Kinetics and mechanism of transfer of reduced and carboxymethylated apolipoprotein A-II between phospholipid vesicles

The transfer of 14C-labeled, reduced and carboxymethylated human apolipoprotein A-II (RCM-AII) between small unilamellar vesicles (SUV) has been investigated. Ion-exchange chromatography was used for rapid separation of negatively charged egg phosphatidylcholine (PC)/dicetyl phosphate donor SUV containing bound 14C-labeled RCM-AII from neutral egg PC acceptor SUV present in 10-fold molar excess. The kinetics of 14C-labeled RCM-AII transfer in incubations of up to 12 h at 37 degrees C are consistent with the existence of fast, slow, and apparently "nontransferrable" pools of SUV-associated apolipoprotein; the transfers from these pools occur on the time scales of seconds or less, hours, and days/weeks, respectively. For donor SUV (0.15 mg of phospholipid/mL reaction mixture) containing about 15 RCM-AII molecules per vesicle, the sizes of the fast, slow, and nontransferrable pools are 13, 69, and 18%, respectively. The transfer of RCM-AII from the slow kinetic pool follows first-order kinetics, and the half-time (t 1/2) is about 3 h. The different kinetic pools of SUV-associated RCM-AII probably reflect apoprotein in different conformations of the SUV surface. Increasing the number of RCM-AII per donor SUV enlarges the size of the fast pool and increases the t 1/2 of transfer from the slow pool. In contrast, raising the incubation temperature reduces the t 1/2 of slow transfer. The t 1/2 of RCM-AII transfer from the slow kinetic pool is inversely proportional to the acceptor/donor SUV ratio which suggests that the transfer of apoprotein molecules in this kinetic pool is mediated by SUV collisions.(ABSTRACT TRUNCATED AT 250 WORDS)     

Transbilayer diffusion of phospholipids: dependence on headgroup structure and acyl chain length

The kinetics and thermodynamics of the transmembrane movement (flip-flop) of fluorescent analogs of phosphatidic acid (PA), phosphatidylglycerol (PG), phosphatidylcholine (PC), and phosphatidylethanolamine (PE) were investigated to determine the contributions of headgroup composition and acyl chain length to phospholipid flip-flop. The phospholipid derivatives containing n-octanoic, n-decanoic or n-dodecanoic acid in the sn-1 position and 9-(1-pyrenyl)nonanoic acid in the sn-2 position were incorporated at 3 mol% into sonicated single-bilayer vesicles of 1-palmitoyl-2-oleoyl-sn-glycerol-3-phosphocholine (POPC). The kinetics of diffusion of the pyrene-labeled phospholipids from the outer and inner monolayers of the host vesicles to a large pool of POPC acceptor vesicles were monitored by the time-dependent decrease of pyrene excimer fluorescence. The observed kinetics of transfer were biexponential, with a fast component due to the spontaneous transfer of pyrenyl phospholipids in the outer monolayer of labeled vesicles and a slower component due to diffusion of pyrenyl phospholipid from the inner monolayer of the same vesicles. Intervesicular transfer rates decreased approx. 8-fold for every two carbons added to the first acyl chain. Correspondingly, the free energy of activation for transfer increased approx. 1.3 kcal/mol. With the exception of PE, the intervesicular transfer rates for the different headgroups within a homologous series were nearly the same, with the PC derivative being the fastest. Transfer rates for the PE derivatives were 5-to 7-fold slower than the rates observed for PC. Phospholipid flip-flop, in contrast, was strongly dependent on headgroup composition with a smaller dependence on acyl chain length. At pH 7.4, flip-flop rates increased in the order PC less than PG less than PA less than PE, where the rates for PE were at least 10-times greater than those of the homologous PC derivative. Activation energies for flip-flop were large, and ranged from 38 kcal/mol for the longest acyl chain derivative of PC to 25 kcal/mol for the PE derivatives. Titration of the PA headgroup at pH 4.0 produced an approx. 500-fold increase in the flip-flop rate of PA, while the activation energy decreased 10 kcal/mol. Increasing acyl chain length reduced phospholipid flip-flop rates, with the greatest change observed for the PC analogs, which exhibited an approx. 2-fold decrease in flip-flop rate for every two methylene carbons added to the acyl chain at the sn-1 position.(ABSTRACT TRUNCATED AT 400 WORDS)     

