Apolipoprotein-B subclasses as acceptors of cholesteryl esters transferred by CETP.

BACKGROUND: Five apolipoprotein (apo)-defined apoB-containing lipoprotein (Lp) subclasses designated LpB, LpB:C, LpB:E, LpB:C:E and LpA-II:B:C:D:E are present in human plasma. This study was to determine whether these subclasses functioned equally as acceptors of cholesteryl esters (CE) transferred from high-density lipoproteins (HDL) by CE transfer protein in healthy subjects with normal and mildly increased plasma triglyceride (TG) levels. MATERIALS AND METHODS: After 4 h incubation of plasma from 14 subjects at 37 degrees C, apoB-containing lipoproteins were separated from HDL by heparin-Mn++ precipitation and fractionated by immunochemical methods into these five subclasses. The neutral lipid (NL) composition for each subclass was measured by gas chromatography (GC) and compared between 0 h and 4 h. A subclass was considered to be a CE acceptor if its CE content increased more than 5% at 4 h and a non-acceptor if no change was observed. RESULTS: Employing the above definition, TG-rich LpB:C and LpB:E + LpB:C:E functioned as CE acceptors and TG-poor LpB and LpA-II:B:C:D:E as non-acceptors. Both LpB:C and LpB:E + LpB:C:E could only actively accept CE as long as they retained their TG-rich character and displayed neutral lipid profiles similar to those of very low-density lipoproteins (VLDL) and intermediate density lipoproteins (IDL). When, as a result of lipolysis their TG content dropped below 25%, they ceased to function as CE acceptors. In subjects with elevated plasma TG, LpB:C was the dominant CE acceptor, a condition that may have pro-atherogenic consequences. CONCLUSIONS: Among the apoB-containing particles, LpB:C and LpB:C:E + LpB:E functioned as CE acceptors while LpB and LpA-II:B:C:D:E did not.

Accelerated cholesteryl ester transfer in plasma of patients with hypercholesterolemia.

To discern the mechanism(s) that underlie abnormal cholesteryl ester transfer (CET) in patients with hypercholesterolemia, we have studied this dysfunctional step in reverse cholesterol transport in 13 subjects with genetically heterogeneous forms of hypercholesterolemia (HC). In all HC patients, the mass of CE transferred in whole plasma from HDL to VLDL and LDL increased rapidly initially and was significantly greater than in controls at 1, 2, and 4 h (P less than 0.005). To further characterize this disturbance, we performed a series of recombination experiments. Combining HC d less than 1.063 containing acceptor VLDL + LDL with the d greater than 1.063 fraction from controls containing donor HDL + CE-transfer protein (CETP) and not the converse combination showed the same characteristics of accelerated CET noted with intact HC plasma, indicating that abnormal transfer was associated with the HC acceptor lipoproteins. When HC VLDL and its subfractions and LDL were isolated separately and then combined with control d greater than 1.063 fractions, accelerated CET was only associated with VLDL1. Consistent with an acceleration of the neutral lipid transfer reaction occurring between HDL and VLDL1 in HC in vivo, we found that the triglyceride/CE ratio was decreased in HC VLDL1 (P less than 0.001), and increased in HDL (P less than 0.25). CETP mass was significantly increased in HC plasma (HC 2.3 +/- 4 micrograms/ml vs. control 1.3 +/- 0.3 micrograms/ml; mean +/- SD; P less than 0.025). This series of observations demonstrate that CET is accelerated in the plasma of HC patients, and this disturbance results from dysfunction of the VLDL1 subfraction rather than an elevation of CETP levels. Since an abnormality of this type in vivo can lead to the accumulation of potentially atherogenic CE-enriched apoB-containing lipoproteins in plasma, it may be an additional previously unrecognized factor that increases cardiovascular risk in HC patients.

The significance of basal insulin levels in the evaluation of the insulin response to glucose in diabetic and nondiabetic subjects.

