and LDL showed strong binding with proteoglycans, whereas no appreciable binding was observed when incubation experiments were done with HDL. There were significant differences in the lipoprotein binding capacity of proteoglycan of control and injured animals, indicating that injury induced changes in proteoglycan composition exert profound influences on their ionic interactions.     

Apoprotein B in high density lipoproteins: variations according to the type of hyperlipoproteinemia

Two techniques for the separation of high density lipoproteins are compared: ultracentrifugation and concanavalin A precipitation. There is a significant correlation for the high density lipoproteins cholesterol levels found by the two techniques (r = 0.66, p less than 0.001). Nonetheless ultracentrifugation gives higher values than precipitation. When the ultracentrifuged high density lipoproteins fraction was precipitated with concanavalin A, a highly significant correlation was found between the amounts of apoprotein B and cholesterol present in the precipitate (r = 0.90, p less than 0.001). These results indicate the presence of a subfraction of the ultracentrifuged high density lipoproteins fraction which is precipitated by concanavalin A and which contains apoprotein B. This fraction is present in varying proportions according to the type of hyperlipoproteinemia, types iib and IV having the greatest amounts. In conclusion, only the concanavalin A precipitation technique results in a separation of apoprotein B-containing and non-apoprotein B-containing lipoproteins. This renders it preferable to ultracentrifugation for determining atherogenicity risks.     

Evaluation of the classical methods for the diagnosis of type III hyperlipoproteinemia

Summary Familial type III hyperlipoproteinemia is characterized by the presence of elevated plasma levels of very low density lipoproteins (VLDL) which contain an increased amount of cholesterol and by the presence of a significant amount of lipoproteins with an intermediate density between that of VLDL and low density lipoproteins (LDL); the intermediate density lipoproteins, designated IDL or Lp III, have a slower electrophoretic migration rate than VLDL, and are found in the ultracentrifugal top fraction as a contaminant. Classically, the diagnosis of type III is based on the demonstration of beta-migrating lipoproteins in the ultracentrifugal top fraction (density <1.006), thus floating beta-lipoprotein. More recently, it has been proposed that an elevated VLDL-cholesterol to triglyceride ratio is diagnostic of the disorder. In the present report, we have compared the two methods for their diagnostic value and have concluded that the chemical index definition is the more reliable method for the diagnosis of type III hyperlipoproteinemia.     

Distribution of apolipoprotein E isoforms between very low and high density lipoproteins of normal subjects and hyperlipidemic patients with special reference to type III hyperlipidemia

Serum very low (VLDL) and high density lipoproteins (HDL) from 17 hyperlipidemic patients and 10 normal subjects have been isolated by preparative ultracentrifugation, and the electrophoretic patterns of apolipoprotein E (apo E) isoforms in the lipoproteins have been examined by isoelectric focusing. Type III hyperlipidemia (dyslipoproteinemia) has been suggested to be a disease caused by an abnormal mutant of the apo E-3 isoform. In accordance with this, all patients with type III hyperlipidemia in the present study showed lack of apo E-3 in VLDL. However, all these patients demonstrated a protein zone corresponding to apo E-3 in their HDL fraction. Patients with other types of hyperlipidemia or normal subjects who showed an apo E-4 variant in VLDL had an HDL that lacked apo E-4. The results support the hypothesis that type III hyperlipidemia is due to an abnormal composition of the VLDL particles rather than a result of an abnormal mutant of apo E-3.     

Lipoproteins in human atherosclerotic vessels. I. Biochemical properties of arterial low density lipoproteins, very low density lipoproteins, and high density lipoproteins.

