Structure-specific inhibition of cholesteryl ester transfer protein by azaphilones.

The effect of thirteen different fungal azaphilones, which have a common 6-iso-chromane-like ring, was tested on cholesteryl ester transfer protein (CETP) activity in vitro. Chaetoviridin B showed the most potent inhibitory activity with an IC50 value of < 6.2 microM, followed by sclerotiorin with an IC50 value of 19.4 microM. Rotiorin, chaetoviridin A and rubrorotiorin had moderate inhibitory activity (IC50 ; 30 approximately 40 microM), but others showed very weak or no inhibitory activity. The relationship between the structures and their inhibitory activity indicated that the presence of an electrophilic ketone(s) and/or enone(s) at both C-6 and C-8 positions in the isochromane-like ring is essential for eliciting CETP inhibitory activity. The transfer activity of both CE and TG was inhibited by sclerotiorin to approximately the same extent (IC50: 14.4 and 10.3 microM, respectively). A model of the reaction suggested that sclerotiorin reacts with a primary amine of amino acids such as lysine in the protein to form a covalent bond.

Low level quantification of cholesteryl ester transfer protein in plasma subfractions and cell culture media by monoclonal antibody-based immunoassay.

Sensitive immunoradiometric (IRMA) and ELISA assays for cholesteryl ester transfer protein (CETP) have been developed using two different monoclonal antibodies (MAbs). The MAbs were prepared against human plasma CETP and demonstrated specificity by their inhibition of cholesteryl ester transfer activity and by immunoblots of crude plasma fractions and whole media from transfected CHO cells. For these sandwich-type assays, one MAb, 2F8, is used for capture, and the second MAb, 2E7, is iodinated (IRMA) or conjugated with alkaline phosphatase (ELISA) and used for detection. Both assays are linear and provide sensitivities much greater than previously reported. The IRMA allows for the accurate quantification of CETP in the range of 0.5-20 ng/assay (5-200 ng/ml), the ELISA 0.05-5 ng/assay (0.5-50 ng/ml). Using the IRMA, the mean plasma CETP concentration in 44 normolipidemic individuals was determined to be 2.10 +/- 0.36 micrograms/ml; 2.05 +/- 0.33 for males (n = 25) and 2.16 +/- 0.40 for females (n = 19). Values ranged from 1.28 to 2.97 micrograms/ml and CETP mass correlated well with cholesteryl ester transfer activity (r = 0.913, n = 23). The distribution of CETP in human plasma was examined both by gel permeation fast protein liquid chromatography (FPLC) and by native gel electrophoresis. For FPLC using agarose resins, a low ionic strength, isotonic buffer system resulted in near total recoveries of CETP, and demonstrated a peak for CETP mass centered at molecular masses of 150 to 180 kilodaltons, larger than that for free monomeric CETP. Native acrylamide gel electrophoresis of plasma from six individuals, followed by 2F8/2E7 sandwich immunoblotting, showed CETP migrating within a size range of 170-220 kilodaltons. This result is consistent with suggestions that plasma CETP is associated with small-sized HDL. Agarose gel electrophoresis showed plasma CETP, as well as purified recombinant CETP, to be prebeta migrating. For determining the concentration of CETP in the media of cultured HepG2 cells, advantage was taken of the high sensitivity of the ELISA. CETP levels were found to increase linearly over the 100-h culture period, reaching 8.0 +/- 0.4 ng/ml (18.0 +/- 1.3 ng/mg cell protein). These sensitive, direct immunoassays for CETP mass should be valuable aids for examining the behavior of CETP in plasma and other complex systems, as well as for studying the synthesis and secretion of CETP by different cells and tissues.

Inhibition of cholesteryl ester transfer protein activity in hamsters alters HDL lipid composition.