The level of insulin after an overnight fast (basal) in 37 obese and nonobese male subjects with normal and abnormal carbohydrate tolerance was directly related to the increase in insulin concentration during a 3 hr 100 g oral glucose tolerance test. Obesity, but not diabetes, was associated with an elevation of this basal insulin level. Thus obesity predicted with the magnitude of the insulin response to glucose ingestion. When the individual insulin values were expressed as per cent change from the basal level, this effect of obesity was excluded. The insulin levels of all subjects with normal carbohydrate tolerance promptly rose 5-7-fold, and reached peak values 1 hr after oral glucose. In contrast, the diabetic response (as per cent increase) was markedly reduced during the 1st hr, and maximal (but still subnormal) insulin levels were not attained until 2 hr. In all subjects the insulin response (quantitated by calculation of the area circumscribed by a plot of the per cent change in insulin with time) showed a significant inverse correlation with the glucose response. Thus increasing degrees of carbohydrate intolerance were associated with decreasing insulin responses. Elevated levels of insulin, in both the basal state and in response to glucose, were related to obesity.

Evidence for the chylomicron origin of lipids accumulating in diabetic eruptive xanthomas: a correlative lipid biochemical, histochemical, and electron microscopic study.

Plasma lipoprotein alterations in nine insulin-dependent diabetics with hyperlipemia have been related to the lipid accumulating in eruptive xanthomas evolving in these patients. Histochemical and electron microscopic examination of xanthomas have been correlated with the lipid analyses in order to obtain additional evidence regarding the lipoprotein origin of lipids accumulating in the lesions. Both analytical and morphologic evidence suggested that circulating chylomicrons significantly contribute to the xanthoma lipids. All the patients had large quantities of circulating triglyceriderich chylomicrons which carried approximately 70% of the triglyceride found in the plasma. The fatty acid pattern of chylomicron and xanthoma triglycerides were similar. Triglyceride constituted the major lipid found in the xanthomas when they were sampled during their eruption. These findings, take in conjunction with histochemical and electron microscopic evidence of chylomicron particles in the dermal capillary walls, support the theory that blood lipoproteins, and particularly chylomicrons, permeated the vascular walls and the triglycerides carried by these lipoproteins apparently accumulated in tissue macrophages and perithelial cells which evolved into foam cells. Initiation of appropriate therapy resulted in clearance of the chylomicronemia and a concomitant resolution of the xanthomas as reflected by a decrease in total xanthoma lipid. Sequential studies of resolving xanthomas in five patients revealed that xanthoma triglyceride was mobilized more rapidly than cholesterol, resulting in a redistribution of the xanthoma lipids, so that the resolving lesions were cholesterol rich. Consistent with this change in lipid composition, correlative electron microscopy revealed loss of amorphous material from many of the foam cell vacuoles.

Association of lysolecithin acyltransferase with the high density lipoproteins and its activation by the low density lipoproteins in normal human plasma.

The lysolecithin acyltransferase of human plasma is shown to be associated with the high-density lipoprotein fraction. Although the low density lipoproteins do not have intrinsic enzyme activity, their presence activated the enzyme 3--7-fold. This activation is not affected by heat-treatment of the low density lipoproteins, but is abolished by the addition of heparin.

Impaired granulocyte adherence in mildly diabetic patients: effects of tolazamide treatment.

Granulocyte adherence in 10 of 15 untreated asymptomatic hyperglycemic diabetic outpatients (mean fasting glucose +/- SEM, 289 +/- 16 mg/100 ml) was 62 +/- 7% of control values. After treatment (2--4 wk) with tolazamide (500 mg daily), adherence to the nylon fiber columns employed in this study returned to control levels in the seven patients whose fasting glucose levels fell, (mean, 192 +/- 16 mg/100 ml) and deteriorated in the three in whom fasting hyperglycemia worsened. Results of this study indicate that, even in some mildly diabetic patients, a reduction in granulocyte adherence similar to that found in insulin-dependent diabetics may impair the inflammatory response.

Whole-plasma and high-density lipoprotein subfraction surface lipid composition in IDDM men.

