A singular state of membrane lipids at cell growth temperatures.

Cells adjust their membrane lipid composition when they adapt to grow at different temperatures. The consequences of this adjustment for membrane properties and functions are not well understood. Our report shows that the temperature dependence of the diffusion of a probe molecule in multilayers formed from total lipid extracts of E. coli has an anomalous maximum at a temperature corresponding to the growth temperature of each bacterial preparation (25, 29, and 32 degrees C). This increase in the lateral diffusion coefficient, D, is characteristic of membrane lipids in a critical state, for which large fluctuations of molecular area in the plane of the bilayer are expected. Therefore, changes in lipid composition may be due to a requirement that cells maintain their membranes in a state where molecular interactions and reaction rates are readily modulated by small, local perturbations of membrane organization.

Temperature dependence of membrane lipid composition in early blastula embryos of Lytechinus pictus: selective sorting of phospholipids into nascent plasma membranes.

Lytechinus pictus eggs were fertilized and incubated at 10, 16, and 23 degrees C until the early blastula stage of embryonic development. The phospholipid composition of the embryos and control unfertilized eggs remain identical and unchanged as incubating temperatures are varied; thus, neither incubating temperature, fertilization nor membrane assembly affect their total phospholipid composition. This result agrees with metabolic studies by others, using only a single incubation temperature, and indicates that embryonic development to the early blastula stage occurs with little, if any, de novo phospholipid biosynthesis. However, as in all poikilotherms, the phospholipid composition of the nascent plasma membranes varies with the incubation temperature. Thus, until the blastula stage of embryonic development, the lipids of these newly formed plasma membranes are derived from lipid pools within the embryo whose phospholipid composition is static. The variation of plasma membrane composition is primarily reflected in an increase in the phosphatidylethanolamine (PE): phosphatidylcholine (PC) ratio as incubating temperatures decrease; this is achieved by an exchange of PE for PC. Several mechanisms are considered for the specificity of the selective sorting and assembly of these phospholipids into the nascent plasma membranes.

Membrane instability, plasmalogen content, and Alzheimer's disease.

The normal stability of the cell membrane bilayer depends on its lipid composition being appropriate to the ambient (physiological) temperature, Tp. Membrane lipid composition may be altered by disease such that the bilayer is only stable at a new critical temperature, T*, which may differ from Tp. In Alzheimer's disease (AD) temporal cortex, a defect of lipid composition has previously been identified, namely, a decrease in the ratio of plasmalogen to nonplasmalogen ethanolamine glycerophospholipids. Furthermore, for AD temporal cortex neural membranes, T* << Tp, a finding confirmed in the present study in a larger series than previously, using a new method for obtaining T*. This inequality between T* and Tp has been proposed as a putative contributory pathogenetic mechanism leading to membrane destabilisation in AD brain. The plasmalogen deficiency could account for the change in T* in AD, as shown by experiments where T* was measured for artificial lipid mixtures simulating brain membranes with varying plasmalogen/nonplasmalogen ratios. The critical temperature was found to be very sensitive to small alterations in plasmalogen content.

Probing the critical unilamellar state of membranes.

Aqueous dispersions of membrane phospholipids comprised of multilamellar vesicles (MLVs) will spontaneously transform to a stable unilamellar structure equivalent to the membrane bilayer, but only at a critical temperature T*. Since much of the evidence for this transformation derives from equilibrium thermodynamic studies, a description of the molecular and topological events occurring as the critical unilamellar state assembles has not previously been possible. Here we report experiments that provide evidence of a spontaneous topological change from MLVs towards unilamellar vesicles (ULVs) at T*. By applying a shearing stress to vesicle suspensions we have observed a decrease of approximately 25% in the force required to cause bilayers to leak; this decrease is confined to temperatures near T*. The T* values observed agree with those previously obtained by equilibrium methods. Using the method with total lipid extracts from normal biological membranes confirms that T* equals the physiological temperature of the original membranes.

Disease and anatomic specificity of ethanolamine plasmalogen deficiency in Alzheimer's disease brain.

