Calorimetric study of microsomal membrane

Coconut oil induces short-term changes in lipid composition and enzyme activity of chick hepatic mitochondria

We studied the short-term effects of a 20% coconut oil supplementation to the chick diet on lipid composition of liver and hepatic mitochondria, and changes that occurred in mitochondrial-associated enzymes as a result of this diet. No significant differences were observed in the lipid contents of liver when young chicks were fed the experimental diet, whereas hepatic mitochondria rapidly changed in response to this diet. Total cholesterol significantly increased in mitochondria at 24 hours of coconut oil diet feeding and decreased when dietary treatment was prolonged for 5 to 14 days. Changes in total mitochondrial phospholipids showed an inverse profile. A significant decrease in phosphatidylethanolamine and an increase in sphingomyelin were found at 24 hours. The cholesterol/phospholipid molar ratio significantly and rapidly (24 hours) increased in mitochondria from treated animals. Cytochrome oxidase activity drastically increased after 24 hours of experimental diet feeding and lowered to the control values when dietary manipulation was prolonged for 5 to 14 days. ATPase activity showed an inverse profile. Changes in cytochrome oxidase activity were parallel to changes in the cholesterol/phospholipid molar ratio, whereas changes in ATPase activity showed an inverse correlation with changes in this molar ratio. To our knowledge, this is one of the first reports on the very rapid response (24 hours) of mitochondrial lipid composition and function to saturated fat feeding.

Arrhenius plots of membrane-bound enzymes of mitochondria and microsomes in the brain cortex of developing and old rats

To study changes in lipid-protein-interaction and fluidity in mitochondrial and microsomal membranes of brain during development and aging, the Arrhenius plots of marker enzymes for mitochondrial inner and outer membranes as well as those of microsomal membranes were compared at different ages. The enzymes were, beta-hydroxybutyrate dehydrogenase (BDH) for the inner mitochondrial membrane and rotenone-insensitive-NADH-cytochrome c reductase (Mit-NADH-CytR) for the outer membrane; also antimycin-insensitive-NADH-cytochrome c reductase (Micr-NADH-CytR) and NADPH-cytochrome c reductase for the microsomal membranes. The ages studied were, 1, 5, 30, 60 days postnatal and 2 years. In the microsomes, the plots of NADH-cytochrome c reductase were found to be biphasic at all ages except in the newborn where no break temperature was observed. The activation energy of this enzyme in the physiological range of temperature was found to be high in the newborn, declining with brain maturation. The plot of this enzyme in the old rat showed no difference when compared with the mature. The specific activity of the enzyme, however, was markedly reduced in the old brain microsomal fraction. In contrast, comparison of the plots of microsomal NADPH-cytochrome c reductase at the various ages revealed no break temperature and very similar energies of activation. For the inner mitochondrial membrane beta-hydroxybutyrate dehydrogenase, the Arrhenius plots were generally biphasic at all ages studied with a break temperature of about 20 degrees C. However, the newborn plot was only barely biphasic showing a high energy of activation in the physiological range of temperature. In contrast, for the outer membrane NADH-cytochrome c reductase, the newborn plots were definitely biphasic, exhibiting low activation energy above the breaks. There was also a significant decline in the break temperature with brain development. No significant differences in the plots of this enzyme were found between the old and mature brain cortex. It is concluded that the enzymes of inner and outer mitochondrial membranes as well as the oxidative enzymes of microsomal membranes may show differential patterns of change in lipid-protein-interaction during development and aging, the changes being more marked in development than in aging.

Effect of dietary cholesterol on microsomal membrane composition, dynamics and kinetic properties of UDPglucuronyl transferase