To determine whether compositional abnormalities are present in high-density lipoprotein (HDL) in patients with insulin-dependent diabetes mellitus (IDDM) that might negate its putatively protective cardiovascular effects, we studied the plasma lipoproteins of 12 men with varying degrees of clinical control (mean fasting glucose 193 +/- 10 mg/dl, mean glycoalbumin greater than 73% above control mean). The diabetic patients' basal plasma triglyceride, total- and free- (unesterified) cholesterol, HDL cholesterol (HDL-chol), and apolipoprotein AI, AII, and B concentrations were similar to those of control subjects, but the free-cholesterol-to-lecithin ratio, a new index of cardiovascular disease risk, was significantly increased in their plasma (0.97 +/- 0.14 vs. 0.88 +/- 0.07, P less than .02) and their very-low-density lipoprotein (VLDL)-low-density lipoprotein (LDL) subfraction (1.50 +/- 0.51 vs. 1.08 +/- 0.15, P less than .005). Although HDL2-chol was similar in diabetic and control groups, the HDL2-chol-to-free-cholesterol ratio (diabetic vs. control, 4.64 +/- 1.7 vs. 1.96 +/- 1.0 mumol/ml, P less than .025) and the sphingomyelin-to-lecithin ratio (0.23 +/- 0.08 vs. 0.20 +/- 0.09, P less than .025) were both significantly increased in the IDDM group. HDL3-chol was higher in the IDDM than in the control subjects (diabetic vs. control, 38.6 +/- 5.2 vs. 32.7 +/- 2.7 mg/dl, P less than .005). In contrast to whole plasma and the VLDL + LDL subfraction, the free-cholesterol-to-lecithin ratio of IDDM and control HDL subfractions were similar.(ABSTRACT TRUNCATED AT 250 WORDS)

Effects of insulin treatment on lipoprotein composition and function in patients with IDDM.

Normolipidemic patients of both sexes with insulin-dependent diabetes mellitus have the same pervasive changes in lipoprotein surface and core lipid composition. The disproportionate increase observed in their lipoprotein free (unesterified) cholesterol relative to the predominant surface phospholipid lecithin (phosphatidylcholine) is reflected by elevation of the FC-L ratio in their whole plasma, VLDL, HDL2, and HDL3. As a possible consequence of this qualitative disturbance, cholesteryl ester transfer is pathologically increased and the mass of cholesteryl ester transferred from HDL to VLDL + LDL is significantly greater in IDDM patients than in control subjects at 1, 2, and 4 hr (P less than 0.001). Consistent with accelerated CET in vivo, the TG-CE core lipid ratio was decreased in VLDL from six subjects (IDDM 9.5 +/- 0.8 vs. control 12.9 +/- 3.4; P less than 0.01) and increased in their HDL (diabetic 0.55 +/- 0.11 vs. control 0.42 +/- 0.04: P less than 0.025). These abnormalities in lipoprotein composition and CET do not correlate with glycemic control and persist after intensive management with s.c. insulin. They may be related to the peripheral hyperinsulinemia that is an unavoidable consequence of conventional s.c. insulin administration because preliminary studies indicate that these disturbances in lipoprotein composition and function are reversed when systemic insulin levels are lowered and insulin is delivered into the portal circulation from an i.p. catheter connected to an implanted programmable s.c. insulin pump.

Impaired granulocyte adherence. A reversible defect in host defense in patients with poorly controlled diabetes.

When the ability of granulocytes from 10 poorly controlled diabetic patients with fasting hyperglycemia and no evidence of ketoacidosis (mean fasting glucose 293 +/- 20 mg. per 100 ml.; mean +/- S.E.M.) to adhere to a nylon fiber column was assessed, the number of adherent granulocytes from whole blood was only 53 +/- 6 per cent of the values observed in controls. After antidiabetic treatment for one to two weeks and lowering of fasting glucose levels (mean 198 +/- 29 mg. per 100 ml.), adherence improved significantly (p less than 0.01) in the diabetics; however, their values were still subnormal (diabetic 74 per cent +/- 8 of control; p less than 0.02). Adherence values before and after treatment correlated with the fasting glucose level (r = 0.88, p less than 0.001). These findings suggest that, in addition to previously reported abnormalities in migration and the ingestion and killing of bacteria, granulocyte adherence may also be impaired in poorly controlled diabetic patients. This functional abnormality correlates directly with the fasting glucose and is reversed by insulin treatment. A defect of this type may compromise the normal inflammatory response in some diabetics and impair their capacity to resist infection.

