By using synchrotron radiation, a movie was made of the X-ray scattering pattern from a biological liquid crystal undergoing a phase transition induced by a pressure jump. The system studied includes the fully hydrated phospholipid dihexadecylphosphatidylethanolamine in the lamellar gel (L beta') phase at a temperature of 68 degrees C and a pressure of 9.7 MPa (1400 psig). Following the rapid release of pressure to atmospheric the L beta' phase transforms slowly into the lamellar liquid crystal (L alpha) phase. The pressure perturbation is applied with the intention of producing a sudden phase disequilibrium followed by monitoring the system as it relaxes to its new equilibrium condition. Remarkably, the proportion of sample in the L alpha phase grows linearly with time, taking 37 s to totally consume the L beta' phase. The time dependencies of radius, peak intensity, and width of the powder diffraction ring of the low-angle (001) lamellar reflections were obtained from the movie by image processing. The concept of an "effective pressure" is introduced to account for the temperature variations that accompany the phase transition and to establish that the observed large transit time is indeed intrinsic to the sample and not due to heat exchange with the environment. The reverse transformation, L alpha to L beta', induced by a sudden jump from atmospheric pressure to 9.7 MPa, is complete in less than 13 s. These measurements represent a new approach for studying the kinetics of lipid phase transitions and for gaining insights into the mechanism of the lamellar order/disorder transition.

We present here the first study of the use of a pressure-jump to induce the ripple (P beta')/lamellar liquid crystal (L alpha) phase transition in fully hydrated 1,2-dimyristoyl-sn-glycero-3-phosphocholine (DMPC). The transition was monitored by using time-resolved x-ray diffraction (TRXRD). Applying a pressure-jump from atmospheric to 11.3 MPa (1640 psig, 111.6 atm) in 2.5 s induces the L alpha to P beta' phase transition which takes place in two stages. The lamellar repeat spacing initially increases from a value of 66.0 +/- 0.1 A (n = 4) to a maximum value of 70.3 +/- 0.8 A (n = 4) after 10 s and after a further 100-150 s decreases slightly to 68.5 +/- 0.3 A (n = 4). The reverse transition takes place following a pressure jump in 5.5 s from 11.3 MPa to atmospheric pressure. Again, the transition occurs in two stages with the repeat spacing steadily decreasing from an initial value of 68.5 +/- 0.3 A (n = 3) to a minimum value of 66.6 +/- 0.3 A (n = 3) after 50 s and then increasing by approximately 0.5 A over a period of 100 s. The transition temperature increases linearly with pressure up to 14.1 MPa in accordance with the Clapeyron relation, giving a dT/dP value of 0.285 degrees C/MPa (28.5 degrees C/kbar) and an associated volume change of 40 microliters/g. A dynamic compressibility of 0.13 +/- 0.01 A/MPa has been determined for the L alpha phase. This value is compared with the equilibrium compressibilities of bilayer and nonbilayer phases reported in the literature. The results suggest testable mechanisms for the pressure-induced transition involving changes in periodicity, phase hydration, chain order, and orientation. A more complete understanding of the transition mechanism will require improvement in detector spatial resolution and sensitivity, and data on the pressure sensitivity of phase hydration.

A beryllium rod with an internal bore diameter of 1.5 mm and a wall thickness of 2.4 mm constitutes an x-ray compatible cell for hydrostatic pressures rated up to 1800 bar (180 MPa, 26 000 psi) at temperatures up to 90 °C. The system can be used to produce static and oscillating pressures, pressure ramps, and pressure jumps. Pressure is generated by compressing water, the pressure medium, in a syringe-type high-pressure pump. Commercially available pressure valves serve to produce sinusoidal pressure oscillations and pressure jumps for any initial and final pressure between atmospheric pressure and 1800 bar. Typical rise times for a minimally damped and damped pressure jump from atmospheric pressure to 1600 bar are ∼2 and 5 ms, respectively. A depressurization jump from 450 bar to atmospheric has a decay time of order 5 ms. This equipment has the advantage of being relatively easy to build and to operate. However, it does not provide the fastest possible pressure jumps. The assembly of thermocouples and wire feedthroughs for operation under high pressure is detailed. The apparatus described here has been used in time-resolved x-ray diffraction studies of lipid phase transitions using synchrotron radiation. We anticipate using it to collect static and slow kinetic data on a conventional x-ray source.

