Nina Borochov

Ligand-specific oligomerization of T-cell receptor molecules

T cells initiate many immune responses through the interaction of their T-cell antigen receptors (TCR) with antigenic peptides bound to major histocompatibility complex (MHC) molecules. This interaction sends a biochemical signal into the T cell by a mechanism that is not clearly understood. We have used quasi-elastic light scattering (QELS) to show that, in the presence of MHC molecules bound to a full agonist peptide, TCR/peptide–MHC complexes oligomerize in solution to form supramolecular structures at concentrations near the dissociation constant of the binding reaction. The size of the oligomers is concentration dependent and is calculated to contain two to six ternary complexes for the concentrations tested here. This effect is specific as neither molecule forms oligomers by itself, nor were oligomers observed unless the correct peptide was bound to the MHC. These results provide direct evidence for models of T-cell signalling based on the specific assembly of multiple TCR/peptide-MHC complexes in which the degree of assembly determines the extent and qualitative nature of the transduced signal. They may also explain how T cells maintain sensitivity to antigens present in only low abundance on the antigen-presenting cell.

Isolation and physical studies of the intact supercoiled: The open circular and the linear forms of CoIE1-plasmid DNA

For the study of DNA conformations, conformational transitions, and DNA-protein interactions, covalently closed supercoiled ColE1-plasmid DNA has been purified from cultures of Escherichia coli harboring this plasmid and grown in the presence of chloramphenicol according to the method of D.B. Clewell [J. Bact. 110 (1972)667]. The open circular and linear forms of the plasmid were prepared by digestion of the covalently closed, supercoiled form with pancreatic deoxyribonuclease and EcoRI-restriction endonuclease, respectively. The linear form was found to be very homogeneous by electron microscopy and sedimenting boundary analysis. Its physical properties (s0 20,w=16.3 S,D0 20,W=1.98 X 10(-8) cm2 s-1 and [eta]=2605 ml g-1) have been carefully determined in 0.2 M NaCl, 0.002 M NaPO4 pH 7.0,0.002 M EDTA, at 23 degrees C. Combination of s0 20, w (obtained by quasielastic laser light scattering) gave Ms,D=4.39 x 10(6). This value is in reasonable agreement with the molecular weight from total intensity laser light scattering M=4.30 x 10(6). The covalently closed and open circular forms of the ColE1-plasmid are less homogeneous due to slight cross-contamination and the presence of small amounts of dimers in these preparations. The weight fractions of the various components as determined by boundary analysis or electron microscopy are given together with the average quantities obtained in the same solvent for the supercoiled form ((s0 20,w)w=25.4 S, (D0 20,w)z=2.89 x 10(-8) cm2 s-1, [eta]= 788 ML G-1,Ms,D=4.69 x 10(6) and Mw=4.59 x 10(6)) and the open circular form (s0 20, w)w=20.1 S, (D0 20,w)z=2.45 x 10(-8) cm2 s-1, [eta]=1421 ml g-1,Ms,D=4.37 x 10(6) and Mw=4.15 x 10(6)). Midpoint analysis of the sedimenting boundaries allows unambiguous determination of the sedimentation coefficients of these two forms: s0 20,w=24.5 S and s0 20,w=18.8 S, respectively. Also deduced from total intensity light scattering were radii of gyration Rg (103.5, 134.2 and 186 nm) and second virial coefficients A2 (0.7, 4.8 AND 5.4 x 10(-4) mole ml/g2) for the supercoiled, the open circular and linear forms, respectively. The data presented are discussed in relation to the conformational parameters for the three forms in solution.

Cholesterol Crystalline Polymorphism and the Solubility of Cholesterol in Phosphatidylserine

There is a marked hysteresis between the heating and cooling polymorphic phase transition of anhydrous cholesterol. At a scan rate of 0.05 degrees C/min the difference in transition temperatures between heating and cooling scans is approximately 10 degrees C. This phenomenon also occurs with mixtures of cholesterol with phosphatidylserine and can result in an underestimation of the amount of crystalline cholesterol in a sample that has not been cooled sufficiently. With 1-palmitoyl-2-oleoyl phosphatidylserine and 1-stearoyl-2-oleoyl phosphatidylserine the cholesterol crystallites form while the lipid remains in the L(alpha) phase. Sonication of dimyristoyl phosphatidylserine with a 0.4 mol fraction cholesterol results in the loss of cholesterol crystallite diffraction, but only a partial loss of the polymorphic transition detected by calorimetry. We therefore conclude that the thermal history of the sample can have profound effects on the appearance of the polymorphic phase transition of cholesterol by differential scanning calorimetry. Depending on the morphology of the vesicles, diffraction methods may underevaluate the amount of cholesterol crystallites present.