Alterations in cholesteryl ester transfer, lipoprotein lipase, and lipoprotein composition after combined pancreas-kidney transplantation.

IDDM patients treated with conventional subcutaneous insulin have an abnormal increase in cholesteryl ester transfer (CET), the proatherogenic step in reverse-cholesterol transport that results in the enrichment of the apolipoprotein (apo) B-containing lipoproteins (VLDL, LDL) with cholesteryl ester (CE). This disturbance is closely linked to iatrogenic hyperinsulinemia and the nonphysiologic stimulation of lipoprotein lipase (LpL), a physiologic activator of CET, because lowering systemic insulin levels by administering insulin through the intraperitoneal insulin route normalizes LpL and CET. Hyperinsulinemia persists in IDDM patients who undergo successful pancreas-kidney transplantation (PKT) when their allografts are placed in the pelvis and drain into the iliac vein. Therefore, to determine whether hyperinsulinemia promotes CET in this setting, we studied CET, LpL, and insulin levels in 14 euglycemic normolipidemic IDDM PKT patients with near-normal kidney function (creatinine 1.5 +/- 0.4 mg/dl). Consistent with our prediction, the net mass of CE transferred from HDL to VLDL + LDL was significantly increased in the PKT group (P < 0.001) compared with nondiabetic renal transplant patients receiving the same immunosuppressive drugs and healthy control subjects. Both basal and arginine-stimulated insulin levels were increased above the kidney transplant group's levels and correlated with the mass of CE transferred at 2 h (r = 0.71, P < 0.05; r = 0.66, P < 0.05, respectively). Total basal LpL activities, LpL and hepatic triacylglycerol lipase activities, and LpL mass all tended to be higher than levels in healthy control subjects. Consistent with these changes in lipase activity, VLDL particle size was significantly reduced (P < 0.025) compared with that of control subjects. These findings indicate that PKT patients with systemically draining allografts have a persisting profile of potentially atherogenic disturbances in insulin levels, LpL, and CET that resemble IDDM patients treated with conventional subcutaneous insulin injections.

Effects of multiple daily insulin injections and intraperitoneal insulin therapy on cholesteryl ester transfer and lipoprotein lipase activities in NIDDM.

Although the relationship between the actions of cholesteryl ester transfer protein (CETP) and atherosclerosis is complex, a strong body of evidence suggests that its activity (cholesteryl ester transfer [CET]) is proatherogenic. We have previously shown that CET is increased in IDDM patients receiving conventional subcutaneous insulin treatment and normalized when systemic insulin levels are lowered with intraperitoneal insulin delivery (IP). Since CET has been found by many observers to also be accelerated in NIDDM, we sought to determine whether the same salutary effect could be achieved in insulin-requiring NIDDM men before and 7 months after randomization to an intensive treatment regimen (Rx) of either IP (n = 9) or multiple daily insulin injections (MDI; n = 13). HbA1c improved to the same degree in both groups (MDI group: 9.4 +/- 1.1% pre-Rx vs. 7.2 +/- 0.7% post-Rx [P < 0.001]; IP group: 9.2 +/- 1.3% pre-Rx vs. 7.1 +/- 0.5% post-Rx [P < 0.001]). Compared with pre-Rx levels, plasma triglycerides were not significantly changed by either treatment (MDI group: 136 +/- 80 mg/dl pre-Rx vs. 139 +/- 87 mg/dl post-Rx; IP group: 157 +/- 63 mg/dl pre-Rx vs. 188 +/- 89 mg/dl post-Rx), though an upward trend followed IP. Before randomization, CET estimated with both mass and isotopic assays was greater in the NIDDM subjects than in nondiabetic control subjects (P < 0.001). With improved glycemic control, CE mass transfer declined in both groups, but only reached normal levels in the IP group (MDI group at 2 h: 49.0 +/- 13.7 [mean +/- SD] pg pre-Rx vs. 29.5 +/- 15.3 microg post-Rx [-39.7%, P < 0.01]; IP group at 2 h: 40.8 +/- 23.3 microg pre-Rx vs. 10.9 +/- 6.5 microg post-Rx [-73.2%, P < 0.05]) and remained abnormally increased (P < 0.005) in the subjects receiving MDI. Total lipolytic activity after intensive treatment was unchanged from pretreatment levels, which were similar to those of the reference group. Although directional changes in lipoprotein lipase (LpL) and hepatic triglyceride lipase (HTGL) similar to those found in IDDM after MDI and IP were observed, they were not statistically significant. Thus, while improved glycemic control alone achieved by either MDI or IP reduced the pathological increase in CET in these insulin-treated NIDDM men, normalization was only achieved in those treated with IP. Despite near-normal HbA1c levels, CET remained abnormally increased in NIDDM patients treated rigorously with conventional subcutaneous insulin delivery.