The effect of cholesterol administration in vivo on the lipid composition, dynamic properties of the microsomal membrane of guinea pig livers and the kinetic properties of UDPglucuronyl transferase were studied. Cholesterol administration in the diet evoked an increase of microsomal cholesterol, but no significant changes in the fatty-acid composition of total lipids or of each phospholipid class. Instead, the phosphatidylethanolamine, phosphatidylcholine molar ratio of the membrane was markedly decreased from 0.57 to 0.38. This decline was not enough to counterbalance the overall 'ordering' effect of cholesterol and consequently, the fluorescence anisotropy of the membranes labeled with 1,6-diphenylhexatriene was increased. The lateral diffusion evaluated by measuring the pyrene excimer formation was decreased by the cholesterol incorporation. These physical changes were associated with changes in the kinetic properties of UDPglucuronyl transferase: Vmax increased, while the Km of the different steps of the reaction decreased in the modified microsomes. Furthermore, a shift of the non-michaelian kinetics to michaelian, equivalent to a decrease of a negative homotropic effect and apparent cooperativity of UDPglucuronic acid was observed since the Hill coefficient changed, approaching 1. A non-michaelian kinetics of this enzyme is an indication of boundary lipids in the gel phase and a shift to michaelian, a change of the surrounding lipids to a liquid-crystalline structure. In consequence, our results suggest that cholesterol incorporation in the microsomal membrane while producing a condensing effect of bulk lipids would produce an opposite effect on the UDPglucuronyl transferase boundary lipids.

Breaks in arrhenius plots of reactions involving membrane-bound and solubilized sialyltransferases, due to temperature dependence of kinetic parameters

Temperature dependence of asialomucin-sialyltransferase (CMP-N-acetylneuraminate:D-galactosyl-glycoprotein N-acetylneuraminyltransferase, EC 2.4.99.1) activity is investigated. Discontinuities in Arrhenius plots are observed, whether the enzyme is membrane-associated or solubilized. These discontinuities cannot be firmly correlated with the phase-transition temperatures of either endogenous or exogenous phospholipids. Arrhenius plots of the kinetic parameters also exhibit sharp discontinuities, so that it is concluded that a significant change in Km and Vmax values occurs with varying temperature. Our results suggest that the biphasic behavior of Arrhenius plots may be attributed to the temperature dependence of the kinetic parameters for both membrane-associated and solubilized sialyltransferase activities.

Temperature dependence of membranous and solubilized sialyltransferase activities in the presence of 1-palmitoyl-sn-glycero-3-phosphorylcholine and fatty acids

The temperature dependence of sialyltransferase (CMP-N-acetylneuraminate: D-galactosyl-glycoprotein N-acetyl-neuraminyltrasferase, EC 2.4.99.1) inhibition is described when 1-palmitoyl-sn-glycero-3-phosphorylcholine, or a saturated fatty acid (lauric, myristic or palmitic acid) or an equimolar mixture of the two components are added. Lysophospholipid and fatty acids have no appreciable effect on the optimal temperature for sialyltransferase activity. In the presence of lysophospholipid, the membranous sialyltransferase activity is decreased for all the temperature range tested. In contrast, the solubilized sialyltransferase activity is decreased for temperatures exceeding 29 degrees C. In the presence of saturated fatty acids, the membranous activity is decreased above a chain-length dependent temperature: 22 degrees, 25 degrees and 30 degrees C for lauric, myristic and palmitic acids, respectively. In contrast, the solubilized activity remains unchanged. In the presence of equimolar mixtures of lysophospholipid and fatty acid, the membranous activity is decreased above the same critical temperature as that described for fatty acids added alone. In contrast, the solubilized activity is decreased above 29 degrees C. From these observations, it is suggested that lysophospholipid inhibits the solubilized enzyme when the temperature exceeds the critical micellar temperature of this lipid. The fatty acids inhibit the microsomal enzyme probably by incorporating into the membrane. It is also suggested that equimolar mixtures of lysophospholipid and fatty acid give rise to molecular analogs of 1,2-dipalmitoyl-sn-glycero-3-phosphorylcholine.

Thermotropic change in phospholipid packing in microsomal membranes sensed by phospholipase A2

A reversible, temperature-dependent change in phospholipid packing occurring between 0 degree C and 12 degrees C has been identified in microsomal membranes by the use of phospholipase A2 from Crotalus atrox. It manifests itself as a drastic increase in susceptibility to the phospholipase and depends on non-lipid (presumably protein) membrane components. It is suggested that this change could underlie the change in transmembrane mobility of phospholipids which occurs in the same temperature range.