Effects of apolipoprotein structure on the kinetics of apolipoprotein transfer between phospholipid vesicles

The kinetics and mechanism of transfer of 14C-labeled human apolipoproteins A-I, A-II and C-III1 between small unilamellar vesicles (SUV) have been investigated. Ion exchange chromatography was used for rapid separation of negatively charged egg phosphatidylcholine (PC)/dicetyl phosphate donor SUV containing bound 14C-labeled apoprotein from neutral egg PC acceptor SUV present in 10-fold molar excess. The transfer kinetics of these apolipoproteins at 37 degrees C are consistent with the existence of fast, slow and apparently 'nontransferrable' pools of SUV-associated lipoprotein: the transfers from these pools occur on timescales of seconds (or less), minutes/hours and days/weeks, respectively. For donor SUV containing about 15 apoprotein molecules per vesicle and at a donor SUV concentration of 0.15 mg phospholipid/ml incubation mixture, the sizes of the fast kinetic pools for apolipoproteins A-I, A-II and C-III1 associated with donor SUV are 2, 10 and 11%, respectively. The sizes of the slow kinetic pools for these apolipoproteins are 16, 71 and 50%, respectively. The transfer of the various apolipoproteins from the slow kinetic pool follows first order kinetics and the half-time (t1/2) values are in the order: apo C-III1 less than apo A-I. Increasing the number of apoprotein molecules per donor SUV enlarges the size of the fast pool and increases the t1/2 of slow transfer. The differences in the kinetics of apolipoprotein transfer between SUV are consequences of the variations in the primary and secondary structures of the apolipoprotein molecules. The slow transfer of apoprotein molecules is mediated by collisions between donor and acceptor SUV; the rate is dependent on the apoprotein molecular weight with larger molecules transferring more slowly from donor SUV containing the same lipid/protein molar ratio. The hydrophobicity of the apoprotein molecule is also significant with less hydrophobic molecules transferring more rapidly. Further understanding of the differences in the kinetics of transfer of these apolipoproteins will require more knowledge of their secondary and tertiary structures.     

Kinetics of transfer of alpha-tocopherol between model and native plasma lipoproteins

The rate of spontaneous transfer of alpha-tocopherol, cholesterol and beta-carotene between model and native lipoproteins was measured to determine the mechanism and kinetics of equilibration of these lipids in plasma. Cholesterol and alpha-tocopherol transfer from apolipoprotein A-I/1-palmityl-2- oleoylphosphatidylcholine ( POPC ) recombinants to bovine brain ganglioside/ POPC single bilage vesicles with half-times of approximately 20 min and 70 min, respectively. Under identical conditions, there is no significant transfer of beta-carotene even after an 18-h incubation period. alpha-Tocopherol transfers from apolipoprotein A-II/dimyristoylphosphatidylcholine recombinants with a half-time of 40 min and an activation energy of 17.2 kcal/mol. Incubation of high-density lipoproteins containing alpha-[3H]tocopherol with low-density lipoproteins or very-low-density lipoproteins results in the equilibration of the labelled lipid between the lipoprotein classes in 1 h. A comparison of the rates of transfer indicates that alpha-tocopherol equilibrates 2-3-times more slowly than cholesterol but on a time scale much shorter than the lifetime of lipoproteins in the circulation. Thus, the distribution of alpha-tocopherol is not kinetically controlled but determined thermodynamically by the partitioning between the total amount of lipid in each compartment. The spontaneous transfer of beta-carotene is too slow for this equilibration to occur.     

Phase behavior of cerebroside and its fractions with phosphatidylcholines: calorimetric studies

Bovine brain cerebroside and its kerasin (beta-D-galactosyl-N-acyl-D-sphingosine) and phrenosin (beta-D-galactosyl-N-(2-D-hydroxyacyl)-D-sphingosine) fractions were mixed with diacylphosphatidylcholines (PCs) to form fully hydrated lamellar phases. These mixtures were examined by differential scanning calorimetry, and phase diagrams for cerebroside/diacylPC mixtures were constructed from the data. Cerebroside was found to be miscible with egg PC at low mole fractions X of cerebroside; the mixture behaves non-ideally for X greater than 0.25. The non-ideal behavior appears to be a superposition of separate interactions of kerasin and phrenosin with egg PC. Strikingly, phrenosin mixes nearly ideally with egg PC. Kerasin mixed with egg PC yields a peritectic phase diagram. Cerebroside and phrenosin were found to be immiscible with dimyristoylphosphatidylcholine (DMPC) in the gel state in low proportions. Both stable and metastable gel phases of kerasin were detected in different endotherms of kerasin/PC mixtures. Kerasin in the stable and metastable gel states exhibits discontinuous and continuous ranges of miscibility, respectively, with DMPC. The stable gel phase of kerasin does not segregate in natural cerebroside. Natural kerasin was found to act isomorphic to semi-synthetic (natural configuration) D-kerasins but not completely to synthetic DL-kerasins of single acyl chain lengths.     