Effects of omega-3 fish oils on plasma lipids, lipoprotein composition, and postheparin lipoprotein lipase in women with IDDM.

Because the apparent reduction in cardiovascular risk noted in nondiabetic populations that ingest diets rich in marine lipids containing omega-3 fatty acids is believed to result in part from their capacity to modify the composition and physicochemical behavior of lipoproteins, we sought to determine whether dietary supplementation with marine lipids might favorably affect lipoprotein composition in insulin-dependent diabetes mellitus (IDDM). Eight normolipidemic IDDM women (mean +/- SD age 29.8 +/- 4.7 yr) were studied before and 3 mo after receiving a marine-lipid concentrate (Super-EPA) containing 6 g omega-3 fatty acids and a total of 12 mg of cholesterol daily. Weight, insulin requirements, and glycosylated hemoglobin remained stable. After treatment, mean +/- SD plasma triglyceride (TG) levels fell (before, 81.7 +/- 22 mg/dl; after, 69.19 +/- 17; P less than 0.025). High-density lipoprotein2 (HDL2) cholesterol (before, 10.98 +/- 5.45 mg/dl; after, 18.43 +/- 7.93; P less than 0.01), its major apolipoprotein A-I (apoAI), and the major phospholipids (sphingomyelin and lecithin) all rose significantly. ApoB and plasma and low-density lipoprotein cholesterol levels and HDL3 composition were unchanged. Postheparin hepatic and lipoprotein lipase activities were unaffected by marine lipids. These data indicate that women with IDDM experience apparently beneficial effects on TG and HDL2 from dietary supplementation with omega-3 fatty acids administered in a low-cholesterol-containing oil without adversely affecting overall diabetes management. If these changes in lipoprotein concentration and composition prove to have antiatherogenic consequences and are free of long-term toxicity, these agents may have a role in the therapy of IDDM patients.

Probucol normalizes cholesteryl ester transfer in IDDM.

