|
|
||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Journal of Lipid Research, Vol. 44, 567-575, March 2003
Copyright © 2003 by Lipid Research, Inc.


* Max-Planck Institute for Colloids and Interfaces, c/o HASYLAB, Notkestrasse 85, D-22603 Hamburg, Germany
Molecular Cell Biology and Biochemistry, Department of Biology, University of the Balearic Islands, E-07071 Palma de Mallorca, Spain
Published, JLR Papers in Press, December 16, 2002. DOI 10.1194/jlr.M200356-JLR200
2 S. S. Funari and F. Barceló contributed equally to this work. ![]()
1 To whom correspondence should be addressed. e-mail: sergio.funari{at}desy.de
| ABSTRACT |
|---|
|
|
|---|
The effects of free fatty acids (FFAs) on membrane structure are relevant for several reasons: i) some biological membranes contain very high levels of FFAs. ii) Mediterranean diets with high OA intake have been shown to exert protective effects against tumoral and hypertensive pathologies. iii) FFA derivatives have been developed as antitumoral and antihypertensive drugs.
Supplementary key words free fatty acids phosphatidylcholine lipid membrane structure lamellar phases nonlamellar phases hexagonal HII phase epitaxial relationship
| INTRODUCTION |
|---|
|
|
|---|
PE is the major phospholipid species in bacterial membranes and in the inner leaflet of mammalian plasma membranes (3). PE is mainly organized into lamellar structures that define the cell, constituting a physical boundary and support for the membrane proteins. PEs are hexagonal (HII)-prone lipids at high temperatures because of their "molecular shape," which resembles a truncated cone and confers a negative curvature strain to model (17, 18) and biological membranes (19).
Numerous studies have demonstrated that localized or mobile, stable, or transient nonlamellar lipid structures exert defined membrane functions (4, 20, 21). Some of the roles attributed to HII-prone phospholipids include facilitation of fusion and fission of bilayers (22, 23), modulation of membrane permeability and elasticity (2), protein transport (8), the regulation of G-protein and protein Kinase C (PKC) localization and activity (57), as well as chaperone-like activity (3) among others. Thus, the special features conferred to membranes by PE are crucial to membrane structure/function.
The present work was designed to study the effect of the C-18 FAs, OA (18:1 cis
9), elaidic acid (EA, 18:1 trans
9), and stearic acid (SA, 18:0) on the structural properties of lamellar and nonlamellar PE bilayers. In this context, X-ray scattering is the most appropriate technique to characterize the mesomorphic behavior of membrane phospholipids and the alterations induced by FAs or other molecules. Here, it has been shown that OA modulated the membrane structure, inducing negative membrane curvature strain (HII-phase facilitation) on PE lipids, whereas the closely related FAs, EA and SA, did not alter markedly the phospholipid bilayer and HII properties. Because the plasma membrane structure regulates a wide variety of cell functions (see above), the present results are relevant to understanding the relationship between membrane structure and function (e.g., 8, 24, 25).
| EXPERIMENTAL PROCEDURES |
|---|
|
|
|---|
Sample preparation
Multilamellar lipid vesicles (15% phospholipids; 85% water, by weight) were prepared in 20 mM Hepes, pH 7.4, in the absence or presence of the above FAs and at the molar ratio indicated below. Samples were thoroughly homogenized with a pestle-type minihomogenizer (Sigma) and vortexed until a homogeneous mixture was obtained. The suspensions were submitted to three temperature cycles (heated up to 70°C and cooled down to 4°C). Then, they were immediately stored at -80°C under argon until use. Before X-ray scattering experiments, samples were allowed to equilibrate at 4°C for 72 h.
X-ray scattering
Small- and wide-angle (SAXS and WAXS) synchrotron radiation X-ray scattering data were collected simultaneously using standard procedures on the Soft Condensed Matter beamline A2 (2628) at the storage ring DORIS III of the Deutsches Elektronen Synchrotron (DESY).