The effect of dietary lipids on the thermotropic behaviour of rat liver and heart mitochondrial membrane lipids

Diets supplemented with relatively high levels of either saturated fatty acids derived from sheep kidney fat (sheep kidney fat diet) or unsaturated fatty acids derived from sunflower seed oil (sunflower seed oil diet) were fed to rats for a period of 16 weeks and changes in the thermotropic behaviour of liver and heart mitochondrial lipids were determined by differential scanning calorimetry (DSC). The diets induced similar changes in the fatty acid composition in both liver and heart mitochondrial lipids, the major change being the omega 6 to omega 3 unsaturated fatty acid ratio, which was elevated in mitochondria from animals on the sunflower seed oil diet and lowered with the mitochondria from the sheep kidney fat dietary animals. When examined by DSC, aqueous buffer dispersions of liver and heart mitochondrial lipids exhibited two independent, reversible phase transitions and in some instances a third highly unstable transition. The dietary lipid treatments had their major effect of the temperature at which the lower phase transition occurred, there being an inverse relationship between the transition temperature and the omega 6 to omega 3 unsaturated fatty acid ratio. No significant effect was observed for the temperature of the higher phase transition. These results indicate that certain domains of mitochondrial lipids, probably containing some relatively higher melting-point lipids, independently undergo formation of the solidus or gel phase and this phenomenon is not greatly influenced by the lipid composition of the mitochondrial membranes. Conversely, other domains, representing the bulk of the membrane lipids and which probably contain the relatively lower melting point lipids, undergo solidus phase formation at temperatures which reflect changes in the membrane lipid composition which are in turn, a reflection of the nature of the dietary lipid intake. These lipid phase transitions do not appear to correlate directly with those events considered responsible for the altered Arrhenius kinetics of various mitochondrial membrane-associated enzymes.

Hepatic microsomal NADPH-cytochrome P-450 reductase from little skate, Raja erinacea. Comparison of thermolability and other molecular properties with a mammalian enzyme

Components of little skate (an elasmobranch) and rabbit hepatic microsomal cytochrome P-450 dependent monooxygenase systems were examined for differences which might explain the decreasing xenobiotic-metabolizing activity of little skate microsomes assayed at temperatures above 30 degrees C. The proportion of saturated fatty acids in microsomal lipids and the habitat temperature are both lower in skate as compared to rabbit, which is consistent with the known adaptive pattern. The more thermolabile enzyme of the skate system in microsomal preparations is NADPH-cytochrome P-450 reductase. The optimal assay temperature for purified skate reductase (30 degrees C) is 10 degrees C lower than that for the purified rabbit reductase. The purified skate reductase differs from rabbit reductase in monomeric molecular weight, in peptides produced by partial proteolysis, in immunochemical properties, but not in flavin content.

Guanidinium- and temperature-induced conformational changes in glucose dehydrogenase

The conformational changes in glucose dehydrogenase are studied as a function of temperature and guanidinium chloride (GdmCl) concentration. The data were analyzed assuming a two-conformer model which gave similar results using either circular dichroism or enzyme activity. The free energy of denaturation was 0.94 kcal/mol from specific activity and 1.64 kcal/mol from circular dichroism measurements. The mid-point of the denaturation curve was 0.65 or 0.63 M GdmCl, as determined by specific activity or circular dichroism, respectively. The transition temperature, 6.4 degrees C, is close to that of a microsomal membrane phase change, a result that is consistent with the fact that glucose dehydrogenase contains lipid materials when isolated with a non-ionic detergent such as Triton X-114. As the temperature increased, the amount of beta-pleated sheet increased, and the alpha-helical content decreased, suggested that glucose dehydrogenase contains a stable core of beta-pleated sheet.

Lipid domains in plasma membranes from rat liver

The existence of fluid and solid lipid domains in isolated rat-liver plasma membranes was evaluated using the fluorescent fatty acids trans-parinaric and cis-parinaric acid as probe molecules for solid and fluid membrane areas, respectively. The fluorescence probe 1,6-diphenyl-1,3,5-hexatriene indicated that a phase transition was present in the liver plasma membrane between 18 degrees C and 30 degrees C. At intermediate temperatures, cis-parinaric acid, which partitioned approximately equally into fluid and solid lipid areas, detected two lipid domains: the mole fractions of fluid and solid lipid domains at 24 degrees C were 0.32 and 0.68 while the mole fractions of cis-parinaric acid in each domain were 0.34 and 0.66, respectively. The dissociation constant, aqueous to membrane lipid partition coefficient, and bound to free ratio for trans-parinaric acid were 7.0 +/- 0.7 microM, 4.0 +/- 0.6 x 10(6), and 83:17, respectively. The affinity of the membrane for cis-parinaric acid was twofold lower than for trans-parinaric acid. The trans-parinaric acid partitioned preferentially into solid lipid, Ksp/f = 3.30, while the cis-parinaric acid partitioned equally between fluid and solid phases Ksp/f = 0.92. Thus, the data demonstrate the coexistence of fluid and solid domains in rat liver plasma membranes.