Lipoproteins extracted from the human aortic intima into 1.65 M NaCl were quantitated and characterized biochemically and by electron microscopy following separation in the preparative ultracentrifuge. The arterial lipoproteins, although separated and designated according to the density classes used for the serum lipoproteins, were distinctly different from their serum counterparts. The amount of lipoproteins in the low density range of d 1.063 to 1.006 (arterial LDL) and in the very low density range of d < 1.006 (arterial VLDL) extracted from arterial intima increased with increasing intimal lipid content. In contrast, the concentration of lipoproteins in the high density range of d 1.210 to 1.063 (arterial HDL) was small and did not change with the severity of atherosclerosis.Arterial VLDL, LDL and its subfractions, LDL₁ (d 1.006 to 1.019) and LDL₂ (d 1.019 to 1.063), were markedly heterogenous and contained unusually large particles, which were isolated by Bio-Gel A-150. These particles showed a pitted and cratered appearance by scanning electron microscopy and were immunochemically unreactive and had no electrophoretic mobility. The lipid and amino acid composition of the arterial VLDL and LDL fractions as well as their electrophoretic, chromatographic and analytical flotation behavior was distinctly different from that of their serum lipoprotein counterparts. Arterial VLDL, in sharp contrast to serum VLDL, was rich in cholesteryl ester and poor in triglycerides. Arterial VLDL also showed no electrophoretic mobility and only half of the preparations reacted to LDL antisera. Acid mucopolysaccharides were detected in the arterial VLDL and LDL fractions in association with the large size particles which lacked electrophoretic mobility and immunochemical reactivity and showed only a “saw tooth” pattern in the analytical ultracentrifuge. Arterial LDL and LDL₂ contained a smaller sized population of particles as separated by Bio-Gel A-150. These particles exhibited a reaction of complete identity with serum LDL when reacted against LDL antiserum. However these particles had a greater electrophoretic mobility and different amino acid composition than did serum LDL and LDL₂. An asymmetrical peak with a mean SF of 7.3 was demonstrated by these particles in the analytical ultracentrifuge.The over-all studies suggest that lipid deposition in atherosclerotic plaques is associated with the accumulation of lipoproteins with biochemical and ultrastructural properties unlike those of serum lipoproteins. The presence of these lipoproteins in the arteries may be a result of the interaction of serum and arterial lipoproteins with acid mucopolysaccharides and of lipoprotein synthesis and degradation in the arteries.     

Demonstration of an atypical pre- ß-lipoprotein in human serum

An atypical pre-β-lipoprotein of human serum has been detected by agarose gel electrophoresis in four patients, three of whom were siblings. This lipoprotein differed from the pre-β lipoprotein previously observed with this technique by sedimenting on ultracentrifugation at density 1.006.     

Cryosopic determination of renal tissue osmolality by an ultramicro method

Summary A micromethod for the determination of tissue osmolality is described. In essence it consists of ultracentrifugation of tissue and cryoscopic analysis of osmolarity of the supernatant in a nanoliter osmometer. The method was applied to renal cortical and medullary tissues in rats with and without Diabetes insipidus, and in dogs.Predoctoral Trainee in Physiology, supported by National Heart Institute Training Grant HE-5322.     

Myelomatosis with type III hyperlipoproteinemia: clinical and metabolic studies

We investigated the metabolism of intermediate-density lipoproteins (IDL [1.006 to 1.019 g per milliliter]) and low-density lipoproteins (LDL [1.019 to 1.063 g per milliliter]) in two men with Type III hyperlipoproteinemia associated with myelomatosis. In vivo kinetic studies using radiolabeled autologous lipoproteins demonstrated a greatly reduced fractional catabolic rate of IDL, relative to control values (patients vs. normal, 0.006 and 0.025 per hour vs. 0.20 +/- 0.08 per hour [mean +/- S.E.M]) and a greatly prolonged IDL-to-LDL conversion time (45 and 17 hours vs. 5.4 +/- 1.6 hours). In studies in vitro, LDL from both patients failed to bind to the LDL receptor of normal blood lymphocytes, whereas LDL from subjects with familial Type III hyperlipoproteinemia bound normally to the receptor. In one patient immunoglobulin was shown to be associated with IDL and LDL. Thus, hyperlipoproteinemia reflected an impaired metabolism of IDL, probably secondary to the binding of immunoglobulin to the lipoproteins. A similar impairment of receptor-mediated LDL catabolism did not elevate the plasma LDL concentration because of the low IDL-to-LDL conversion rate.     