OBJECTIVE: Cholesteryl ester transfer (CET) is pathologically increased in insulin-dependent diabetes mellitus (IDDM), and the resulting enrichment of the apolipoprotein B-containing lipoproteins with cholesteryl ester (CE) is believed to increase their atherogenicity. Because we have shown previously that treatment with the lipid-modifying antioxidant probucol normalizes CET in nondiabetic patients with hypercholesterolemia, we sought to determine whether the same beneficial effects could be achieved in IDDM. RESEARCH DESIGN AND METHODS: CET was measured by both mass and isotopic assay in eight normolipidemic (triglyceride, 102; cholesterol, 192; high-density lipoprotein [HDL] cholesterol, 45 mg/dl) IDDM patients (fructosamine 495 +/- 146 mumol/l; normal 174-286) before and after 2 months of treatment with probucol (1.0 g/day). RESULTS: Before treatment, CET was accelerated abnormally (P < 0.001). As expected, probucol decreased plasma (-13%; P < 0.025) and HDL2 cholesterol levels (-52%; P < 0.025) and the concentration of lipoprotein A-I particles (P < 0.025). In conjunction with these changes, CET fell dramatically in all subjects (mass assay: -94%; isotopic assay: -22%, P < 0.001) with no change in the mass of cholesteryl ester transfer protein (CETP) (pretreatment 2.91 +/- 0.97 vs. posttreatment 3.21 +/- 1.03 micrograms/ml). Glycemic control, however, improved significantly (fructosamine 409 +/- 85 mumol/l, P < 0.025). CONCLUSIONS: Because it is believed that accelerated CET promotes the formation of apolipoprotein B-containing lipoproteins enriched with atherogenic CE, the capacity of probucol to reverse this functional abnormality without adversely affecting glycemic control suggests that it has a place in the therapy of IDDM.

Differing effects of pancreas-kidney transplantation with systemic versus portal venous drainage on cholesteryl ester transfer in IDDM subjects.

OBJECTIVE: Cholesteryl ester transfer (CET) is accelerated in patients with IDDM treated with conventional (subcutaneous) insulin therapy (CIT) and a number of other disorders associated with premature cardiovascular disease. We have shown that in IDDM this disturbance is closely linked to iatrogenic hyperinsulinemia (HI), because it was reversed when insulin was administered by the intraportal (i.p.) route. In this study, we sought to determine whether HI after successful pancreas-kidney transplantation (PKT) has the same adverse effect on CET. RESEARCH DESIGN AND METHODS: CET was measured by both mass and isotopic assays and compared in two groups of euglycemic non-insulin-requiring IDDM PKT patients with either systemically draining allografts and persistent HI or grafts with portal vein anastomoses that were normoinsulinemic (PK-P). A third group of eight nondiabetic kidney transplant (KT) patients receiving the same immunosuppressive drugs served as control subjects. RESULTS: CET in pancreas-kidney transplantation subjects with systemic venous drainage (PK-S) was increased (P < 0.001) to the same level we have reported previously in IDDM patients receiving CIT and was significantly higher (P < 0.001) than in those subjects with PK-P. CET in the PK-P group did not differ from that of the KT control patients. CONCLUSIONS: CET is affected by variations in systemic insulin levels in pancreas transplant patients with allografts that have differing venous drainage. Because high systemic insulin levels are linked to the activation of (ET, euglycemic HI IDDM pancreas allograft recipients may continue to be at high risk for macrovascular complications.

Effects of estropipate treatment on plasma lipids and lipoprotein lipid composition in postmenopausal women.

It is generally believed that the cardioprotective benefit of long term treatment of postmenopausal women with estrogen results in part from its capacity to increase high density lipoprotein (HDL) and lower low density lipoprotein (LDL) concentrations. The extent to which the various estrogens employed in replacement treatment affect the composition of lipoproteins, however, is not known. For this reason, we have examined the impact of one such preparation, the synthetic estrone estropipate (1.25 mg/day), on lipoprotein levels and composition in six postmenopausal women. After 6 months of treatment, whole plasma triglyceride (pretreatment, 135 +/- 63; posttreatment, 143 +/- 56 mg/dL), cholesterol (pretreatment, 232 +/- 14; posttreatment, 216 +/- 29 mg/dL), and HDL-C (pretreatment, 57.8 +/- 14.8; posttreatment, 55.6 +/- 13.2) were unchanged. However, plasma free (unesterified) cholesterol (FC) fell (pretreatment, 73.4 +/- 6.2; posttreatment, 53.7 +/- 9.3 mg/dL; P less than 0.05) and lecithin (L) rose significantly (pretreatment, 2.12 +/- 0.29; posttreatment, 2.47 +/- 0.34 mumol/mL; P less than 0.01). The consequence of these changes was a significant decline in the plasma FC/L ratio (pretreatment, 0.91 +/- 0.17; posttreatment, 0.68 +/- 0.12; P less than 0.01) to levels observed in healthy menstruating women. The calculated lipoprotein particle size was unchanged in very low density lipoproteins and increased significantly (P less than 0.05) in LDL after estropipate therapy. Since qualitatively altered lipoproteins enriched in FC and an increased FC/L ratio in plasma are both associated with increased coronary risk, the improvement noted in these parameters after estropipate therapy indicates that its use may be beneficial despite the lack of change in whole plasma lipids.