The use of differential scanning calorimetry and differential thermal analysis in studies of model and biological membranes

Differential scanning calorimetry (DSC), and to a lesser extent differential thermal analysis (DTA), are powerful yet relatively rapid and inexpensive thermodynamic techniques for studying the thermotropic phase behavior of lipids in model and biological membranes, without the introduction of exogenous probe molecules. In this review the principles as well as the scope and limitations of DSC and DTA are discussed first. The application of these techniques to the study of the thermotropic phase behavior of aqueous dispersions of various single synthetic phospholipids are then summarized, and the effects of cholesterol, free fatty acids, lysophospholipids, drugs, anesthetics and proteins on the gel to liquid-crystalline phase transitions exhibited by these model systems are discussed. The phase mixing properties of model membranes consisting of mixtures of two or more synthetic or natural phospholipids are considered next. Finally, the thermotropic phase behavior of prokaryotic plasma membranes and of the plasma, microsomal and mitochondrial membranes of eukaryotic cells are reviewed, and the applications of DSC and DTA to study the thermal behavior of specific membrane proteins, as well as the physical properties of the membrane lipid phase, are summarized.

Effect of temperature and ions on the microviscosity of bilayers from natural phospholipid mixtures

The thermotropic behaviour of two natural extracts of phospholipids differing by their content in saturated fatty acids and acidic polar heads were studied with the aid of the fluorescent probes : perylene and diphenylhexatriene. The main results are that the bilayers are in a fluid state in the range from -- 5 to 50 degrees C, in the presence as well as in the absence of calcium, and that misleading conclusions may be derived with the perylene probe. The effects of pH and Ca2+/H+ interaction on the microviscosity were studied. In the absence of calcium pH increase fluidizes the bilayers ; when Ca2+ is present the microviscosity is constant over a large pH range. Calculations using the Gouy-Chapman theory suggest that the pH-induced microviscosity changes are controlled by the content in phosphadidylserine, diphosphatidylglycerol and phosphatidic acid of the extracts. Similarly the rigidifying effect of Ca2+ seems to be related to the content in these phospholipids. These hypothesis were confirmed by modifying the content of the extracts in these phospholipids. It is proposed that the behaviour of the two studied extracts results from the high insaturation of their fatty acids, which is typical of biological membranes.

Factors affecting the intracellular generation of free radicals from quinones

Isolated hepatocytes do not liberate appreciable amounts of superoxide into the external medium. Simple quinones stimulate the release of superoxide up to 15 nmol/min/10(6) hepatocytes. Superoxide release stimulated by a variety of simple quinones and more complex antitumor quinones was maximal at a quinone one-electron reduction potential of -70 mV. This was qualitatively similar to the pattern of superoxide formation seen with NADH-cytochrome b5 reductase and NADH: ubiquinone oxidoreductase. Superoxide production by NADPH-cytochrome P-450 reductase was maximal at a quinone single-electron reduction potential at -200 mV. Phenobarbital pretreatment had no effect on superoxide formation by hepatocytes suggesting that NADPH-cytochrome P-450 reductase activity is not rate limiting for quinone stimulated superoxide formation. Sulfonated stilbenes, specific inhibitors of anion exchange, had no effect on the release of superoxide by hepatocytes suggesting that superoxide is not transported through anion channels in the plasma membrane. Pretreatment of hepatocytes with 10(-5) M diethyldithiocarbamate produced over a two fold increase in the release of superoxide.