Phase separation of cholesterol in dimyristoyl phosphatidylserine cholesterol mixtures

Abstract Miscibility of cholesterol in dimyristoyl phosphatidylserine (DMPS) bilayers was investigated by differential scanning calorimetry and X-ray scattering. The onset of cholesterol phase separation as detected by X-ray diffraction is at the molar ratio DMPS:cholesterol of 2:1 and 1.7:1 in the gel and in the liquid crystalline states of the phospholipid, respectively.

PHASE BEHAVIOR OF MIXTURES OF CHOLESTEROL AND SATURATED PHOSPHATIDYLGLYCEROLS

The interaction of cholesterol with a series of saturated phosphatidylglycerols was investigated using differential scanning calorimetry and X-ray diffraction. We find that the miscibility of cholesterol in phosphatidylglycerol bilayers is lower than in the corresponding phosphatidylcholine bilayers and decreases with increasing acyl chain length of the phospholipid. The influence of the negative charge of the phosphatidylglycerol on cholesterol miscibility is discussed.

Phase behaviour of heteroacid phosphatidylserines and cholesterol

Abstract The phase behavior of mixtures of cholesterol with 1-palmitoyl-2-oleoyl phosphatidylserine or with 1-stearoyl-2-oleoyl phosphatidylserine has been examined using differential scanning calorimetry and X-ray diffraction. We found that the miscibility of cholesterol in the bilayer is strongly affected by the small difference in length between the fully saturated fatty acid chains of the two phospholipids. Cholesterol crystallites are detected in the 1-palmitoyl-2-oleoyl-phosphatidylserine/cholesterol mixtures at a mol fraction of cholesterol of approximately 0.36, whereas for the mixtures with 1-stearoyl-2-oleoyl-phosphatidylserine they appear at a cholesterol mol fraction of approximately 0.2. A phase diagram, based on differential scanning calorimetry experiments without cryoprotectant, is presented for the 1-stearoyl-2-oleoyl phosphatidylserine/cholesterol system. This phase diagram indicates the marked immiscibility of cholesterol with either the gel or the liquid crystalline phase of this phospholipid.

The effect of protons or calcium ions on the phase behavior of phosphatidylserine-cholesterol mixtures

The influence of protons or calcium ions on the miscibility of cholesterol in phosphatidylserine has been examined using differential scanning calorimetry and X-ray diffraction. At pH 2.6, where the carboxyl group of the serine moiety is protonated, two endothermic transitions are observed in cholesterol-phosphatidylserine mixtures. The midpoint of the first is at 35 degrees C in the absence of cholesterol and decreases to approx. 15 degrees C for molar fraction of cholesterol 0.5. The second transition is centered at approx. 44 degrees C, almost independent of cholesterol content. The two lower temperature phases are lamellar and the high temperature phase has hexagonal symmetry. Cholesterol is more miscible in protonated phosphatidylserine than in the sodium form: cholesterol crystals are detected at a molar ratio of phosphatidylserine to cholesterol of about 1.7:1 as compared to about 2.3:1 at neutral pH. In the presence of calcium ions (1.3 Ca2+ per phosphatidylserine), a lamellar phase is observed with layer spacing 53 A which is independent of temperature (25 degrees C-65 degrees C) and of cholesterol content. Calcium ions cause reduced cholesterol solubility: crystallites are detected already at a molar ratio of 4:1.

A New High-Temperature Transition of Crystalline Cholesterol in Mixtures with Phosphatidylserine