Positions of the observed peaks were converted into distances, d, after calibration using standards with well-defined scattering patterns. Silver behenate and polyethylene theraphtalate were used to calibrate the SAXS and WAXS regions, respectively. Data from each sample were acquired continuously for 15 s at each temperature, followed by a waiting time of 45 s with a local shutter closed. Linear detectors with delay line readout were used (29). The measurements were designed to compare the effects of the various FAs used on the HII-phase propensity. During data collection, samples were heated from 27°C to 75°C (DEPE), 2°C to 32°C (DOPE), or 27°C to 80°C (POPE) at a scan rate of 1°C/min. Then, they were kept at the highest temperature for 5 min and finally cooled down to the lowest temperature at the same scan rate. At the end of the experiment, the samples were maintained at the lowest temperature for 5 min to confirm the phase structure. The experimental conditions did not affect the phase sequence structures or their parameters (e.g., their characteristic dimensions). Interplanar distances, dhkl, were calculated according to Eq. 1:
where s is the scattering vector, 2
is the scattering angle,
(0.15 nm) is the X-ray wavelength, and hkl are the Miller indexes of the scattering planes. For HII-phases, the unit-cell dimension, a, was calculated using the following relationship: a = 2d10 /31/2. This parameter (a) also corresponds to the diameter of the rods forming the hexagonal lattice.
The lamellar L
phase was identified in the temperature scans by a single reflection peak at s
0.19 nm-1, with a very good signal-to-noise ratio, occurring between patterns associated with the gel (Lß) and HII phases.
| RESULTS |
|---|
|
|
|---|
DEPE-OA mixtures
In the absence of OA, DEPE clearly showed a phase sequence from lamellar gel Lß to lamellar liquid crystalline L
to hexagonal HII phase, with increasing temperature. The presence of OA inhibited the occurrence of the L
phase (Fig. 1)
. This effect was concentration dependent, since the range of temperatures where lamellar and nonlamellar phases occurred depended on the DEPE-OA molar ratio. The greater the OA content, the narrower the L
-phase temperature range. At a molar ratio of 10:1, the L
phase was not observed at all.
|
42 wt%. At high amounts of OA in the mixture, say PE-OA 1:1, one would expect a different effect, i.e., a separation of the mixture into two homogeneous phases: one, HII, containing almost pure DEPE and another, L
, containing almost only OA.
Lamellar phases
All samples studied showed a lamellar gel Lß phase up to 38°C with an approximately constant repeat distance of 6.4 nm. The similarity observed between gel phases of pure DEPE and DEPE-OA mixtures indicated a complete incorporation of the FA into the lipid bilayer, as previously described (31). DEPE, in the absence of FAs, showed a single L
phase between 3860°C. From 61°C to 66°C, L
and HII phases coexisted. Above 66°C DEPE was organized into an HII structure. Along the temperature range L
was observed, the repeat distance measured decreased linearly from 5.48 nm (38°C) to 5.11 nm (66°C). The lattice parameter for HII phases also decreased linearly in a temperature-dependent manner. Compression coefficients for these structures are shown in Table 1.
|
phase (Fig. 1). OA, at a molar ratio of 40:1 (DEPE-OA), induced a significant decrease (about 15°C) in the L
-to-HII phase transition temperature (TH = 46°C), and both phases coexisted in the range of 46°C to 52°C. Note the
6°C temperature range of coexistence of these phases, as observed in the single DEPE system. The Lß-to-L
transition, however, was not markedly altered by the presence of the FA (Tm = 37°C in the absence or presence of OA). The influence of OA was further evidenced by the enhancement of the thermal sensitivity of the lamellar lattice parameter (
d/
T), whose value changed from -0.012nm/°C (in the absence of OA) to -0.016 nm/°C (in the presence of OA). Higher OA content (DEPE-OA 20:1, mol/mol) (Figs. 2, 3)
induced greater effects on the structural properties of DEPE dispersions. First, the occurrence was observed of a broad peak alongside the reflections characteristic of the gel Lß phase. Heating induced a shift of this broad peak toward smaller angles, accompanied by a loss in resolution. Further characterization of this peak was not possible with the available data. Second, the Lß phase was clearly identified by a sharp and intense SAXS reflection, accompanied by a well-defined reflection in the WAXS region (Fig. 2). The corresponding lamellar spacing remained constant in the temperature range of its occurrence.
|
|
phase, which was observed only from 38°C to 53°C and coexisted with either Lß or HII phases (DEPE-OA, 20:1 mol/mol). The thermal sensitivity of the lamellar lattice parameter (
d/
T) was -0.014 nm/°C. An interesting aspect of this transition was the continuity of the interplanar distance between Lß and HII phases, which led us to consider an epitaxial relationship between the (10)-planes of both phases (see below). In this situation, the phase transition was identified by the WAXS region peaks (Fig. 2). The complete vanishing of the peak in the WAXS characterized the change from a well-organized all-trans to liquid-like conformation of the acyl chains of the phospholipid and OA present in the mixture. It is also interesting to note that in the temperature range where L
and HII phases coexisted, the lattice parameter of the hexagonal phase HII remained basically constant (6.4 nm), decreasing only after the system turned monophasic (Fig. 3).