Cholesterol distribution and movement in the Mycoplasma gallisepticum cell membrane.

The time course and extent of transfer of [14C]-cholesterol from resting Mycoplasma gallisepticum cells or membrane preparations to high-density lipoproteins were studied. More than 90% of the total cholesterol in isolated, unsealed membrane preparations was exchanged in a single kinetic process. In intact cells, however, cholesterol exists in two different environments. Cholesterol in one environment, representing approximately 50% of the total unesterified cholesterol, is readily exchanged with the cholesterol of high-density lipoproteins, with a half-time of about 4 h at 37 degrees C. The rate of exchange of [14C]cholesterol from the other environment was exceedingly slow, with a half-time of about 18 days. The fraction of the total cholesterol in the readily exchangeable cholesterol pool in intact cells increased somewhat upon aging of the culture. Electron spin resonance spectra of nitroxide-labeled stearic acids incorporated into membranes of M. gallisepticum cells indicated increased rigidity at the late exponential phase of growth. These results suggest that cholesterol is present in approximately equal concentrations on both surfaces of the M. gallisepticum membrane and that in resting cells the rate of movement of cholesterol molecules from the inner to outer halves of the lipid bilayer is exceedingly slow or nonexistent.     

Molecular conformations of cerebrosides in bilayers determined by Raman spectroscopy.

Vibrational Raman spectra of the solid and gel phases of bovine brain cerebrosides and the component fractions, kerasin and phrenosin, provide conformational information for these glycosphingolipids in bilayer systems. The carbon-carbon stretching mode profiles (1,150-1,000 cm-1) indicate that at 22 degrees C the alkyl chains assume an almost all-trans arrangement. These spectral data, combined with those from the C-H stretching region (3,050-2,800 cm-1), show that phrenosin forms the most highly ordered polycrystalline solid and kerasin the most ordered gel phase. The conformation of the unsaturated, 24-carbon acyl chains is monitored independently by a skeletal stretching mode at 1,112 cm-1. The alkyl chains in the kerasin and phrenosin gels are sufficiently extended to allow interdigitation of the 24-carbon acyl chains across the midplane of the bilayer. The amide I vibrational mode occurs at a lower frequency in solid phrenosin than kerasin, a shift consistent with stronger hydrogen bounding. This band is broadened and shifted to higher frequencies, however, in the phrenosin gel phase. In both the solid and gel phases natural cerebroside exhibits a composite amide I mode. The disruptive effects on cerebroside chain packing and headgroup orientation arising from mixing with dimyristoyl phosphatidylcholine are examined. Vibrational data for cerebroside are also compared to those for ceramide, sphingosine, and distearoyl phosphatidylcholine structures. Spectral interpretations are discussed in terms of calorimetric and X-ray structural data.     

Lipid transfer between phosphatidylcholine vesicles and human erythrocytes: exponential decrease in rate with increasing acyl chain length

The rate of phospholipid transfer from sonicated phospholipid vesicles to human erythrocytes has been studied as a function of membrane concentration and lipid acyl chain composition. Phospholipid transfer exhibits saturable first-order kinetics with respect to both cell and vesicle membrane concentrations. This kinetic behavior is consistent either with transfer during transient contact between cell and vesicle surfaces (but only if the fraction of the cell surface susceptible to such interaction is small) or with transfer of monomers through the aqueous phase. The acyl chain composition of the transferred phospholipid affects the transfer kinetics profoundly; for homologous saturated phosphatidylcholines, the rate of transfer decreases exponentially with increasing acyl chain length. This behavior is consistent with passage of phospholipid monomers through a polar phase, which might be the bulk aqueous phase( as in the monomer transfer model) or the hydrated head-group regions of a cell-vesicle complex (transient collision model). Collisional transfer also predicts that intercell transfer of phospholipids should be slow compared to cell-vesicle transfer, as surface charge and steric effects should prevent close apposition of donor and acceptor membranes. This is not found; dilauroylphosphatidylcholine transfers rapidly between red cells. Thus, the observed relationship between acyl chain length and intermembrane phospholipid transfer rates likely reflects the energetics of monomer transfer through the aqueous phase.     

