| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Biophys J, April 2001, p. 1649-1658, Vol. 80, No. 4
Department of Chemistry, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina 27599 USA
 |
ABSTRACT |
|---|
 TOP
 ABSTRACT
 INTRODUCTION
METHODS
RESULTS
CONCLUSIONS
REFERENCES
|
|---|
Five molecular dynamics computer simulations were performed on different phospholipid:sterol membrane systems in order to study the influence of sterol structure on membrane properties. Three of these simulated bilayer systems were composed of a 1:8 sterol:phospholipid ratio, each of which employed one of the sterol molecules: cholesterol, ergosterol, and lanosterol. The two other simulations were of a bilayer with a 1:1 sterol:phospholipid ratio. These simulations employed cholesterol and lanosterol, respectively, as their sterol components. The observed differences in simulations with cholesterol and lanosterol may have their implication on the form of the phospholipid/sterol phase diagram.
| |
INTRODUCTION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
CONCLUSIONS
REFERENCES
|
|---|
Cholesterol and closely related sterols play an
important role in the function of plasma membranes in most eukaryotic
cells. The incorporation of cholesterol into the phospholipid membrane usually: a) broadens and eventually eliminates gel to
liquid-crystalline phase transition of phospholipid bilayers b)
decreases (increases) the area per molecule of the liquid-crystalline
(gel) phase of monolayers c) increases (decreases) the orientational
ordering of the hydrocarbon chains in liquid-crystalline (gel) phase of bilayers (although, as was shown recently, for some lipids, such as
dimyristoylphosphatidylcholine (DMPC), cholesterol induces an increase
in the orientational order both in the gel and liquid crystalline
phases (McMullen et al., 1994
)) and d) decreases
(increases) the passive permeability of the bilayer above (below) the
main transition temperature. These effects were investigated using different physicochemical techniques (for recent reviews of cholesterol in membranes see Finegold (1993) and McMullen and
McElhaney (1996)) and computer simulations (Scott,
1991; Robinson et al., 1995; Tu et al.,
1998; Smondyrev and Berkowitz, 1999c;
Nielsen et al., 1999; Pasenkiewicz-Gierula et
al., 2000). Changes and alternations in structure of the
cholesterol molecule are responsible for some loss of the ability to
produce the above-mentioned effects. As a result, the physical
properties of the bilayer are modified. Thus, it was suggested that a
change in physical properties of the bilayer due to the replacement of
cholesterol in the membrane by another sterol-ergosterol (this sterol
can be found in the membranes of fungi, yeasts, and protozoans) may be
responsible for the difference in the interaction between membrane and
an antibiotic such as amphotericin B (Bolard, 1986;
Brajtburg et al., 1990). The structures of cholesterol
and ergosterol are displayed in Fig. 1
and, as one can see, they are not that different from each other. If
there is a difference in the specific antibiotic/membrane interaction
when sterol structure is modified, it would be of interest to
understand its physical origin.
|
In Fig. 1, the structure of another sterol molecule, lanosterol, is
also displayed. It is very close in its structure to both cholesterol
and ergosterol. Contrary to the situation with cholesterol and
ergosterol, lanosterol is not present in naturally occurring membranes,
but it is a common precursor of cholesterol and ergosterol in the
evolutionary pathway. The question is therefore: why has nature spent
thousands of years to convert from lanosterol to other two sterols? It
was suggested (Bloom and Mouritsen, 1988) that
cholesterol is designed to optimize thermodynamic and mechanical properties of phospholipid membranes. Apparently cholesterol can do
this, but not lanosterol. The difference in thermodynamics of
cholesterol/phospholipid and lanosterol/phospholipid mixtures can be
understood on the basis of the phase diagram of the sterol/phospholipid mixture. The phase diagram of a dipalmitoylphosphatidylcholine (DPPC)/cholesterol mixture in the presence of water was inferred by
Vist and Davis (1990). According to their phase diagram
a new phase, which is called liquid ordered (lo), appears when the
concentration of cholesterol is high (Xchol > 25%). This phase is characterized by a simultaneous presence of a
high degree of conformational order of phospholipid molecules and
lateral disorder. At low cholesterol concentration
(Xchol < 10%) and temperatures above the
main transition temperature, the membrane is in a phase where both
conformational and lateral degrees of freedom are disordered. This
phase is called the liquid disordered phase (ld). At temperatures below
main transition the membrane is in a gel phase where both translational
and conformational order is high, such a phase is called the solid
ordered (so) phase. The presence of the cholesterol in small amounts
only slightly reduces the temperature of the main phase transition.
Recent computer modeling data (Nielsen et al., 2000)
indicate that the phase diagram of DPPC/lanosterol is very different.