In the presence of higher OA concentrations (DEPE-OA, 10:1, mol/mol), the L
phase was not observed over the temperature range studied (2775°C) (Figs. 4, 5)
. The Lß phase appeared between 27°C and 35°C, coexisting with the hexagonal phase. Above this temperature, the Lß phase vanished and the lipids reorganized directly into HII phase (3575°C) without transit through the liquid crystalline L
phase. It is noteworthy that the Lß phase showed also a constant lattice parameter of 6.35 nm along the temperature range where it was observed, but the HII showed a different behavior (Fig. 3). In the two-phase temperature range (2735°C), HII phases expanded linearly (
d/
T = 0.13 nm/°C), whereas in the single phase region, there was a temperature-dependent contraction (
d/
T = -0.020 nm/°C). In addition, this phase contraction was one order of magnitude smaller than the HII phase expansion observed in the two-phase region.
|
|
or Lß) and HII structures coexisted. The threshold temperature for HII phases, either alone or forming a binary system, greatly depended on OA concentration. For low OA concentrations (DEPE-OA, 40:1 and 20:1, mol/mol), the binary system consisted of HII and lamellar L
phases, while for high OA concentrations (DEPE-OA, 10:1, mol/mol), the binary system was formed by HII and Lß structures. The temperature range of coexistence of these phases also depended on the DEPE-OA ratio in the mixture. For all DEPE-OA systems studied, the compressibility of the HII lattice parameter was inversely proportional to temperature increase (
d/
T
-0.022 nm/°C). The HII phase contraction factor was about 2-fold greater than that of the lamellar L
phase (
d/
T
-0.014 nm/°C) (Table 1). This effect is indicative of a phase dehydration with a concomitant reduction of the surface area per molecule. The negative value of the thermal coefficient (
d/
T < 0) for both phases (L
and HII) can be attributed to a continuous coiling up of the acyl chains forming the mesomorphic units of the structures. At the initial formation of the hexagonal phase, a typical lattice spacing of 6.4 nm was observed, with the exception of DEPE-OA at a molar ratio of 10:1 for which this value was smaller (i.e., 5.9 nm). This reduced value of the lattice parameter implies a more effective packing of the molecules, and may account for the direct transition between Lß and HII phases, without the formation of the liquid crystalline lamellar L
phase.
"Apparent" epitaxial relationship
Structural aspects of the Lß-to-HII phase transition of DEPE-OA (20:1, mol/mol) are complex. The sequence of diffraction patterns collected during the temperature scans showed defined phase transitions that allowed unequivocal characterization of the structures and their respective lattice parameters. The diffraction patterns always showed a peak at s = 0.155 nm-1 (see Eq. 1 in Experimental Procedures), which apparently related epitaxially the Lß and HII phases. Although through the transition the broadness and intensity of this peak changed drastically, its position could still be clearly identified. It was associated with the (10)-planes of both L
and HII phases. Changes in the peak features marked the Lß-to-HII phase transition, where significant morphological changes in the mesogenic units take place; therefore, neither positional nor orientational order can be maintained properly, causing the peak broadening. Simultaneously, the phase transition was also observed between Lß and L
phases, with onset at the same temperature. In this situation, the gel phase initially develops into two phases, L
and HII phases. At higher temperatures, L
also turns into a HII phase. The mechanism of such complex phase transition is still unclear and beyond the scope of the present study. With the data available, we could not quantify the relative amount of each phase formed upon the gel transition. However, the peak intensities (s = 0.155 nm-1) suggested a greater abundance of L
phase with respect to HII phases when both phases coexisted in the DEPE-OA sample at a molar ratio of 20:1. This could be due to the melting of the acyl moiety that induced a negative curvature strain necessary to form rods characteristic of hexagonal phases. In this system, the lamellar phase plays the role of an intermediate, as in chemical reactions, in which the transformation among reagents and products follows a competitive path between kinetic and energetic dominance. For the DEPE-OA (20:1) system, the L
phase is kinetically favorable but with higher total energy, therefore soon after its formation, it transforms into the energetically favorable hexagonal phase. Although the phase transition contained the elements required for an epitaxial relationship between these phases, additional events taking place simultaneously, e.g., the formation of the L
phase, impaired a proper classification as such. Therefore, we called it an "apparent" epitaxial relationship between gel and HII phases.