Changes in hepatic microsomal membrane fluidity with age

There are changes in the mixed function oxidase enzymatic activities of rat hepatic microsomal membranes with age. However, the protein components of the mixed function oxidase system do not appear to change with age. The purpose of this study was to detect possible changes in the fluidity of the lipid component of the microsomal membrane with age. Hepatic microsomes were isolated by differential centrifugation from uninduced, male CFN rats aged 3, 12 and 26 mo. The microsomal membrane fluidity was measured using electron paramagnetic resonance after incorporation of a 5-nitroxide stearic acid spin label into the membrane. The order parameter S decreased with age from 0.586 +/- 0.003 (3 mo) to 0.581 +/- 0.002 (12 mo) to 0.569 +/- 0.003 (26 mo) at 30 degrees C. This indicated an increase in membrane fluidity with age. In membranes labeled with the 16-nitroxide stearic acid, a similar increase in membrane fluidity with age was observed. The order parameter of microsomal membranes from 3 and 26 mo rats was measured over the temperature range 10 degrees to 31 degrees C in steps of 0.9 degrees C. A plot of the log of S versus the reciprocal temperature revealed a phase transition at 24 degrees C in membranes from 26 mo rats, but no phase transition was observed in 3 mo old rats in this temperature range. The change in fluidity of the hepatic microsomal membrane with age may account for some of the observed changes in membrane-bound mixed function oxidase activities with age.

The effects of various incubation temperatures, particulate isolation, and possible role of calmodulin on the activity of the base exchange enzymes of rat brain

The involvement of calmodulin in the choline, ethanolamine, and serine exchange activities of rat brain microsomes was investigated. Calmodulin stimulated choline exchange activity to a greater extent than ethanolamine and serine exchange activities. The three base exchange activities were inhibited by antipsychotic drugs believed to prevent calmodulin interaction, but not by calmodulin-binding protein. The solutions employed for tissue homogenization and subsequent isolation of microsomes greatly influenced the base exchange activities. The process of resuspending isolated microsomes and recentrifugation, or "washing," produced major losses of detectable activity. The base exchange enzyme activities were maximal at 45 degrees, and Arrhenius plots revealed a common transition temperature of 31 degrees. The activation energies for the base exchange reactions decreased at temperatures above the observed transition temperature. Kinetic data, Km and Vmax, for the base exchange activities at 27, 37, and 45 degrees are presented.

Effect of fatty acids on physical properties of microsomes from isolated perfused rat liver

A computer-centered spectrofluorimeter was used to examine the physicochemical properties of hepatic microsomes and microsomal lipids obtained from isolated rat livers perfused with medium containing palmitate or oleate. The fatty acid composition and degree of unsaturation of the liver microsomal lipids reflected that the fatty acid present in the perfusate. The absorption corrected fluorescence, relative fluorescence efficiency, polarization, and fluorescence anisotropy of several fluorescent probe molecules were measured to determine if their different microenvironments may be altered by the type of fatty acid infused. The probe molecules beta--parinaric acid and 1,6-diphenyl-1,3,5-hexatriene had higher values for each of these parameters when incorporated into microsomes obtained from livers perfused with a medium containing palmitate than with oleate. The same parameters measured for cholesta-5,7,9(11)-trien-3 beta-ol and N-phenyl-1-naphthylamine were not altered. These differences appeared to be primarily due to alterations in microviscosity of the probe microenvironments since the rotational correlation time of 1,6-diphenyl-1,3,5-hexatriene was 25% lower in the microsomes from livers perfused with oleate as compared to livers perfused with palmitate. Thermal discontinuities in Arrhenius plots were noted in the intact microsomes but not in the isolated microsomal lipids with the fluorescence probe molecule beta-parinaric acid. Break points occurred at 10 degrees C and 26 degrees C for microsomes from livers perfused with palmitate and at 12 degrees C and 17 degrees C for microsomes from livers perfused with oleate containing medium. These results suggest that the physicochemical properties of liver microsomes were determined in part by the fatty acid in the perfusate.

Induction of gel-phase lipid in plasma membrane of chick intestinal cells after coccidial infection

When chickens are infected with the coccidial parasite Eimeria necatrix, the plasma membrane of intestinal cells harbouring second-generation schizonts becomes refractory to mechanical shearing, hypotonic shock and ultrasonication. Plasma membrane from these infected cells was isolated to high purity as judged by enriched levels of ouabain-sensitive (Na+ + K+)-stimulated Mg2-dependent ATPase activity and sialic acid content, the lack of detectable cytochrome oxidase and glucose-6-phosphatase activities and electron microscopic analysis of the final preparation. Wide-angle X-ray diffraction patterns recorded from the isolated membranes revealed that during the later stages of parasite maturation the host cell plasma membrane acquires increasing proportions of gel-phase lipid. By contrast, purified membrane from isolated parasites is in a liquid-crystalline state. The transition temperature of host cell plasmalemma at 100 h postinfection is 61 degrees C, about 20 degrees C above physiological temperature. By contrast, liposomes of plasma membranes from infected cells undergo a thermal transition at about 28 degrees C. The accumulation of gel-phase lipid in the host cell plasma membrane is not attributable either to an increase in the constituent ratio of saturated to unsaturated fatty acids or to a significant change in the cholesterol to phospholipid ratio. During the late stages of infection, the cells become stainable with trypan blue which suggests that the acquisition of crystalline phase lipid disrupts the permeability of the host cell plasmalemma.