We investigated the role of cholesteryl ester transfer protein (CETP) in hamsters by using a monoclonal antibody (MAb) that inhibited hamster CETP activity. MAbs were prepared against partially purified human CETP and screened for inhibiton of 3H-cholesteryl oleate (CE) transfer from LDL to HDL in the presence of human plasma bottom fraction (d > 1.21 g/ml). Antibody 1C4 inhibited CE transfer activity in both human plasma bottom fraction (IC50 = approximately 4 micrograms/ml) and in whole plasma from male Golden Syrian hamsters (IC50 = approximately 30 micrograms/ml). Purified MAb 1C4 was injected into chow- and cholesterol-fed hamsters, and blood was collected for analysis of plasma CETP activity and HDL lipid composition. Plasma CETP activity was inhibited by 70%-80% at all and HDL lipid composition. Plasma CETP activity was inhibited by 70%-80% at all times up to 24 h following injection of 500 micrograms MAb 1C4 (approximately 3.7 mg/kg). The amount of antibody required for 50% inhibition at 24 h post-injection was 200 micrograms (approximately 1.5 mg/kg). Inhibition of hamster CETP activity in vivo increased hamster HDL cholesterol by 33% (P < 0.0001), increased HDL-CE by 31% (P < 0.0001) and decreased HDL-triglyceride by 42% (P < 0.0001) (n = 36) as determined following isolation of HDL by ultracentrifugation. An increase in HDL cholesterol and a redistribution of cholesterol to a larger HDL particle were also observed following fast protein liquid chromatography (FPLC) gel filtration of plasma lipoproteins.(ABSTRACT TRUNCATED AT 250 WORDS)

Possible role of the Golgi apparatus in the assembly of very low density lipoprotein.

Transit of newly synthesized triacyl[3H]-glycerol through organelles of the secretory system leading to assembly into nascent very low density lipoproteins (VLDLs) or to cytoplasmic storage was investigated in chick hepatocytes. Cells in monolayer culture were pulse-labeled with [2-3H]glycerol, and after different periods of chase with unlabeled glycerol, the movement of triacyl[3H]glycerol through the endoplasmic reticulum (ER) and Golgi and the incorporation into nascent VLDL and cytoplasmic triacylglycerol-rich vesicles (TGRVs) were determined. Initially, newly synthesized triacyl[3H]glycerol is tightly associated with the ER. Movement from the ER of triacyl[3H]glycerol destined for cytoplasmic storage (as TGRVs) is extremely rapid and virtually complete within 8 min of chase. After 8 min of chase, triacyl[3H]glycerol lost from organelles of the secretory system was accounted for entirely as triacyl[3H]glycerol secreted as VLDL. Comparison of the rates of movement of triacyl[3H]glycerol, apolipoprotein B, apolipoprotein II, and apolipoprotein A-I through the ER and Golgi and of their secretion in nascent VLDL suggests that assembly of triacyglycerol with apolipoproteins occurs in the Golgi. Experiments with permeabilized hepatocytes supplemented with cytosol show that newly synthesized triacyl[3H]glycerol and [3H]phospholipid moves from the ER through the full-density range of Golgi elements and is dependent upon supplementary ATP.

Assembly of very low density lipoprotein in the hepatocyte. Differential transport of apoproteins through the secretory pathway.

The transport of the apolipoprotein (apo) constituents of hepatic very low density lipoprotein (VLDL) through the secretory pathway was investigated with estrogen-induced chick hepatocytes in primary culture. Cell monolayers were pulse-labeled with [3H]leucine and, after differing periods of chase with unlabeled leucine, were subjected to subcellular fractionation for 3H-apoprotein analysis. The first-order rate constants for transit of apoB, apoA-I, and apoII through the endoplasmic reticulum (ER) and Golgi were estimated using a three-compartment (ER, Golgi, and extracellular medium) kinetic analysis. The results indicate that apoB resides in the ER (t1/2 = 26 min) for a shorter period of time than in the Golgi (t1/2 = 43 min). For apoII, the t1/2 for transport through the ER and Golgi are 43 and 49 min, respectively. ApoA-I transits the ER at a rate (t1/2 = 6 min) much faster than apoB, apoII, and virtually all other secretory proteins. Upon reaching the Golgi, the rate of movement of apoA-I is markedly reduced (t1/2 = 28 min). Thus, in contrast to current models of protein secretion, the rate-limiting step in the secretion of VLDL apoproteins from the cell is transport through the Golgi, not the ER. Examination of the steady-state distribution of the apoproteins in the ER and Golgi support this conclusion. To characterize the intracellular transport process further, the distribution of apoproteins between the lumenal contents of the ER and Golgi and the membranes which delineate these compartments was determined after steady-state labeling with [3H]leucine. Approximately 50% of the apoB in the ER and in a dense, early Golgi fraction was membrane-associated, whereas in a less dense or late Golgi compartment, only 20% was bound to membranes. ApoII was also associated with the membranes of the ER and Golgi to a significant extent. In contrast, apoA-I was primarily localized lumenally throughout the secretory pathway. The occurrence of membrane-associated apoproteins in the Golgi, coupled with their slow rate of transit through this compartment suggests a major role for the Golgi in the assembly of the constituents of VLDL, and suggests that interaction of apoproteins (apoB) with the membranes of the Golgi is required for the maturation of VLDL.