Influence of low estrogen-containing oral contraceptives on lipoprotein phospholipid composition and mononuclear cell membrane fluidity.

Since the effects of the new low estrogen-containing oral contraceptive (OC) preparations on lipoprotein phospholipid (PL) composition are unknown, we studied 3 groups of 10 young women before and after 6 months of use of 3 commonly prescribed agents containing almost identical amounts of ethinyl estradiol (0.30-0.35 microgram) and differing progestogens, and correlated these changes with their estrogen to progestin (E/P) ratio. The directional changes in both plasma neutral lipid and PL concentrations tended to correlate with the E/P ratio, with plasma HDL-cholesterol (HDL-C) falling slightly and the low density lipoprotein-cholesterol (LDL-C)/HDL-C ratio increasing in the women taking the OC with the lowest E/P ratio; in contrast, plasma HDL-C increased and the LDL-C/HDL-C ratio fell in those receiving the preparation with the highest E/P ratio. In HDL, the ratio of the 2 principal PL, sphingomyelin and lecithin, an index of lipid fluidity, tended to increase, suggesting that the surface of this lipoprotein class had become more rigid. This change was most apparent in women receiving the agent in which the progestin was predominant; in women receiving the preparations with the higher E/P ratios the sphingomyelin/lecithin ratio actually declined. The membrane fluidity of mononuclear cells obtained from five women taking an OC with a relatively high E/P ratio, however, was significantly increased (P less than 0.007) compared to that in normal women. These findings demonstrate that, even with substantial reductions in their estrogen content, the use of these newer OC is associated with quantitative and qualitative changes in lipoprotein PL composition that parallel their E/P balance and are associated with altered fluidity of mononuclear cell membranes.

The effects of hypothyroidism and replacement therapy on cholesteryl ester transfer.

To characterize further the impact of thyroid dysfunction on the transport of cholesterol in plasma, we studied plasma lipids and cholesteryl ester transfer (CET) in 10 hypothyroid women before and 3 months after thyroid replacement therapy. CET, estimated as the net mass transfer of CE from HDL to the apolipoprotein B-containing lipoproteins (very low density and low density lipoproteins) was significantly decreased at 4 h (P < 0.05) and 6 h (P < 0.001) when the patients were hypothyroid (T4, 2.01 +/- 1.4; TSH, 55.5 +/- 39.9 microIU/mL) and increased to normal levels after hormone replacement and restoration of eumetabolism. Plasma lipid levels in the hypothyroid state closely resembled those in a female reference group, although total plasma cholesterol fell significantly [pretreatment, 218 +/- 36 vs. posttreatment, 192 +/- 49 (P < 0.025); control, 218 +/- 28 mg/dL (mean +/- SD)] after treatment. Concentrations of cholesteryl ester transfer protein (CETP) were unchanged (pretreatment, 2.35 +/- 0.83 vs. posttreatment, 2.30 +/- 1.19 mg/dL). The results of recombination studies using different lipoprotein fractions suggest that decreases in CET during hypothyroidism may be secondary to acceptor lipoprotein (low density and very low density lipoprotein) changes in the hypothyroid state and not to changes in the concentration of CETP itself.

Effects of tamoxifen treatment on plasma lipids and lipoprotein lipid composition.