Vascular collagens. General review

The analysis of vascular collagens proved difficult due to high insolubility. In the seventies, compositional studies concerned a fraction of 30% of total collagen. Type I collagen, far less soluble than type III, was not fully extracted, resulting in an overevaluation of type III collagen. When the proportions of collagens are measured on the totality of the material, type I collagen represents 60%, type III, 30% and the remaining 10% are constituted by type V and minor collagens. Atheromatous plaques contain a little more collagen than normal arteries and the proportion of type I remains 60%. Whereas several experiments demonstrated that slices of atheromatous arteries in vitro synthesize more collagen than slices of normal arteries, the possible mechanism of interaction between plasma lipoproteins (normal or abnormal) and collagen metabolism remains unknown. Vein and capillary collagens have been rarely studied. Basement membrane collagen does not seem to be increased in quantity in the thickened basement membrane of diabetic patients capillaries.     

The pathophysiology of cholesterol metabolism in man

Summary The concentration of low density lipoprotein in human plasma depends on the balance between its rates of synthesis and catabolism. Although both processes appear to be independently regulated they occur side by side in the liver and may be linked via the activity of the high affinity low density lipoprotein receptor on hepatocyte membranes. Dietary changes such as cholesterol feeding or variation in fat content can promote synthesis of the lipoprotein without changing catabolism while other interventions (e.g. sequestrant resin therapy) have the opposite effect. These different responses may be explained on the basis of compartmentalisation of regulatory sterol pools in the liver cell.Abbreviations apo apolipoprotein - FH familial hypercholesterolaemia - HDL high density lipoproteins (d=1.063–1.21 kg/l) - HMG CoA reductase 3-hydroxy-3-methylglutaryl Coenzyme A reductase - IDL intermediate density lipoproteins (d=1.006–1.019 kg/l) - LDL low density lipoproteins (d=1.019–1.063 kg/l) - VLDL very low density lipoproteins (d<1.006 kg/l)     

Overexpression of lipoprotein lipase in transgenic rabbits leads to increased small dense LDL in plasma and promotes atherosclerosis

Lipoprotein lipase (LPL) is a key enzyme in the hydrolysis of triglyceride-rich lipoproteins. Previous studies using transgenic mice and rabbits have demonstrated that high level of LPL activity in adipose and skeletal muscle protects against diet-induced hypercholesterolemia and subsequently prevents aortic atherosclerosis. However, it is unknown, per se, whether increased LPL activity itself is antiatherogenic, or whether the antiatherogenic effect of LPL is dependent upon the LPL lipid-lowering effect. To address this issue, we fed LPL transgenic and littermate rabbits diets containing different amounts of cholesterol (0.3-0.6%) adjusted to maintain their plasma cholesterol concentrations at similarly high levels for 16 weeks. We analyzed their lipoprotein profiles and compared their susceptibility to atherosclerosis. The results showed that the overexpression of LPL in transgenic rabbits reduced remnant lipoproteins (beta-VLDL, d<1.006 g/ml) but concomitantly led to a significant increase of the large (d=1.02-1.04 g/ml) and small LDLs (d=1.04-1.06 g/ml) compared to the amounts in control rabbits. Furthermore, we found that with equally high hypercholesterolemia, transgenic rabbits developed 1.8-fold more extensive aortic atherosclerosis than control rabbits. To examine the hypothesis that altered lipoprotein profiles may be responsible for the enhanced atherosclerosis in transgenic rabbits, we studied the atherogenic properties of apoB-containing lipoproteins in vitro. These studies revealed that small-sized LDLs of transgenic rabbits were more susceptible to copper-induced oxidation and had higher affinity to biglycan than large remnant lipoproteins. We conclude, therefore, that LPL exerts a dual function in terms of its atherogenicity, namely antiatherogenicity, through enhancing receptor-mediated remnant lipoprotein catabolism and proatherogenicity via the generation of a large amount of small-sized LDLs. At an equal atherogenic-cholesterol level, small and dense LDLs are more atherogenic than large remnant lipoproteins.     