Symmetrical distribution and rapid transbilayer movement of cholesterol in Mycoplasma gallisepticum membranes

The exchange of cholesterol between [¹⁴C]cholesterol-labeled Mycoplasma gallisepticum cells and an excess of sonicated egg phosphatidylcholine/cholesterol vesicles (molar ratio of 0.9) was measured. More than 90% of the radioactive cholesterol underwent transfer from intact cells to the vesicles. The kinetics of the transfer was biphasic. About 50% of the radioactive cholesterol was exchanged with a half-time of about 4 h. The residual was exchanged at a slower rate with a half-time of about 9 h at 37°C. Bovine serum albumin had a pronounced effect in enhancing both the fast and slow rates of cholesterol exchange, but did not affect the pool sizes significantly. The half-time for equilibration of the two pools in the presence of 2% albumin, calculated using a reversible two-pool method of analysis, was 6.2 h. The effect of albumin was also obtained with isolated membrane preparations and with cells treated with growth inhibitors, suggesting that this effect is independent of albumin preservation of cell viability. The rate enhancement of albumin was concentration dependent with maximal effects observed with 2%, where the rates of exchange of both the rapidly and slowly exchanging pools were twice as fast. The mechanism by which albumin may affect the exchange rates is discussed.     

Transmembrane movement and distribution of cholesterol in the membrane of vesicular stomatitis virus.

The transmembrane movement and distribution of cholesterol in the vesicular stomatitis virus membrane were studied by following the depletion of cholesterol from virions to interacting phospholipid vesicles and by exchange of radiolabeled cholesterol between virions and phospholipid-cholesterol vesicles. The kinetics of the cholesterol exchange or depletion reactions revealed the presence of two exponential rates: a rapid rate, dependent on the vesicle to virus ratio, and a slower rate, independent of the vesicle to virus ratio. The kinetics of cholesterol movement could be best interpreted by a model of the virion membrane considered as a two pool system in which approximately 30% of the cholesterol resides in the outer monolayer and approximately 70% in the inner monolayer. The half-time for equilibration of the two pools was calculated to be 4--6 h and was assumed to represent the time required for transmembrane movement of cholesterol across the bilayer. The initial rate of transfer of cholesterol from virus into vesicles increased when vesicle phospholipids contained more unsaturated and shorter chain fatty acids. Furthermore, the transfer of cholesterol appeared to occur by a collisional mechanism requiring membrane-membrane contact. Interaction with lipid vesicles did not significantly affect the integrity of the virion membrane as assessed by the relative inaccessibility of internal proteins to lactoperoxidase-catalyzed iodination and by the small loss of [3H]amino acid labeled protein from the virus.     

Kinetics and mechanism of exchange of apolipoprotein C-III molecules from very low density lipoprotein particles

Transfer of apolipoprotein (apo) molecules between lipoprotein particles is an important factor in modulating the metabolism of the particles. Although the phenomenon is well established, the kinetics and molecular mechanism of passive apo exchange/transfer have not been defined in detail. In this study, the kinetic parameters governing the movement of radiolabeled apoC molecules from human very low density lipoprotein (VLDL) to high density lipoprotein (HDL3) particles were measured using a manganese phosphate precipitation assay to rapidly separate the two types of lipoprotein particles. In the case of VLDL labeled with human [14C]apoCIII1, a large fraction of the apoCIII1 transfers to HDL3 within 1 minute of mixing the two lipoproteins at either 4 degrees or 37 degrees C. As the diameter of the VLDL donor particles is decreased from 42-59 to 23-25 nm, the size of this rapidly transferring apoCIII1 pool increases from about 50% to 85%. There is also a pool of apoCIII1 existing on the donor VLDL particles that transfers more slowly. This slow transfer follows a monoexponential rate equation; for 35-40 nm donor VLDL particles the pool size is approximately 20% and the t1/2 is approximately 3 h. The flux of apoCIII molecules between VLDL and HDL3 is bidirectional and all of the apoCIII seems to be available for exchange so that equilibrium is attained. It is likely that the two kinetic pools of apoCIII are related to conformational variations of individual apo molecules on the surface of VLDL particles. The rate of slow transfer of apoCIII1 from donor VLDL (35-40 nm) to acceptor HDL3 is unaffected by an increase in the acceptor to donor ratio, indicating that the transfer is not dependent on collisions between donor and acceptor particles. Consistent with this, apoCIII1 molecules can transfer from donor VLDL to acceptor HDL3 particles across a 50 kDa molecular mass cutoff semipermeable membrane separating the lipoprotein particles. These results indicate that apoC molecules transfer between VLDL and HDL3 particles by an aqueous diffusion mechanism.     