A significant and selective deficiency of ethanolamine plasmalogen (PPE) relative to phosphatidylethanolamine was identified in post mortem brain samples from patients with Alzheimer's disease (AD). This lipid defect showed anatomic specificity, being more marked at a site of neurodegeneration in AD brain than in a region relatively spared by the disease (mid-temporal cortex vs. cerebellum) and disease specificity for AD: it was not observed at the primary site of neurodegeneration in Huntington's disease (caudate nucleus) nor Parkinson's disease (substantia nigra). PPE deficiency parallels an inherent tendency towards membrane bilayer instability previously detected in AD brain which is necessarily due to a change in membrane lipid composition, and which may contribute to AD pathogenesis.

Critical temperature for unilamellar vesicle formation in dimyristoylphosphatidylcholine dispersions from specific heat measurements.

Using a heat conduction calorimeter with very high resolution (+/- 0.00005 J/degrees C.cm3), we have measured the specific heat CpL between 25 and 35 degrees C of dimyristoylphosphatidylcholine (DMPC) in aqueous dispersions. Previous studies of the temperature dependence of the chemical potential of DMPC in the L alpha phase (lamellar, liquid crystalline) indicated that a dispersion consisting only of unilamellar vesicles forms spontaneously at a critical temperature T* of 29.0 degrees C. Our present measurements show an anomaly in CpL between 28.70 and 29.50 degrees C: the curve for CpL versus T first decreases and then exhibits an inflection point at 28.96 degrees C before it flattens. This anomaly is attributed to the transformation from multilamellar dispersion to unilamellar vesicles at T* = 28.96 degrees C. Two independent properties of the CpL data also indicate T* is a critical point for the formation of unilamellar vesicles: (a) the time to reach equilibrium upon changing temperature increased dramatically between 28.7 and 28.96 degrees C, increasing as (T* - T)-1; at T > T* the dramatic "slowing-down" phenomenon was not observed. This slowing-down near T* is a general characteristic of critical phenomena. (b) The free energy change for the multilamellar-unilamellar transformation was obtained from the CpL-T data over this temperature interval and found to be 3.2 J/mol or 0.016 ergs/cm2 of bilayer, in agreement with other estimates of the interaction energy between neutral bilayers. We conclude with a discussion of the implications for membrane bilayer stability of these newly identified dynamic properties of the transformation.

Regional specificity of membrane instability in Alzheimer's disease brain.

We report an inherent tendency towards the destabilisation of cellular membranes in Alzheimer's disease (AD) brain. This tendency is a natural consequence of abnormal membrane lipid composition, which has previously been documented in AD. Membrane destabilisation may contribute to AD pathogenesis in its own right and may also facilitate amyloid beta-protein deposition, which is potentially neurotoxic. The instability was found to co-localise selectively with areas of neurodegeneration in AD brain, thereby possibly accounting for the focal pathology observed in this disorder.

A differential heat-conduction microcalorimeter for heat-capacity measurements of fluids.

A heat-conduction calorimeter has been developed for measuring small changes in heat capacity of milligram samples of membrane lipid dispersed in water as a function of temperature. The operation of the instrument is based on the principle that the thermal response of the sample to a short (10 s), electrically generated heat burst is a function of the diffusivity of the sample. Modeling studies of the instrument's performance have revealed that the output response after the heat burst is a function of only the heat capacity, rho Cp. Calibration of the instrument experimentally confirmed this behavior. This feature obviated the need to measure the thermal conductivity in order to determine rho Cp from the diffusivity equation, eta = lambda/rho Cp. The calorimeter has the following characteristics: reproducibility of loading: +/- 400 microJ/C degrees.cm3; baseline stability: +/- 10 microJ/C degrees.cm3 per 36 h; resolution (+/- 1 S.D.): +/- 50 microJ/C degrees.cm3; sample size 600 microliters.

Evidence for a membrane lipid defect in Alzheimer disease.