By studying the behavior of the NMR order parameter, Nielsen et
al. (2000) concluded that the most significant characteristics
of the phospholipid/cholesterol phase diagram is a stable coexistence
region between the lo and ld phases, while such a region is absent from
the phase diagram of lipid/lanosterol system. Based on NMR measurements
and theoretical modeling, Nielsen et al. concluded that the difference
in the phase diagrams is due to the difference in the interaction
strength between different sterols (cholesterol vs lanosterol) and
lipids. Since cholesterol and lanosterol have different molecular
shape, Nielsen et al. proposed that molecular smoothness is the
determining factor in the sterol/lipid interaction.
In this paper we report on a molecular dynamics computer simulation study performed on membranes containing phospholipid (DMPC) and sterols at low and high content. We study how the substitution of cholesterol with ergosterol and lanosterol influences the structural and dynamical properties of membranes with low sterol content (8:1 phospholipid:sterol ratio). In order to study the sterol/phospholipid interactions more accurately we want to perform simulations where sterol molecules are far from each other. At the same time the number of sterol molecules should be sufficiently large in order to obtain statistically significant results. This justified our choice in performing simulations at a 8:1 ratio. At high sterol content where the ratio of phospholipid to sterol was chosen as 1:1, we studied membranes with cholesterol and lanosterol.
| |
METHODS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
CONCLUSIONS
REFERENCES
|
|---|
At low sterol concentrations (11 mol %) the lipid membrane used
in our simulations consisted of 64 DMPC and 8 sterol molecules. We
performed three simulations of lipid bilayer with different sterols:
cholesterol, ergosterol and lanosterol. To set up the simulation system
we used coordinates of DMPC molecules determined by Vanderkooi
(1991). Coordinates of cholesterol molecules (Shieh et
al., 1981) and for ergosterol (Hull and Woolfson,
1976) were taken from the crystal structure. Lanosterol was
created from similar compound found in the Cambridge Structural
Database (CSD) using model builder in Spartan software package. Initial
configuration and equilibration protocol was the same as in our recent
simulation of DPPC:Cholesterol bilayer at 11 mol % sterol. Initially
four sterol molecules were distributed in each leaflet of the membrane so that the distance between them was at its maximum. We used the
united atom force field for DMPC molecules that we employed in our
previous simulations of various DPPC membranes (Smondyrev and
Berkowitz, 1999a,c,d). Lipid membranes were surrounded by 1476 TIP3P (Jorgensen et al., 1983) water molecules, which
corresponds to 20.5 waters per lipid molecule. Membranes with high
sterol concentrations (50 mol %) were composed of 32 DMPC and 32 sterol molecules surrounded by 1312 water molecules. Initially lipid and sterol molecules were placed in regular arrays, which corresponds to structure A in our recent simulations of DPPC:Cholesterol membranes (Smondyrev and Berkowitz, 1999c). Parameters for the
sterol molecules were taken from the united atom AMBER force field
(Weiner et al., 1984). Partial atomic charges were
calculated using the Gaussian 98 program at the 6-31G(d) basis set
level and the Milliken population analysis (Frisch et al.,
1998). It was found that hydroxyl group atoms and C3 carbon
atom had the largest charges. For cholesterol molecule these charges
are: 0.343e (in electron units) on the hydroxyl hydrogen, 
0.694e on
the hydroxyl oxygen, and 0.347e on C3 carbon atom. Similar charge
distributions were found for ergosterol (0.333e, 0.673e, and 0.361e)
and lanosterol (0.340e, 0.675e, and 0.350e). Charges on other carbon
atoms in sterol rings and tails were close to zero.
After initial equilibration we performed simulations on nanosecond time
scale at constant pressure (P = 0 atm) and temperature (T = 308 K) with periodic boundary conditions.
Dimensions of the rectangular simulation cell were controlled using the
Hoover barostat. Thermostat and barostat relaxation times were 0.2 and
0.5 ps, respectively. All bond lengths were constrained using the SHAKE algorithm with a tolerance of 10
4, allowing for the use
of 0.002 ps time step. The Ewald summation technique was employed in
the calculation of electrostatic contributions with a tolerance of
104. The real space part of the Ewald sum and van der
Waals interactions were cut off at 10 Å. Calculations were performed
on a Cray-T3E computer at the Texas Advanced Computing Center and an
IBM-SP computer at the North Carolina Supercomputer Center using the DL_POLY simulation package, version 2.8, developed in Daresbury Laboratory, England (Smith and Forester, 1996).