Although little is known about the modulation of membrane structural properties by FAs, there is evidence that the incorporation of amphiphilic molecules into the lipid matrix induces drastic changes in the bilayer behavior. For example, the addition of nonionic surfactants into phosphatidylcholine bilayers modifies the structure of the lipid self-assembly when compared with the phospholipid alone (32). Thus, in the presence of C12EO4 (tetraethyleneglycol-mono-n-dodecylether), dipalmitoyl phosphatidylcholine (a lamellar-prone lipid) formed a pseudobinary mixture in excess of water (33). Similarly, POPC organizes into HII phases in the presence of C12EO2, although neither molecule aggregates into nonlamellar structures separately (34). C12EO2-POPC (1:2, mol/mol) mixtures are arranged into Lß and HII structures below 20°C, but only a single HII phase is observed above this temperature. Interestingly, in the two-phase region, the HII phase expands while heating, but when the system turns into a single HII phase, it undergoes contraction (35). A similar behavior was observed here for DEPE-OA, 10:1 (mol/mol).
Recently, Yang and Huang, studying diphytanoyl phosphatidylcholine (DPhPC) supported on a silicon nitride window using 2D diffraction patterns, were able to determine the structure of a stalk (rhombohedral), an intermediate structure between L
and HII (36). This structure is part of a current model for cell fusion, thus the strong interest in determining its parameters and conditions of occurrence. The spontaneous question that arises is if we would get the same structure using similar preparations and conditions with our samples. We do not expect so for different reasons. First, our samples contain much more water, 85 wt%, which brings the system to an excess of water condition. This means both lipid and FA in the mixture are fully hydrated, against a relative humidity of 7080% in the study on DPhPC. Finally, in multi-component systems, one would expect that at a point of large changes in topology, as is necessary for the formation of the stalk structure, segregation between the components should occur. In this situation, the system can no longer be considered homogeneous on a microscale; therefore, comparison between these systems has to be viewed with great care and attention to their differences in preparation. Moreover, the system based on the polyoxyethylene glycol-alkyl ether, C16EO6, in water when studied under very slow temperature scan rate, also showed a rhombohedral phase (37), but with no indication of stalk formation, although the X-ray data were collected with a linear detector. Despite large differences in structure, C16EO6 and DPhPC have in common a similar chain length and a large uncharged head group different from our samples.
DEPE-EA and DEPE-SA mixtures
EA and SA are FAs closely related to OA in terms of chemical structure (SA is a saturated 18:0 FA and EA is an isomer of OA, both 18:1 FAs). Conversely to OA, its congeners EA and SA induce smaller changes in the structural properties of DEPE. The mixtures DEPE-EA (20:1 and 10:1, mol/mol) and DEPE-SA (20:1, mol/mol) (Table 1, Fig. 6)
followed a similar trend. In DEPE-SA mixtures, L
was observed as a single phase up to 60°C, and it was present until 66°C along with an HII phase. In the absence of SA, the temperatures for these phases where 61°C and 66°C, respectively. DEPE-EA (20:1, mol/mol) mixtures showed a single L
phase up to 57°C and L
+HII up to 59°C. Over 66°C for DEPE-SA, and over 59°C for DEPE-EA, a single HII phase was observed (Table 1). On the other hand, the HII lattice parameter of DEPE was altered by OA (6.27 nm and 5.87 nm for DEPE and DEPE-OA, 20:1, mol/mol, respectively), but EA and SA did not induce important changes in this parameter at 72°C (6.21 nm and 6.33 nm, respectively) (Table 1).
|
d/
T
-0.021 nm/°C) for DOPE and DOPE-EA was similar to that of DEPE samples. In contrast, the mixture DOPE-OA had a significantly different value (
d/
T
-0.014 nm/°C), indicating that the thermal sensitivity of DOPE is altered by the presence of OA. Comparison of the systems DOPE-OA and DEPE-OA at the same molar ratio (20:1) also demonstrated the influence of the cis double bond on the phospholipid structure, inducing the formation of HII phase in mixtures containing DOPE, but not DEPE at 30°C (Table 1).