Presence of Apo B48, and relative Apo CII deficiency and Apo CIII enrichment in uremic very-low density lipoproteins

In the present study, lipid and apolipoprotein composition of very low density lipoprotein (VLDL) was analyzed in 39 patients with end-stage renal failure by comparison with 41 healthy subjects. Uremic patients had an increase of serum triglycerides (TG) concentration by comparison with control values. This increase of serum TG was associated with an increase of VLDL which had a normal percent amount of main components. Furthermore a mid-band between VLDL and low density lipoproteins (LDL) on polyacrylamide gel was observed in 22 out of 39 uremic patients but in only 1 out of 41 control subjects. In uremic VLDL Apo B48 was more frequently observed than in control VLDL (p less than 0.05). Furthermore, the content of Apo CII expressed as percent of total Apo C was significantly (p less than 0.001) decreased in uremic VLDL (19.13 +/- 4.54 p. cent) as compared to normal VLDL (23.57 +/- 4.40 p. cent). Apo CIII-O was significantly (p less than 0.001) increased (9.58 +/- 7.19 p. cent vs 5.55 +/- 6.12 p. cent, whereas Apo CIII-1 and Apo CIII-2 distribution was not modified in uremic VLDL. These anomalies were present in uremic patients even when no elevation of fasting serum TG was present. No significant change was observed in uremic patients before their fifth as compared to their first hemodialysis (HD) session, respectively, for any of the parameters studied. Advanced chronic renal failure is associated with a variety of anomalies of TG-rich lipoproteins isolated at d less than 1.006 g/ml which are not reflected by the degree of hypertriglyceridemia and are not corrected by the first four HD sessions.     

Canine hyperlipoproteinemia and atherosclerosis. Accumulation of lipid by aortic medial cells in vivo and in vitro.

Dogs maintained for 1 year on a semisynthetic diet containing hydrogenated coconut oil and cholesterol developed hypercholesterolemia. In those cases where plasma cholesterol levels exceeded 750 mg/100 ml, the animals also developed severe atherosclerosis. This atherogenic hyperlipoproteinemia was characterized by the presence of beta very low density lipoproteins (B-VLDL), increased levels of low density lipoproteins (LDL), and the occurrence of the HDLc lipoproteins. In all of these cholesterol-rich lipoproteins the arginine-rich apoprotein (ARP) was prominent. Moreover, the HDLc (d = 1.006-1.02) contained the ARP as the only detectable apoprotein. The atherosclerosis involved the abdominal aorta, coronary and cerebrovascular arteries, and many of the peripheral arteries. Histologically, the aortic lesions were characterized by a variable intimal proliferative response and extensive medial lipid deposition. In the peripheral, coronary, and cerebral arteries, the lesions were more extensive and involved primarily the media of the vessel wall, with little intimal reaction in many cases. The correlation between the in vivo disease process and the response of aortic smooth muscle cells (SMC) grown in tissue culture to the various cholesterol-induced lipoproteins was examined. B-VLDL, LDL, and HDLc (but not HDL2) caused a marked accumulation of free and esterified cholesterol in the SMC. The cholesterol accumulation was found to be more extensive in canine SMC than in swine smooth muscle cells or smooth muscle cells of other species in response to a similar lipoprotein cholesterol concentration. The enhanced sterol uptake appeared to be a property of canine smooth muscle cells rather than a property of the canine lipoproteins. These in vitro results may be related to the observed propensity for the development of medical disease that was demonstrated in the in vivo studies.     