| |
RESULTS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
CONCLUSIONS
REFERENCES
|
|---|
Simulations at small sterol concentrations
Membrane geometry
Each leaflet of the membrane in our simulation contained 32 phospholipid molecules and 4 sterol molecules. In order to determine the average area per phospholipid and sterol molecule we employed the following strategy. The total volume of the membrane V is given by V = npVp + nsVs where np is the number of phospholipid molecules, ns is the number of sterol molecules, and Vp and Vs are the volumes per phospholipid and sterol molecules. We assume that the volume of a sterol molecule does not depend on the composition of the membrane, since the sterol molecule is rather rigid. The volume of the phospholipid molecule is composition dependent. For the areas we can write a similar relationship A = npAp + nsAs where the area of each compound is obtained from the relationship A = 2V/L; L is the effective thickness of the membrane. The membrane thickness can be estimated by taking the distance between phosphorus atoms in the opposite leaflets. This distance depends on the temperature, pressure, and composition of the membrane. Note that in the case of a pure DPPC membrane this kind of an estimate for the area per phospholipid results in a value of 64.8 Å2 (V = 1232 Å3 (Nagle and Weiner, 1988) and
L = 38 Å) which is in a reasonable agreement with the
experimentally determined value 62.9 Å2 (Nagle et
al., 1996) and the value of 61.6 Å2 obtained
directly from simulations (Smondyrev and Berkowitz, 1999d). In simulations with sterols, we assume that the volume of the sterol molecule is the same as the molecular volume of cholesterol which can be found from the crystal structure (Shieh et al., 1981) or (CSD) and is equal to 612 Å3. The
effective membrane thickness (distance between P atoms) of
DMPC:Cholesterol membrane obtained from our simulations is 36.4 Å (Table 1). Therefore, the effective area
per sterol molecule is 33.6 Å2. This value is close to the
value we used in our previous simulations of DPPC bilayers with
cholesterol. It is also similar to the area per cholesterol molecule
(32.4 Å2) obtained in recent simulations of the
DPPC/Cholesterol bilayer (Tu et al., 1998). Using the
estimated area per sterol molecule we can now calculate the average
area per DMPC molecule. Fig. 2 shows the
time evolution of the average areas per DMPC molecule in membranes with
cholesterol, ergosterol, and lanosterol. The average values determined
from the last 3.0 ns of simulations are: 57.9 ± 0.9 Å2 for the membrane with cholesterol, 58.4 ± 1.1 Å2 for the membrane with ergosterol, and 57.3 ± 1.2 Å2 for the membrane with lanosterol. We observe that these
three values are very close to each other. It should be emphasized that the method used to obtain this estimate is not unique. Other methods of
estimating the area per cholesterol molecule in lipid bilayers will be
described in greater details elsewhere (A. M. Smondyrev and
M. L. Berkowitz, manuscript in preparation). Thus the
values of the area per DMPC molecule should be viewed as an estimate only.
|
|
Structure of lipid bilayers
We calculated the average distances from the bilayer center to different DMPC and sterol atoms (Table 1) in order to characterize the structural properties of lipid bilayers with various sterols. These values are compared to the data for the pure DMPC membrane (Smondyrev and Berkowitz, 1999b). It is clear that
inclusion of sterol molecules into membrane results in the increase of
the membrane thickness (as evident by examining the distance from the
bilayer center to phosphorus atoms). Such an increase represents a main
manifestation of the sterol condensing effect on the membranes. The
effects of three different sterols on the membrane structure were
similar. The electron density profiles calculated for lipid bilayers
with cholesterol, ergosterol, and lanosterol (not shown here) were
almost identical at 11 mol % sterol concentration. At the same time we
found that the lanosterol hydroxyl group is located closer to the
bilayer center compared to cholesterol and ergosterol. We also measured
the average tilt of sterol molecules, which was defined as the angle
between the bilayer normal and the vector connecting carbon atoms
C3 and C17 in sterol rings. The average
cholesterol tilt in DMPC membrane (22.2°) is slightly higher than the
one observed in membrane composed of DPPC and cholesterol (20°)
(Smondyrev and Berkowitz, 1999c). DMPC membrane is
thinner than the one composed of DPPC, so that cholesterol is more
tilted in order to adjust its length to match the DMPC hydrophobic
thickness. Interestingly the average sterol tilt in DMPC bilayer
becomes higher for ergosterol (25.2°) and for lanosterol (30.1°).
The distributions of the tilt angles for three different sterols in
DMPC membrane are shown in Fig. 3. In
Figs. 4,
5, and 6
we show the time evolutions of the tilt angle for individual sterol
molecules. It is evident that lanosterol and, to a lesser extent
ergosterol can tilt significantly with respect to the bilayer normal
and such an orientation may be stable for several hundred picoseconds.