Concern has been raised that long term treatment with the antiestrogen tamoxifen might predispose women to the rapid development of cardiovascular disease. Since estrogen-induced changes in plasma lipids confer protection to females from coronary heart disease, we have examined the impact of tamoxifen on lipoprotein levels and composition on eight posmenopausal women. After 3 months of tamoxifen treatment (10 mg, twice daily), no significant changes were observed in either whole plasma triglyceride (pre-Rx, 137 +/- 59; post-Rx, 157 +/- 110 mg/dL) or cholesterol (pre-Rx, 193 +/- 23; post-Rx, 204 +/- 14 mg/dL); plasma free (unesterified) cholesterol (FC), however, fell significantly (pre-Rx, 66.5 +/- 6.5; post-Rx, 59.6 +/- 4.6 mg/dL; P less than 0.05). Since plasma lecithin (L) was unchanged, the FC/L ratio declined significantly to levels observed in healthy menstruating women (pre-Rx, 95 +/- 0.16; post-Rx, 0.74 +/- 0.12 molar ratio; P less than 0.025). In low density lipoprotein (LDL), the concentrations of cholesterol and FC and the FC/L ratio all fell significantly (P less than 0.025, P less than 0.05, and P less than 0.025, respectively). Despite a tendency for high density lipoprotein2 cholesterol (HDL2-C) to increase (pre-Rx, 9.7 +/- 3.6; post-Rx, and 14.4 +/- 13.3 mg/dL; P less than 0.4) and phosphoinositol to fall, there were few clear-cut alterations in either HDL2 or HDL3 surface or core lipid composition. The combination of reduced HDL3 lysolecithin (P less than 0.025) associated with a posttreatment trend toward increased triglyceride/cholesterol esters ratios in both HDL subfractions are findings consistent with tamoxifen-induced inhibition of hepatic lipase. These changes in lipoprotein composition together with the fall in LDL cholesterol and increase in sex hormone-binding globulin (P less than 0.005) indicate that tamoxifen acts as an estrogen agonist on the liver. Since elevated LDL cholesterol levels and qualitatively altered lipoproteins enriched in FC are both associated with increased coronary risk, the improvement noted in these parameters after tamoxifen should allay to some degree anxiety about its use with regard to cardiovascular risk.

Incorporation of dietary n-3 fatty acids into molecular species of phosphatidyl choline and cholesteryl ester in normal human plasma.

To understand the differences in the antiatherogenic actions of eicosapentaenoic acid (EPA, 20:5) and docosahexaenoic acid (DHA, 22:6), we determined their incorporation into molecular species of phosphatidylcholine (PC) and cholesteryl ester (CE) after feeding 12 g marine lipid concentrate/d to six normolipidemic males for 28 d. The time course of incorporation of EPA into plasma PC and CE showed a precursor-product relationship. In contrast, the DHA concentration of CE was markedly lower than that in PC, and the EPA-DHA ratio was 2-6-fold higher in CE than in PC at all time intervals. Three PC species--16:0-20:5, 16:0-22:6, and 18:0-20:5--increased, whereas 18:1-18:2, 18:0-18:2, and 16:0-20:3 decreased. In vitro formation of CE species in plasma by lecithin-cholesterol acyltransferase (LCAT) showed an increased formation of 20:5 CE but not 22:6 CE, indicating that DHA is a poor substrate for LCAT. These results demonstrate a differential incorporation of EPA and DHA into plasma lipids, which may be related to the differences in their biological effects.

Chronic renal failure and atherogenesis--Serum factors stimulate the proliferation of human arterial smooth muscle cells.

To assess the role of serum factors in the genesis of accelerated vascular disease in chronic renal failure, human arterial smooth muscle cells (SMC) and dermal fibroblasts were grown in culture and the effects of serum from chronic dialysis patients on cell proliferation was studied. Exposure to serum from these renal failure patients was associated with significantly greater growth of both SMC and fibroblasts than that observed with control serum. A portion of this mitogenic effect appears to be related to the presence of a factor(s) which is heat stable, dialysable, and is contained in the lipoprotein deficient fraction of plasma of density greater than 1.25 g/dl. These findings suggest that circulating substances which stimulate the proliferation of SMC may contribute to accelerated cardiovascular disease in patients undergoing chronic dialysis treatment.