POPE systems
We also studied the effect of OA on POPE to further determine the effect of this FA on another PE derivative with two different acyl chains. POPE dispersions organized into lamellar phases over most of the temperature range studied (27°C to 80°C). Lß and L
phases coexisted and could be individually identified over a large range of temperatures up to 37°C. Between 37°C to 71°C, we observed a single lamellar (L
) phase. Over 71°C, L
and HII phases coexisted. In the temperature range of this study, a single- HII phase was not observed. OA lowered the L
-to-HII phase transition temperature from 71°C to 51°C. Above 51°C, both L
and HII phases coexisted up to 62°C when the system turned into a single HII phase (Fig. 7)
. This result clearly indicated that OA also facilitated the formation of POPE hexagonal phases. The lattice parameter of POPE was also altered by the presence of OA (6.34 nm and 5.92 nm in the absence or presence of OA, respectively). The compressibility factor for the HII phase of POPE (-0.029 nm/°C) increased in the presence of OA (-0.019 nm/°C), similar to what it was observed for DEPE and DOPE. It should be noted that for phospholipid-OA mixtures with molar ratio 10:1, DEPE and POPE showed essentially the same compressibility factor (Table 1).
|
| DISCUSSION |
|---|
|
|
|---|
HII-prone phospholipids, such as PE, are involved in a number of cellular functions, including the development of endocytic (membrane fission) and exocytic (membrane fusion) processes (42) and the regulation of activities of membrane proteins (2, 48). PE has been shown to accumulate at the cleavage furrow during cell division in E. coli (23) and exhibits a chaperone-like activity in E. coli (3). The high proportion of PE in membranes and the precise regulation of their levels indicate that this lipid has a great functional relevance (49). Because PE is a major membrane lipid species capable of organizing into nonlamellar phases, and the FA composition modulates the membrane properties, this work was designed to study the effects of OA (and its congeners EA and SA) on membrane mesomorphism. In this context, the FA concentration and the conformation of its double bond appeared to be important factors that influenced the properties of macrostructures formed by the mixtures used in this study. Thus, OA induced important concentration-dependent alterations in the supramolecular organization of PE derivatives, whereas the closely related FAs, EA and SA, did not. OA probably exerted a lateral pressure on PE FA moieties, favoring a negative-curvature strain. This effect induced the formation of inverted tubular micelles, which are the basic supramolecular units of the HII-phase lattice. This hypothesis is consistent with the marked decrease of the L
-to-HII phase transition temperature induced by OA. In contrast, EA and SA exerted very modest effects on PE structural properties. The different effects promoted by OA, EA, and SA on lipid mesomorphism could be also explained from the point of view of the lipid packing parameter (50). This property has been used to explain secondary structures formed by membrane lipids (51). Thus, lamellar phases are favored by cylinder-shaped phospholipids (e.g., PC with a bulky polar head), and inverted micelles, such as HII phases, are formed by truncated cone-shaped phospholipids (e.g., PE with a small polar head). With respect to the FA structure, OA (18:1 cis
9) has a "molecular shape" similar to a boomerang, whereas EA (18:1 trans
9) and SA (18:0) resemble a rod. This is the main structural difference between OA and its congeners EA and SA. In fact, the chemical compositions of OA and EA are identical (C18O2H34), while SA (C18O2H36) has only 2 H more because of the absence of double bonds. The effect of OA cannot be only attributed to the presence of a double bond, since EA did not exert similar effects on PE structure. Moreover, EA and SA, which differ in chemical composition but are closer in terms of "molecular shape," appeared to have similar (modest) effects on the membrane structure. Then, OA effects on PE lipid mesomorphism are most probably due to its molecular structure. Epand et al. (31) observed an enhanced ability of OA, with respect to other FAs, for lowering the bilayer-to-hexagonal phase temperature at different pHs. Our hypothesis, based on the FA "molecular shape," also explains the ability of OA to facilitate nonlamelar phases from a complementary point of view. Moreover, the present study identifies each phase coexisting in multi-phase regions, whose extent is dependent not only on the FA conformation but also on its concentration.