Lipoproteins, post-heparin lipoprotein lipase and hepatic triglyceride lipase in patients with and without severe hyperlipemia caused by alcoholism

Because of the high incidence for development of a secondary hyperlipemia during chronic alcohol intake, this study was performed to look for a possible reason, why some patients produce severe hyperlipemia and other ones not. 15 male patients with chronic alcoholism (group I) who produce under influence of alcohol a secondary type-V hyperlipoproteinemia (type-V HLP) were compared with 15 male controls. Additionally, 8 male patients with chronic alcoholism (group II) who were normolipemic under alcohol abuse, and 7 male patients (group II) who had also produced type-V HLP under chronic alcohol abuse, but were teetotal since at least 6 months, were investigated. In comparison with controls, patients of group I showed significantly (p less than 0.01) increased plasma concentrations of very low-density lipoproteins (VLDL) and significantly decreased plasma concentrations of low-density lipoproteins (LDL), high-density lipoproteins2 (HDL2) and HDL3 (all p less than 0.01). Furthermore, the activities of postheparin lipoprotein lipase (LPL) and hepatic lipase (HTGL) were significantly decreased (both p less than 0.01). In patients of group III, the plasma concentrations of lipoproteins did not differ significantly from controls, but the activity of LPL was also significantly impaired (p less than 0.01), whereas the activity of HTGL was distinctly (p less than 0.01) increased. No significant difference between patients of group II and controls could be demonstrated. It is concluded that severe alcohol intake strongly impairs LPL in patients with chronic alcoholism. The pronounced increase of HTGL in patients of group III seems to protect these individuals from producing severe hyperlipemia under the influence of alcohol.     

Methods for analyzing lipoproteins

In recent years there have been considerable advances and changes in the investigations available to study lipoprotein metabolic disorders. The methods used up to now were based on physicochemical criteria (electrophoresis, ultracentrifugation, polyanionic precipitation). But the recent demonstration of the preponderant role of apoproteins in lipoprotein metabolism makes it essential to develop a method of analysis lipoproteins at the molecular level. Immunological methods seem to be the most appropriate and should allow us to study lipoprotein metabolism more closely.     

Hyperlipoproteinaemia and atherosclerosis in rabbits fed low-level cholesterol and lecithin

Dutch-Belted rabbits were fed for 18 months an atherogenic semipurified gel diet containing 14% hydrogenated coconut oil and 0.06% cholesterol (approximately 0.15 mg/kcal) or a non-atherogenic basal gel diet containing the same ingredients but with no coconut oil or cholesterol. Rabbits fed atherogenic diet developed hypercholesterolaemia (means 733 mg/dl at 16 months) and plasma lipoprotein (LP) distribution shifted from a pattern in which high-density lipoproteins (HDL) predominated to one in which very-low-density lipoproteins (VLDL) were predominant. Total cholesterol/triglyceride ratio in d less than 1.006 LP changed from 0.3 to 1.8. Plasma cholesterol and LP distribution returned to normal in rabbits fed atherogenic diet for 18 months followed by atherogenic diet plus 3% soya lecithin for an additional 4 months. Rabbits fed atherogenic diet for 18 months had extensive, usually full circumference fibromuscular plaques in main branches of coronary arteries and all portions of aorta which compromised lumen area by almost 50%. These lesions were modified in rabbits fed atherogenic diet plus lecithin. The plaques lacked foam cells and cholesterol clefts, were less cellular with a distinct fibrous surface and occupied less space. Animals fed basal diet did not develop hypercholesterolaemia (means 86 mg/dl at 16 months), although distribution of plasma LP shifted slightly in favour of increased low-density lipoproteins (LDL) and decreased HDL compared with rabbits fed standard commercial diet. Basal diet rabbits had no coronary atherosclerosis and only minimal focal foam cell lesions in proximal aorta. Liver injury including fatty change, cholangitis and portal fibrosis occurred in animals fed atherogenic diet. Thus, rabbits fed appropriate diets low in cholesterol accumulate cholesterol-enriched LP in their plasma and develop lesions in abdominal aorta and main branches of coronary arteries which are similar to those in man. Also, in this experimental model, dietary lecithin promotes a return to normal of the LP distribution profile and removal of lipid from established atherosclerotic plaque.     