Post-translational changes in tertiary and quaternary structure of the insulin proreceptor. Correlation with acquisition of function.

Tertiary and quaternary structural changes that occur during post-translational processing of the insulin proreceptor were examined in 3T3-L1 adipocytes. In pulse-chase experiments with [35S]methionine, labeled insulin receptor species, isolated by immuno- and insulin-affinity adsorption, were analyzed by sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis under conditions where intra- and intermolecular disulfide bonds remained intact or were cleaved by reduction. Reducing SDS-polyacrylamide gel electrophoresis confirmed that the insulin receptor is synthesized as a long-lived (t1/2 = 3 h) proreceptor precursor of 210 kDa which undergoes proteolytic cleavage and carbohydrate maturation to form the alpha- and beta-subunits of the mature receptor. The proreceptor acquires insulin binding activity through a subtle structural change (t1/2 = 45 min) detected only by an autoimmune antibody specific for an epitope of the active insulin binding site. Analysis of insulin receptor species by nonreducing SDS-polyacrylamide gel electrophoresis revealed that the proreceptor undergoes two additional structural changes not detected by reducing SDS-polyacrylamide gel electrophoresis. The proreceptor is synthesized as a monomer (M1) with an apparent molecular mass of 170 kDa that is converted by disulfide rearrangement to another monomeric form of 190-kDa apparent molecular mass (M2). N-Linked glycosylation is required for this transition, since aglycoproreceptor, synthesized in the presence of tunicamycin, does not undergo any detectable tertiary or quaternary structural changes. M2 self-associates to form a disulfide-linked proreceptor dimer (D) which is subsequently proteolytically processed, forming the mature, disulfide-linked alpha 2 beta 2 receptor tetramer. The mature receptor was distinguished from the three proreceptor species (M1, M2, and D) by its cell surface location and its ability to bind tightly to wheat germ agglutinin-agarose, indicating the presence of complex oligosaccharide chains. Subcellular fractionation indicated that both the M1 to M2 and M2 to D conversions occur in the endoplasmic reticulum. Separation of the nonreduced proreceptor species into "active" and "inactive" forms by affinity chromatography on insulin-agarose revealed that neither the transition of M1 to M2, nor of M2 to D, is correlated with the acquisition of insulin binding function. Rather, during its life-time, the M2 species acquires insulin binding activity and an epitope recognized by a binding site specific autoimmune antibody through a subtle structural change not detected by reducing or nonreducing SDS-polyacrylamide gel electrophoresis.

The bidirectional flux of cholesterol between cells and lipoproteins. Effects of phospholipid depletion of high density lipoprotein.