Calf spleen NAD glycohydrolase. Comparison of the catalytic properties of the membrane-bound and the hydrosoluble forms of the enzyme

The catalytic properties of membrane-bound calf spleen NAD glycohydrolase were studied in comparison with previous data obtained with a solubilized hydrosoluble form of the enzyme. When the hydrolysis of NAD catalyzed by membrane-bound NAD glycohydrolase was studied at pH values below 7.5, only insignificant interference by other NAD-hydrolyzing enzymes was detected, and no proton-diffusional inhibition was observed. The kinetics could, therefore, be followed using a titrimetric assay for NAD glycohydrolase activity. The effect of pH, ionic strength on the kinetic parameters, and shifts in binding constants for several ligands of the membrane-bound enzyme indicate that the NAD glycohydrolase activity is influenced by an electrostatic potential due to negative charges (polyelectrolyte effect). No significant changes in kinetic mechanism could be found between both NAD glycohydrolase forms. The association of the enzyme with the membrane results in a remarkably increased thermal stability, in changes in binding properties of the active site and in the emergence of new inhibitor binding sites; e.g. adenosine 3':5'-monophosphate (cyclic AMP) and adenosine, which do not inhibit the hydrosoluble form of NAD glycohydrolase, are good inhibitors (respectively competitive and mixed) of the membrane-bound enzyme. These data (i.e. allotopic changes) probably can be ascribed to enzyme conformational changes induced and stabilized by interaction with membrane constituents.

Evidence for isotropic motion of phospholipids in liver microsomal membranes. A 31P NMR study

1. The motional properties of phospholipids in bovine and rat liver microsomes and aqueous dispersions of the extracted lipids have been investigated employing 31 P NMR techniques. 2. The 31P NMR spectra obtained from the microsomes indicate that a considerable portion of the constituent phospholipids experience isotropic motion on the NMR timescale (10(-5) s). This is in strong contrast to the spectra obtained from aqueous dispersions of the extracted lipids, which display the characteristic lineshape associated with liquid crystalline phospholipids in (large) bilayer structures, which experience restricted anisotropic motion. 3. Evidence is presented which strongly suggests that the isotropic motion of microsomal phospholipids does not arise from tumbling of the microsomal vesicles or from lateral diffusion of phospholipids around these vesicles. 4. These results are discussed in terms of possible transitory formation of intramembrane non-bilayer lipid configurations, with which the bulk (bilayer) phospholipids are in rapid exchange.

Phosphatidylcholine mobility in liver microsomal membranes

Purified phosphatidylcholine exchange protein from bovine liver was used to exchange rat liver microsomal phosphatidylcholine for egg phosphatidylcholine. It was found that at 25 and 37 degrees C rat liver microsomal phosphatidylcholine was completely and rapidly available for replacement by egg phosphatidylcholine. In contrast, phosphatidylcholine in vesicles prepared from total microsomal lipids could only be exchanged for about 60%. At 8 and 0 degrees C complex exchange kinetics were observed for phosphatidylcholine in rat liver microsomes. The exchange process had neither effect on the permeability of the microsomal membrane to mannose 6-phosphate, nor on the permeability of the phosphatidylcholine vesicles to neodymium (III) cations. Purified phospholipase A2 from Naja naja could hydrolyze some 55-60% of microsomal phosphatidylcholine at 0 degrees C, but 70-80% at 37 degrees C. Microsomal phosphatidylcholine, remaining after phospholipase treatment at 37 degrees C, could be exchanged for egg phosphatidylcholine at 37 degrees C, but at a slower rate than with intact microsomes. Microsomal phosphatidylcholine remaining after phospholipase treatment at 0 and 37 degrees C had a lower content of arachidonic acid than the original phosphatidylcholine. These results are discussed with respect to the localization and transmembrane movement of phosphatidylcholine in liver microsomes.