Hyperlipoproteinaemia and atherosclerosis in rabbits fed low-level cholesterol and lecithin.

Dutch-Belted rabbits were fed for 18 months an atherogenic semipurified gel diet containing 14% hydrogenated coconut oil and 0.06% cholesterol (approximately 0.15 mg/kcal) or a non-atherogenic basal gel diet containing the same ingredients but with no coconut oil or cholesterol. Rabbits fed atherogenic diet developed hypercholesterolaemia (means 733 mg/dl at 16 months) and plasma lipoprotein (LP) distribution shifted from a pattern in which high-density lipoproteins (HDL) predominated to one in which very-low-density lipoproteins (VLDL) were predominant. Total cholesterol/triglyceride ratio in d less than 1.006 LP changed from 0.3 to 1.8. Plasma cholesterol and LP distribution returned to normal in rabbits fed atherogenic diet for 18 months followed by atherogenic diet plus 3% soya lecithin for an additional 4 months. Rabbits fed atherogenic diet for 18 months had extensive, usually full circumference fibromuscular plaques in main branches of coronary arteries and all portions of aorta which compromised lumen area by almost 50%. These lesions were modified in rabbits fed atherogenic diet plus lecithin. The plaques lacked foam cells and cholesterol clefts, were less cellular with a distinct fibrous surface and occupied less space. Animals fed basal diet did not develop hypercholesterolaemia (means 86 mg/dl at 16 months), although distribution of plasma LP shifted slightly in favour of increased low-density lipoproteins (LDL) and decreased HDL compared with rabbits fed standard commercial diet. Basal diet rabbits had no coronary atherosclerosis and only minimal focal foam cell lesions in proximal aorta. Liver injury including fatty change, cholangitis and portal fibrosis occurred in animals fed atherogenic diet. Thus, rabbits fed appropriate diets low in cholesterol accumulate cholesterol-enriched LP in their plasma and develop lesions in abdominal aorta and main branches of coronary arteries which are similar to those in man. Also, in this experimental model, dietary lecithin promotes a return to normal of the LP distribution profile and removal of lipid from established atherosclerotic plaque.     

Type III hyperlipoproteinema with apolipoprotein E2/2 genotype in Japan

Limited information is available concerning type III hyperlipoproteinemia (HLP) in the Asian population. Therefore, clinical and biochemical characteristics of type III HLP were examined in 16 Japanese patients. Mean plasma triglyceride (TG) and total cholesterol (chol) levels were 381 mg/dl and 253 mg/dl, respectively, and the mean very low density lipoprotein (VLDL)-chol/plasma TG ratio was 0.27, which were lower than those reported in Western countries. Eighty percent of the patients had high plasma remnant-like particles (RLP)-chol levels above 50 mg/dl and a high RLP-chol/plasma TG ratio above 0.1. Twelve patients (75.0%) were obese. Seven patients (43.8%) had type 2 diabetes mellitus and four patients (25.0%) had impaired glucose tolerance. Six patients (37.5%) had coronary heart disease (CHD), but none had peripheral vascular disease or xanthomas. TG-rich lipoproteins from type III HLP patients with diabetes mellitus stimulated cholesteryl ester synthesis by human macrophages significantly (p < 0.001) more than those from type III HLP patients without diabetes mellitus. In conclusion, the Japanese type III HLP patients had lower plasma TG and total chol levels and a lower VLDL-chol/plasma TG ratio, but CHD was more common. The patients were characterized by a high frequency of obesity and/or glucose intolerance. The TG-rich lipoproteins from type III HLP patients with diabetes mellitus were more atherogenic.