Phosphatidylserine and cholesterol are two major components of the cytoplasmic leaflet of the plasma membrane. The arrangement of cholesterol is markedly affected by the presence of phosphatidylserine in model membranes. At relatively low mol fractions of cholesterol in phosphatidylserine, compared with other phospholipids, cholesterol crystallites are formed that exhibit both thermotropic phase transitions as well as diffraction of x-rays. In the present study we have observed and characterized a novel thermotropic transition occurring in mixtures of phosphatidylserine and cholesterol. This new transition is observed at 96 degrees C by differential scanning calorimetry (DSC), using a heating scan rate of 2 degrees C/min. Observation of the transition requires that the hydrated lipid mixture be incubated for several days, depending on the temperature of incubation. The rate of formation of the material exhibiting a transition at 96 degrees C is more rapid at higher incubation temperatures. At 37 degrees C the half-time of conversion is approximately 7 days. Concomitant with the appearance of the 96 degrees C peak the previously known transitions of cholesterol, occurring at approximately 38 degrees C and 75 degrees C on heating scans of freshly prepared suspensions, disappear. These two transitions correspond to the polymorphic transition of anhydrous cholesterol and to the dehydration of cholesterol monohydrate, respectively. The loss of the 75 degrees C peak takes a longer time than that of the 38 degrees C peak, indicating that anhydrous cholesterol first gets hydrated to the monohydrate form exhibiting a transition at 75 degrees C and subsequently is converted by additional time of incubation to an altered form of the monohydrate, showing a phase transition at 96 degrees C. After several weeks of incubation at 37 degrees C, only the form with a phase transition at 96 degrees C remains. If such a sample undergoes several successive heating and cooling cycles, the 96 degrees C peak disappears and the 38 degrees C transition reappears on heating. For samples of 1-palmitoyl-2-oleoyl phosphatidylserine or of 1-stearoyl-2-oleoyl phosphatidylserine having mol fractions of cholesterol between 0.4 and 0.7, the 38 degrees C transition that reappears after the melting of the 96 degrees C component generally has the same enthalpy as do freshly prepared samples. This demonstrates that, at least for these samples, the amount of anhydrous cholesterol crystallites formed is indeed a property of the lipid mixture. We have also examined variations in the method of preparation of the sample and find similar behavior in all cases, although there are quantitative differences. The 96 degrees C transition is partially reversible on cooling and reheating. This transition is also scan rate dependent, indicating that it is, at least in part, kinetically determined. The enthalpy of the 96 degrees C transition, after incubation of the sample for 3 weeks at 37 degrees C is dependent on the ratio of cholesterol to 1-palmitoyl-2-oleoyl phosphatidylserine or to 1-stearoyl-2-oleoyl phosphatidylserine, with the enthalpy per mole cholesterol increasing between cholesterol mol fractions of 0.2 and 0.5. Dimyristoyl phosphatidylserine at a 1:1 molar ratio with cholesterol, after incubation at 37 degrees C, exhibits a transition at 95 degrees C that reverses on cooling at 44 degrees C, instead of 60 degrees C, as observed with either 1-palmitoyl-2-oleoyl phosphatidylserine or 1-stearoyl-2-oleoyl phosphatidylserine. These findings along with the essential absence of the 96 degrees C transition in pure cholesterol or in cholesterol/phosphatidylcholine mixtures, indicates that the phospholipid affects the characteristics of the transition, and therefore the cholesterol crystallites must be in direct contact with the phospholipid and are not simply in the form of pure crystals of cholesterol. These observations are particularly important in view of recent observations of the presence of cholesterol crystals in biological systems.

The effect of ethanol on the structure of phosphatidylserine bilayers

Thermotrophic and structural effects of ethanol on phosphatidylserine (PS) membranes were investigated by differential scanning calorimetry (DSC) and X-ray diffraction. It was found that up to 15% (v/v) added ethanol, there is little change in the melting temperature of the phospholipid and no change in the interbilayer (d) spacing in the gel phase, indicating that there is no interdigitation of the hydrocarbon chains. Above the melting temperature of the phospholipid, a large decrease of the d spacing, due primarily to a decrease in the thickness of the bilayer, was found. Ethanol molecules located in the headgroup region apparently expand the area available to the headgroups with concomitant coiling of the acyl chains, resulting in marked thinning of the lipid layer.

Phase separation of cholesterol and the interaction of ethanol with phosphatidylserine-cholesterol bilayer membranes.

Thermotropic and structural effects of ethanol on phosphatidylserine (PS) membranes containing up to 0.4 mol fraction cholesterol were investigated by differential scanning calorimetry, X-ray diffraction and fluorescence spectroscopy. It was found that in the presence of cholesterol, 10% (v/v) added ethanol depresses the melting temperature of the phospholipid by approximately 2 degrees C, similar to what was observed in the absence of cholesterol. Below the melting temperature the progressive disordering effect of added cholesterol is weakly enhanced by the presence of ethanol. In the liquid crystalline state, the marked decrease in the thickness of the bilayer which ethanol causes in the absence of cholesterol (Chem. Phys. Lipids 92 (1998) 127), is also observed in its presence. We conclude that, in contrast to what has been observed for zwitterionic phospholipids, high concentrations of cholesterol do not diminish the interaction of ethanol with PS membranes. With addition of 10% (v/v) ethanol, crystalline cholesterol diffraction, an indication of phase separation of the sterol, appears at mol fraction cholesterol 0.34, as compared to 0.3 in the absence of ethanol (Chem. Phys. Lipids 92 (1998) 71).