The kinetics and mechanism of the barotropic lamellar gel (L beta')/lamellar liquid crystal (L alpha) phase transition in fully hydrated 1,2-dihexadecyl-sn-glycero-3-phosphoethanolamine (DHPE) has been studied using time-resolved x-ray diffraction (TRXRD). The phase transition was induced by pressure jumps of varying amplitudes in both the pressurization and depressurization directions at controlled temperature (78 degrees C). Both low- and wide-angle diffracted x rays were recorded simultaneously in live time using an x-ray-sensitive image intensifier coupled to a CCD camera and Super-VHS videotape recorder. Such an arrangement allowed for the direct and quantitative characterization of the long- (lamellar repeat spacing) and short-range order (chain packing) during a kinetic experiment. The image-processed live-time x-ray diffraction data were fitted using a nonlinear least-squares model, and the parameters of the fits were monitored continuously throughout the transition. The pressure-induced transitions from the L alpha to the L beta' phase and from the L beta' to the L alpha phase was two-state (no formation of intermediates apparent during the transition) to within the sensitivity limits of the method. The corresponding transit time (the time during which both phases coexist) associated with the long- and short-range order of the pressurization-induced L alpha-to-L beta' phase transition decreased to a limiting value of approximately 50 ms with increasing pressure jump amplitude. This limiting value was close to the response time of the detector/recording system. Thus, the intrinsic transit time of this transition in fully hydrated DHPE at 78 degrees C was less than or equal to 50 ms. In contrast, the depressurization-induced L beta'-to-L alpha phase transition was slower, taking approximately 1 s to complete, and occurred with no obvious dependence of the transit time on pressure jump amplitude. In the depressurization jump experiment, the lipid responded rapidly to the pressure jump in the L beta' phase up to the rate-determining L beta'-to-L alpha transition. Such behavior was examined carefully, as it could complicate the interpretation of phase transition kinetic measurements.

A pressure-temperature phase diagram for 1,2-dihexadecyl-sn-glycero-3-phosphoethanolamine (DHPE) in a 50% (w/w) mixture with water has been constructed in the range 1-1200 bar and 65-90°C. The lamellar gel (L ′ ) phase prevails at high pressure and low temperature. It gives way successively to the lamellar liquid crystalline (L R ) and inverted hexagonal (H II ) phases as pressure is lowered and temperature is raised. Measurements were made using synchrotron-based X-ray diffraction and a high-pressure beryllium cell. The pressure dependence of the L ′ /L R transition temperature, dT m /dP, is 25 ( 2°C/kbar. Compression/expansion in the lamellar phases was highly anisotropic. The isothermal pressure dependence of the chain-packing and lamellar repeat distance, (∂d/∂P) T , in both the L R and L ′ phases was insensitive to temperature in the range studied. The average (∂d/∂P) T values for the lamellar repeat in the L R and L ′ phases and for the chainpacking repeat in the L ′ phase are 3.0 ( 0.1, 1.18 ( 0.06, and -0.117 ( 0.04 Å/kbar, respectively. In like manner, the isobaric temperature dependence of the chain-packing and lamellar repeat distance, (∂d/∂T) T , in both the L R and L ′ phases is insensitive to pressure in the range studied. The average (∂d/∂T) P values for the lamellar repeat in the L R and L ′ phases and for the chain-packing repeat in the L ′ phase are -0.154 ( 0.005, -0.053 ( 0.003, and 0.0043 ( 0.0002 Å/°C, respectively. The absolute value of the ratio of the isothermal pressure dependence to the isobaric temperature dependence for all structural parameters measured, including area per chain, is remarkably close to the dT m /dP value reported above. A careful evaluation of mesophase structure parameters, a one-dimensional electron density profile, and a wide-angle diffraction profile has enabled us to describe in considerable detail the structure response of the L ′ phase in hydrated DHPE to temperature and pressure. Problems that can arise due to X-ray damage are discussed.