CHANGES IN BRAIN ALCOHOL DEHYDROGENASE ACTIVITY DURING CHRONIC ETHANOL INGESTION AND WITHDRAWAL

Abstract— Chronic ethanol ingestion in rats leads to a slow rise in brain alcohol dehydrogenase activity which levels off after 2 weeks at approximately twice the initial activity. The half-time of the rise is approximately 8 days. Abrupt withdrawal of the ethanol is followed by a rapid decline of the brain alcohol dehydrogenase activity to the normal level with a half-time of approximately 15 h. The difference in time constants between the rise in enzyme activity during ethanol-feeding and its decline following withdrawal suggests that the increased enzyme activity is at least in part the result of a reduced rate constant of enzyme degradation in the presence of ethanol. The effect of ethanol on brain alcohol dehydrogenase activity is not altered by supplementation of the diet with carbohydrate or vitamins. The effect is seen only in the cerebral hemispheres and not in the brain-stem. Acquisition of tolerance to ethanol during chronic ethanol ingestion and its extinction following withdrawal follow almost the same time courses as the changes in brain alcohol dehydrogenase activity.     

Enzymatic sulphation of some galactose- containing sphingolipids in developing rat brain

Abstract— Cerebroside sulphotransferase has been found to catalyze the transfer of sulphate from 3′-phosphoadenosine-5′-phosphosulfate (PAPS) to both the α-hydroxy fatty acid galactosylceramides and the nonhydroxy fatty acid galactosylceramides. The sulphotransferase has a higher affinity for the α-hydroxy fatty acid galactosylceramides than for the nonhydroxy fatty acid galactosylceramides and will also use lactosylceramide as an acceptor for the transfer of sulphate from PAPS. A second sulphotransferase, PAPS: psychosine sulphotransferase, is also present in the developing rat brain and will catalyse the transfer of sulphate from PAPS to galactosylsphingosine and lactosylsphingosine. The sulphate moiety was determined to be on the galactose and most likely on the 3′ position giving a proposed structure of: 3-O-SO₄-galactosylsphingosine. The possible role of this later pathway in the synthesis of sulphogaiactosylceramide remains to be elucidated.     

Stages of the bilayer-micelle transition in the system phosphatidylcholine-C12E8 as studied by deuterium- and phosphorous-NMR, light scattering, and calorimetry.

The perturbation of phospholipid bilayer membranes by a nonionic detergent, octaethyleneglycol mono-n-dodecylether (C12E8), was investigated by 2H- and 31P-NMR, static and dynamic light scattering, and differential scanning calorimetry. Preequilibrated mixtures of the saturated phospholipids 1,2-dipalmitoyl-sn-glycero-3-phosphorylcholine (DPPC), 1,2-dimyristoyl-sn-glycero-3-phosphorylcholine (DMPC), and 1,2-dilauroyl-sn-glycero-3-phosphorylcholine (DLPC) with the detergent were studied over a broad temperature range including the temperature of the main thermotropic phase transition of the pure phospholipids. Above this temperature, at a phospholipid/detergent molar ratio 2:1, the membranes were oriented in the magnetic field. Cooling of the mixtures below the thermotropic phase transition temperatures of the pure phospholipids led to micelle formation. In mixtures of DPPC and DMPC with C12E8, a narrow calorimetric signal at the onset temperature of the solubilization suggested that micelle formation was related to the disorder-order transition in the phospholipid acyl chains. The particle size changed from 150 nm to approximately 7 nm over the temperature range of the bilayer-micelle transition. The spontaneous orientation of the membranes at high temperatures enabled the direct determination of segmental order parameters from the deuterium spectra. The order parameter profiles of the phospholipid acyl chains could be attributed to slow fluctuations of the whole membrane and to detergent-induced local perturbations of the bilayer order. The packing constraints in the mixed bilayers that eventually lead to bilayer solubilization were reflected by the order parameters of the interfacial phospholipid acyl chain segments and of the phospholipid headgroup. These results are interpreted in terms of the changing average shape of the component molecules. Considering the decreasing cross sectional areas in the acyl chain region and the increasing hydration of the detergent headgroups, the bilayer-micelle transition is the result of an imbalance in the chain and headgroup repulsion. A neutral or pivotal plane can be defined on the basis of the temperature dependence of the interfacial quadrupolar splittings.     