The thermal compressibility factor of the different phases also evidenced differences between OA, EA, and SA on the behavior of PE macrostructures. The compressibility of the L
-lattice parameter (
d/
T
-0.013 nm/°C) was similar for all DEPE samples studied. For DEPE HII phases, the thermal compressibility factor was about -0.022 nm/°C. However, the presence of OA in DOPE membranes (DOPE-OA, 20:1, mol/mol) induced a marked alteration of this parameter (
d/
T = -0.014 nm/°C). Conversely, EA did not induce any decrease in the DOPE lattice thermal compressibility factor (
d/
T
-0.022 nm/°C). These results highlight, on one hand, the influence of OA on nonlamellar membrane structures, and suggest cooperative effects between OA (FFA) and the FA moieties (also OA residues) of DOPE. A direct effect of this cooperativity is an improved packing of DOPE molecules in the HII structure.
EA and SA had little effect on the L
-to-HII DEPE phase transition temperature (
T = -4 and -1°C, respectively) at a molar ratio 20:1 (DEPE-FA). In contrast, OA induced very important alterations of this temperature (
T = -16°C) at the same molar ratio (Table 1, Fig. 1). As a matter of fact, for the DEPE-OA mixture at a molar ratio of 20:1, L
phase was not observed as a single phase and at a molar ratio of 10:1 was not observed at all. Similarly, the POPE-OA mixture (10:1, mol/mol) showed a decrease in the L
-to-HII phase transition of about 20°C, albeit here L
phases appeared over a narrow temperature range, indicating that the effect of OA depended also on the FA moieties species of the PEs.
In summary, the present work quantifies the effects of the FAs OA, EA, and SA on membrane structure. From the structural behavior of the model systems studied here, we conclude that the FA molecular shape facilitates the HII phase formation by modulating the bilayer curvature. This regulation of the membrane structure explains in part the modulation exerted by OA on membrane and cell functions: membrane fluidity, permeability, domain formation, exo/endocytosis, cell division, signal transduction, membrane protein (G-proteins, G-protein-coupled receptors, adenylyl cyclase) activities, blood pressure control, and antiproliferative (antitumoral) effects.
| ACKNOWLEDGMENTS |
|---|
Manuscript received September 9, 2002 and in revised form December 10, 2002.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
S. Teres, G. Barcelo-Coblijn, M. Benet, R. Alvarez, R. Bressani, J. E. Halver, and P. V. Escriba Oleic acid content is responsible for the reduction in blood pressure induced by olive oil PNAS, September 16, 2008; 105(37): 13811 - 13816. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. H. Borchert, M. Giggey, F. Kolar, T. M. Wong, P. H. Backx, and P. V. Escriba 2-Hydroxyoleic acid affects cardiomyocyte [Ca2+]i transient and contractility in a region-dependent manner Am J Physiol Heart Circ Physiol, April 1, 2008; 294(4): H1948 - H1955. [Abstract] [Full Text] [PDF] |
||||
![]() |
F. Barcelo, J. Prades, J. A. Encinar, S. S. Funari, O. Vogler, J. M. Gonzalez-Ros, and P. V. Escriba Interaction of the C-Terminal Region of the G{gamma} Protein with Model Membranes Biophys. J., October 1, 2007; 93(7): 2530 - 2541. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Hlavackova, J. Neckar, J. Jezkova, P. Balkova, B. Stankova, O. Novakova, F. Kolar, and F. Novak Dietary Polyunsaturated Fatty Acids Alter Myocardial Protein Kinase C Expression and Affect Cardioprotection Induced by Chronic Hypoxia Experimental Biology and Medicine, June 1, 2007; 232(6): 823 - 832. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. S. Perona, O. Vogler, J. M. Sanchez-Dominguez, E. Montero, P. V. Escriba, and V. Ruiz-Gutierrez Consumption of Virgin Olive Oil Influences Membrane Lipid Composition and Regulates Intracellular Signaling in Elderly Adults With Type 2 Diabetes Mellitus J. Gerontol. A Biol. Sci. Med. Sci., March 1, 2007; 62(3): 256 - 263. [Abstract] [Full Text] [PDF] |
||||