We have previously shown that cells normally maintain their lipid metabolic pools at a critical composition, appropriate for spontaneous assembly of a stable membrane bilayer at their physiological temperature. When disease affects membrane lipids such that the new composition will only form a stable bilayer at a critical temperature (T*), which differs from the physiological value, membrane destabilization and hence cellular damage will necessarily ensue. We have previously tested this pathogenetic mechanism in metachromatic leukodystrophy, a disorder with a known primary lipid metabolic defect. In the present study, we found T* for cerebral cortex lipids from three Alzheimer disease (AD) patients ranged between 19 and 28 degrees C, independent of membrane protein composition. Control cortex lipids yielded a normal value for T* of 37 degrees C. Thus, one possible mechanism for neurodegeneration in AD is membrane destabilization secondary to a lipid compositional aberration, which shifts T* away from 37 degrees C. This lipid defect is brain region-specific as cerebellar lipids from the AD patients gave a normal value for T*. Studies aimed at delineating the nature of the biochemical anomaly are in progress.

Membrane bilayer instability and the pathogenesis of disorders of myelin.

We have previously shown that total lipid extracts from normal nervous tissues spontaneously form a structure in vitro resembling the cell membrane bilayer, but only at a critical temperature, T*, equal to the 'physiological' temperature of the original tissues. In the present study, we found T* for normal human myelin lipids was 37 degrees C, in agreement with the concept that lipid metabolic pools maintain a critical composition in vivo which permits spontaneous formation of the (myelin) membrane bilayer at normal body temperature. But T* for myelin lipids from a patient with metachromatic leukodystrophy was less than 30 degrees C. Thus, myelin lipid composition was inappropriate for normal bilayer stability at this patient's core temperature, suggesting a mechanism whereby defective lipid metabolism in this disease could produce pathological myelin. The shift in T* in this patient was unlikely to be simply secondary to myelin destruction, as myelin lipids from a patient with advanced multiple sclerosis yielded a normal value for T* of 37 degrees C, even when extracted from areas of extensive demyelination.

Membrane bilayer assembly in neural tissue of rat and squid as a critical phenomenon: influence of temperature and membrane proteins.

Cell membrane bilayers have been reconstructed in vitro utilizing total lipid extracts from rat neural tissue (forebrain, cerebellum, brainstem and spinal cord) and from the optic lobe and fin nerve of the squid Loligo pealei. In agreement with the critical state theory of bilayer assembly (Gershfeld, N.L. 1986. Biophys. J. 50:457-461; Gershfeld, NL.L. 1989. J. Phys. Chem. 93:5256-5261), these lipid extracts spontaneously formed purely unilamellar structures in aqueous dispersion, but only at a critical temperature, T*, which was species dependent. For all the rat tissues T* = 37 +/- 1 degrees C; for squid neural extracts T* = 15.5 +/- 1.4 degrees C. These values correspond to 'physiological' temperatures for both organisms, implying that their lipid metabolism is geared to permit spontaneous assembly of unilamellar membranes at the ambient temperature in the tissues. Membrane protein composition had little or no effect on critical bilayer formation.

Thermodynamics of phospholipid bilayer assembly.

Thermodynamic properties of bilayer assembly have been obtained from measurements of the solubility of the sodium salt of dimyristoylphosphatidylglycerol (DMPG) in water. The standard free energy of bilayer assembly delta G degree a is shown to be RT 1n Xs + zF psi 0 where Xs is the mole fraction of dissolved lipid, F is the Faraday constant, z is the valence of the counterion (Na+), and psi 0 is the electrical double-layer potential of the ionized bilayer. The function d 1n Xs/dT was found to be discontinuous at 24 degrees C, the gel-liquid-crystal transition temperature (Tm) for DMPG. This function was unaffected when solubilities were measured in 0.001 M NaCl solutions; thus, psi 0 is constant in the experimental temperature interval (4-40 degrees C). Using a value of psi 0 = -180 mV [Eisenberg et al. (1979) Biochemistry 18, 5213-5223], and the temperature dependence of delta G degrees a, values for delta H degrees a and delta S degree a at 24 degrees C were calculated for the gel and liquid-crystal states of DMPG. For the gel, delta H degrees a and T delta S a are -26.2 and 12.7 kcal/mol, respectively; for the liquid-crystal, delta H degrees a and T delta S degrees a are -19.2 and -5.7 kcal/mol, respectively. The calculated value for the latent heat of the gel-liquid-crystal transition is 7 kcal/mol, in agreement with calorimetric measurements.