The bidirectional surface transfer of free cholesterol (FC) between Fu5AH rat hepatoma cells and human high density lipoprotein (HDL) was studied. Cells and HDL were prelabeled with [4-14C]FC and [7-3H]FC, respectively. Influx and efflux of FC were measured simultaneously from the appearance of 3H counts in cells and 14C counts in medium. Results were analyzed by a computerized procedure which fitted sets of kinetic data to a model assuming that cell and HDL FC populations each formed a single homogeneous pool and that together the pools formed a closed system. This analysis yielded values for the first-order rate constants of FC influx and efflux (ki and ke), from which influx and efflux of FC mass (Fi and Fe) could be calculated. With normal HDL, the uptake and release of FC tracers conformed well to the above-described model; Fi and Fe were approximately equal, suggesting an exchange of FC between cells and HDL. HDL was depleted of phospholipid (PL) by treatment with either phospholipase A2 or heparin-releasable rat hepatic lipase, followed by incubation with bovine serum albumin. PL depletion of HDL had little or no effect on ki, but reduced ke, indicating that PL-deficient HDL is a relatively poor acceptor of cell cholesterol. The reduction in ke resulted in initial Fi greater than Fe and, thus, in net uptake of FC by the cells. This result explained previous results demonstrating net uptake of FC from PL-depleted HDL. In the presence of an inhibitor of acyl coenzyme A:cholesterol acyltransferase, the steady state distribution of FC mass between cells and HDL was accurately predicted by the ratio of rate constants for FC flux. This result provided additional validation for describing FC flux in terms of first-order rate constants and homogeneous cell and HDL FC pools.

Mechanism of phencyclidine binding to the acetylcholine receptor from Torpedo electroplaque.

The mechanism of phencyclidine binding to Torpedo acetylcholine receptor-rich membranes was investigated. The rate of [3H]phencyclidine association is 10(3)- to 10(4)-fold more rapid when phencyclidine and carbamoylcholine are added simultaneously to acetylcholine receptor-rich membranes than when phencyclidine is added to membranes previously equilibrated with carbamoylcholine or membranes in the absence of carbamoylcholine. The mechanism of binding under conditions in which the slower rate was observed was studied with thermodynamic, viscosity, and kinetic experiments. Association and dissociation rates were highly dependent on temperature with activation energies of 26-30 kcal/mole. Viscosity had no effect on the association rate but increased the dissociation rate. These studies suggest that the binding is not diffusion-controlled but rather is limited by a significant energy barrier. The association rate was determined as a function of the concentration of acetylcholine receptor-rich membranes and the concentration of phencyclidine. In the presence of carbamoylcholine, the association rate was highly dependent upon the concentration of acetylcholine receptor but virtually insensitive to the concentration of phencyclidine. In the absence of carbamoylcholine, the association rate seemed to be a hyperbolic function of both the phencyclidine and the acetylcholine receptor concentration. The minimal model capable of explaining the data is a mechanism by which phencyclidine binds to two conformations of the acetylcholine receptor, one conformation having a higher affinity and constituting a lower percentage of receptors and the other having a lower affinity and constituting a higher percentage. The data are consistent with the possibility that the high-affinity conformation is the open-channel state of the acetylcholine receptor.

Interactions of narcotics an their antagonists with human serum esterase. IX: structure-activity relationships of morphinans and morphines.

The interaction of 21 morphinan and 25 morphine derivatives with human serum esterase was examined. The behavior of these drugs closely parallels that of their simpler analogues, benzomorphans, studied previously: all have affinity for an allosteric modifying site at which agonists accelerate enzyme action and, except for 6-ketodihydromorphines, all are also competitive inhibitors of the enzyme. The results suggest that both ring C of morphinans and morphines and the furan ring of morphines contribute to binding to the allosteric site; so does increase in the size of the N-alkyl group, but alkylation at C-6 or C-7 is unfavorable for such binding. A hydroxyl group at C-6 or C-14, or a methoxyl group at C-3 are also unfavorable. As with benzomorphans, we found that the natural (--)-morphine configuration is more sensitive than its optical antipode to structural influences on its affinity for the allosteric site.

Interactions of narcotics and their antagonists with human serum esterase. VIII: structure-activity relationships of benzomorphans.

N-alkylbenzomorphans follow the pattern established in our earlier papers for the interaction of opiates with human serum esterase: they have affinity for both the substrate-binding site of the enzyme and an allosteric site at which agonistic drugs accelerate enzyme action. Affinity for the substrate-binding site is a function of the hydrophobicity of the drug, while affinity to the allosteric site is increased by alkyl groups at N and at C--5 and decreased by a 2'-methoxyl group or by either a methyl or a hydroxyl group in the 9 beta position. The absolute D configuration (corresponding to that of natural morphine) appears to be more sensitive to factors modifying affinity. All but three of the benzomorphans studied are agonists; all agonists have lower affinity for the allosteric site than antagonists.