A potential role for sterol carrier protein-2 in cholesterol transfer to mitochondria

Mitochondrial cholesterol oxidation rapidly depletes cholesterol from the relatively cholesterol-poor mitochondrial membranes. However, almost nothing is known regarding potential mechanism(s) whereby the mitochondrial cholesterol pool is restored. Since most exogenous cholesterol enters the cell via the lysosomal pathway, this could be a source of mitochondrial cholesterol. In the present study, an in vitro fluorescent sterol transfer assay was used to examine whether the lysosomal membrane could be a putative cholesterol donor to mitochondria. First, it was shown that spontaneous sterol transfer from lysosomal to mitochondrial membranes was very slow (initial rate, 0.316 +/- 0.032 pmol/min). This was due, in part, to the fact that 90% of the lysosomal membrane sterol was not exchangeable, while the remaining 10% also had a relatively long half-time of exchange t(1/2) = 202 +/- 19 min. Second, the intracellular sterol carrier protein-2 (SCP-2) and its precursor (pro-SCP-2) increased the initial rate of sterol transfer from the lysosomal to mitochondrial membrane by 5.2- and 2.0-fold, respectively, but not in the reverse direction. The enhanced sterol transfer was due to a 3.5-fold increase in exchangeable sterol pool size and to induction of a very rapidly (t(1/2) = 4.1 +/- 0.6 min) exchangeable sterol pool. Confocal fluorescence imaging and indirect immunocytochemistry colocalized significant amounts of SCP-2 with the mitochondrial marker enzyme cytochrome oxidase in transfected L-cells overexpressing SCP-2. In summary, SCP-2 and pro-SCP-2 both stimulated molecular sterol transfer from lysosomal to mitochondrial membranes, suggesting a potential mechanism for replenishing mitochondrial cholesterol pools depleted by cholesterol oxidation.     

Multiple geminate ligand recombinations in human hemoglobin

The geminate ligand recombination reactions of photolyzed carbonmonoxyhemoglobin were studied in a nanosecond double-excitation-pulse time-resolved absorption experiment. The second laser pulse, delayed by intervals as long as 400 ns after the first, provided a measure of the geminate kinetics by rephotolyzing ligands that have recombined during the delay time. The peak-to-trough magnitude of the Soret band photolysis difference spectrum measured as a function of the delay between excitation pulses showed that the room temperature kinetics of geminate recombination in adult human hemoglobin are best described by two exponential processes, with lifetimes of 36 and 162 ns. The relative amounts of bimolecular recombination to T- and R-state hemoglobins and the temperature dependence of the submicrosecond kinetics between 283 and 323 K are also consistent with biexponential kinetics for geminate recombination. These results are discussed in terms of two models: geminate recombination kinetics modulated by concurrent protein relaxation and heterogeneous kinetics arising from alpha and beta chain differences.     

Measurement and prediction of the rates of spontaneous transfer of phospholipids between plasma lipoproteins