Thermal instability of red blood cell membrane bilayers: temperature dependence of hemolysis.

Rates of human red blood cell hemolysis were measured as a function of temperature. Three distinct temperature intervals for hemolysis were noted: a) At temperatures equal to or less than 37 degrees C no hemolysis was observed for the duration of the incubation (30 hr). b) For temperatures exceeding 45 degrees C hemolysis rates are rapid and are accompanied by gross changes in cellular morphology. The activation energy for hemolysis is 80 kcal/mole; this value is characteristic of protein denaturation and enzyme inactivation suggesting that these processes contribute to hemolysis at these high temperatures. c) Between 38 and 45 degrees C the energy of activation is 29 kcal/mole, indicating that a fundamentally different process than protein inactivation is responsible for hemolysis at these relatively low temperatures. A mechanism based on the concept of the critical bilayer assembly temperature of cell membranes (N.L. Gershfeld, Biophys. J. 50:457-461, 1986) accounts for hemolysis at these relatively mild temperatures: The unilamellar state of the membrane is stable at 37 degrees C, but is transformed to a multibilayer when the temperature is raised; hemolysis results because formation of the multibilayer requires exposing lipid-free areas of the erythrocyte surface. An analysis of the activation energy for hemolysis is presented that is consistent with the proposed unilamellar-multibilayer transformation.

Phospholipid surface bilayers at the air-water interface. III. Relation between surface bilayer formation and lipid bilayer assembly in cell membranes.

Lipid bilayer assembly in cell membranes has been simulated with total lipid extracts from human red blood cells and from mesophilic and thermophilic bacteria grown at several temperatures. Aqueous dispersions of these natural lipid mixtures form surface bilayers, a single bimolecular lipid state, but only at the growth temperature of the source organism. Thus, a single isolated bilayer state forms spontaneously in vitro from lipids that are available in vivo at the growth temperature of the cell. Surface bilayers form at a specific temperature that is a function of hydrocarbon chain length and degree of fatty acid unsaturation of the phospholipids; this property is proposed as an essential element in the control of membrane lipid composition.

Phospholipid surface bilayers at the air-water interface. II. Water permeability of dimyristoylphosphatidylcholine surface bilayers.

Dispersions of dimyristoylphosphatidylcholine (DMPC) in water have been reported to form a structure at 29 degrees C at the equilibrium air/water surface with a molecular density equal to that of a typical bilayer. In this study, the water permeability of this structure has been evaluated by measuring the rate of water evaporation from DMPC dispersions in water in the temperature range where the surface film density exceeds that of a monolayer. Evaporation rates for the lipid dispersions did not deviate from those for lipid-free systems throughout the entire temperature range examined (20-35 degrees C) except at 29 degrees C, where a barrier to evaporation was detected. This strengthens the view that the structure that forms at this temperature has the properties of a typical bilayer.

Phospholipid surface bilayers at the air-water interface. I. Thermodynamic properties.

Dispersions of dimyristoylphosphatidylcholine (DMPC) in water spontaneously form a surface bilayer at the equilibrium air/water surface (Gershfeld, N. L., and K. Tajima, 1979, Nature [Lond.]. 279: 708-709). This phenomenon has now been demonstrated with dispersions of dioleoylphosphatidylcholine (DOPC), dipalmitoylphosphatidylcholine (DPPC), and with a mixture of DMPC and DOPC. Each of these dispersions forms a surface bilayer at a singularity in temperature that is a characteristic of the phospholipid. The surface bilayer formed by the lipid mixture is shown to have the same composition as the bulk liquid-crystal phase of the dispersion, and the surface components have identical partial molar entropies as the bulk lipid components. These properties indicate that the surface bilayer has the same structure as the bilayer in the liquid-crystal phase of the bulk dispersion.

Quantitative analysis of disaturated lecithins in human plasma by ammonia chemical ionization mass spectrometry.

Three physiologically important disaturated lecithins have been analyzed quantitatively as the acetate derivatives using a solids inlet probe and ammonia chemical ionization. Concentration independent response factors have been determined over a tenfold range that brackets human plasma levels. The results obtained serve as an independent corroboration of gas chromatographic analyses.