The purpose of this report is to develop a correlation between the hydrophobicity of a phospholipid as measured by reversed-phase high-performance liquid chromatography and its rate of spontaneous transfer and to use this correlation to predict the rate of transfer of any homologous lipid from any lipoprotein. We have studied the mechanism of transfer of a series of fluorescent or radiolabeled phospholipids among natural and reassembled serum lipoproteins. Fluorescent phosphatidylcholines included those with 9-(1-pyrenyl)nonanoic acid in the sn-2 position and lauric, myristic, palmitic, stearic, oleic or linoleic acid at sn-1. The radioactive phosphatidylcholines contained [3H]oleic acid in the sn-2 position and lauric, myristic, or palmitic acid at sn-1. The kinetics of transfer of the pyrene-labeled lipid were followed by changes in the excimer fluorescence, and that of the radioactive lipids by separation of the donor (lipid-apolipoprotein recombinant) from the acceptor (single bilayer vesicles) on a column of Sephacryl S-200. The retention time of each lipid was measured by high-performance hydrophobic chromatography through a Waters radially compressed C18 column eluted with 75% isopropanol and 25% triethylammonium phosphate (0.15 M). A linear relationship was observed between the rate-constant of transfer and the retention time which suggest that the rate of desorption of phosphatidylcholines from lipoproteins and vesicles is controlled predominately by the hydrophobic effect. For a homologous series of lipids, the rate of transfer can be predicted from retention times obtained from hydrophobic chromatography. The kinetics of transfer of 1-lauroyl-2-[9-(1-pyrenyl)nonanoyl] phosphatidylcholine between isolated human serum lipoproteins exhibits a linear correlation between the transfer half-time and the size of the donor lipoproteins. As a consequence, transfer from very-low-density lipoprotein is 10-times slower than that observed from high-density lipoproteins. The observed correlations between phospholipid transfer rates and both the Stokes radius of the donor and the retention time of the phospholipid on a hydrophobic column permit one to calculate the rate of transfer of homologous molecules between lipid-protein complexes. The results predict that the spontaneous transfer of phospholipids between plasma lipoproteins would be too slow to be a physiologically important phenomena.     

Probing red cell membrane cholesterol movement with cyclodextrin

We probed the kinetics with which cholesterol moves across the human red cell bilayer and exits the membrane using methyl-beta-cyclodextrin as an acceptor. The fractional rate of cholesterol transfer (% s(-1)) was unprecedented, the half-time at 37 degrees C being ~1 s. The kinetics observed under typical conditions were independent of donor concentration and directly proportional to acceptor concentration. The rate of exit of membrane cholesterol fell hyperbolically to zero with increasing dilution. The energy of activation for cholesterol transfer was the same at high and low dilution; namely, 27-28 Kcal/mol. This behavior is not consistent with an exit pathway involving desorption followed by aqueous diffusion to acceptors nor with a simple one-step collision mechanism. Rather, it is that predicted for an activation-collision mechanism in which the reversible partial projection of cholesterol molecules out of the bilayer precedes their collisional capture by cyclodextrin. Because the entire membrane pool was transferred in a single first-order process under all conditions, we infer that the transbilayer diffusion (flip-flop) of cholesterol must have proceeded faster than its exit, i.e., with a half-time of <1 s at 37 degrees C.     

Fatty acid metabolism in sn-glycerol-3-phosphate acyltransferase (plsB) mutants.

Fatty acid metabolism was examined in Escherichia coli plsB mutants that were conditionally defective in sn-glycerol-3-phosphate acyltransferase activity. The fatty acids synthesized when acyl transfer to glycerol-3-phosphate was inhibited were preferentially transferred to phosphatidylglycerol. A comparison of the ratio of phospholipid species labeled with 32Pi and [3H]acetate in the presence and absence of glycerol-3-phosphate indicated that [3H]acetate incorporation into phosphatidylglycerol was due to fatty acid turnover. A significant contraction of the acetyl coenzyme A pool after glycerol-3-phosphate starvation of the plsB mutant precluded the quantitative assessment of the rate of phosphatidylglycerol fatty acid labeling. Fatty acid chain length in membrane phospholipids increased as the concentration of the glycerol-3-phosphate growth supplement decreased, and after the abrupt cessation of phospholipid biosynthesis abnormally long chain fatty acids were excreted into the growth medium. These data suggest that the acyl moieties of phosphatidylglycerol are metabolically active, and that competition between fatty acid elongation and acyl transfer is an important determinant of the acyl chain length in membrane phospholipids.     

Two gel states of cerebrosides. Calorimetric and Raman spectroscopic evidence.

The enthalpy of the gel-to-liquid crystalline phase transition for kerasin (15.8 kcal/mol) is found to be markedly higher than that for phrenosin and unfractionated bovine brain cerebrosides (about 7 kcal/mol). Evidence for a higher degree of order in the hydrocarbon chains and a different configuration in the polar region of kerasin is supplied by Raman spectroscopic parameters for these gel phases. The high transition enthalpy for kerasin is ascribed to a lesser accommodation of gauche conformers in the hydrocarbon chains just below the transition temperature. The thermodynamic behavior of these cerebroside fractions, including hysteresis in kerasin gels, is compared to that previously reported for sphingomyelins.