For example, as we can observe in the Fig. 6, one of the lanosterol
molecules can reorient itself in the bilayer, so that the lanosterol
angle tilt is close to (90°) and therefore the sterol molecule finds
itself in a plane parallel to the membrane surface. Such an arrangement
of lanosterol can be seen in Fig. 7,
where we display a snapshot from a simulation of a DMPC/lanosterol bilayer. As a result of the reorientations of ergosterol and lanosterol molecules the distributions of tilt angles for these sterols broadens compared to the one for cholesterol. In addition, the peaks of lanosterol and ergosterol distributions are shifted with respect to the
cholesterol distribution (Fig. 3).
|
|
|
|
|
; Urbina et al.,
1995). The ordering of hydrocarbon tails is usually
characterized by the deuterium order parameter
SCD for selectively deuterated carbon atoms,
which can be measured using the NMR technique, or by its average
quantity 
SCD
, which is directly related to
the first moment M1 of the NMR spectrum. In
DMPC/sterol membranes at 30 mol % sterol, the ordering effects of
different sterols increased in the following progression:
lanosterol-cholesterol-ergosterol (Urbina et al., 1995).
Experimental (Nielsen et al., 2000) and theoretical
(Polson et al., 2000) studies of DMPC membranes with lanosterol and cholesterol indicate that DMPC tails become more ordered
in bilayers with cholesterol compared to the ones with lanosterol at
all concentrations. At the same time this effect is more pronounced at
high sterol concentrations and more subtle at low sterol levels. In
computer simulations the deuterium order parameter
SCD can be defined using the following
expression (Egberts and Berendsen, 1988):
|
(1) |
1.5 cos
i cos j 0.5 
ij;
ij is the angle between the ith molecular
axis and the bilayer normal (z-axis). In Fig.
8 we compare
SCD values for the Sn-2 chain from our
simulations of DMPC (Smondyrev and Berkowitz, 1999d),
DMPC-Cholesterol, DMPC-Ergosterol and DMPC-Lanosterol (11 mol % sterol). As we can see the effects of three different sterols on the
ordering of DMPC tails are very similar. Similar results were also
obtained for the average number of gauche defects per lipid tail in
bilayers with cholesterol (2.8), ergosterol (2.9), and lanosterol
(2.8).
|
Sterol-lipid interactions
Nielsen et al. (2000) proposed that cholesterol,
which has a smoother molecular surface compared to lanosterol,
stabilizes the phospholipid membranes more effectively. Therefore, in
their Monte Carlo simulations, Nielsen et al. employed a simple model where the depth of a potential well that describes the interaction of a
sterol molecule with the phospholipid molecule was larger for
cholesterol. To verify thus assumption we calculated the
sterol-phospholipid interaction energy. Since this energy depends on
the location of the sterol molecule in the bilayer we calculated this
interaction energy as a function of the center of mass position of
sterol molecules and their tilt. While the average energies have only a
weak dependence on the sterol tilt, they depend on the position of the
sterol molecule relative to the bilayer center. The plots for the total
energies as a function of a distance between the sterol center of mass
and the center of the bilayer for the three sterols are presented in
Fig. 9. As we can see, the total
energy becomes more negative when sterol molecules move closer to the headgroup region. Although the energies are close to each other for a
given distance, nevertheless we observe that lanosterol/DMPC interaction has the lowest energy at any given distance. Therefore, as
Fig. 9 illustrates to be consistent with the assumption of Nielsen et
al. lanosterol molecule should be, on the average, closer to the center
of the bilayer compared to cholesterol molecule.
|
Hydrogen bonding
The largest contribution to the sterol-lipid interaction energy comes from the van der Waals forces as indicated by the energy minimization studies of Vanderkooi (1994). However,
interactions between polar groups of sterol and phospholipid molecules
may also play an important role. Experimental studies of membranes with
cholesterol and epicholesterol indicate that the configuration of the
cholesterol hydroxyl group is important for the sterol-phospholipid interactions (Murari et al., 1986). An energy
minimization simulation of DMPC:Cholesterol bilayer at 1:1 molar ratio
predicted that cholesterol binds to the DMPC Sn-2 carbonyl group via a
hydrogen bond (Vanderkooi, 1994). Several MD simulation
studies also showed the formation of hydrogen bonds between cholesterol
hydroxyl group, phospholipid oxygen atoms, and water
(Robinson et al., 1995; Tu et al., 1998;
Smondyrev and Berkowitz, 1999c;
Pasenkiewicz-Gierula et al., 2000). Existence of the
hydrogen bond between cholesterol and DMPC molecules was also predicted
experimentally using a combination of cross-polarization and NMR
techniques (Guerneve and Auger, 1995). Recent MD
simulation of DMPC/Cholesterol bilayer at 22 mol % cholesterol showed
that cholesterol can bind to phospholipid molecules via a water bridge
(Pasenkiewicz-Gierula et al., 2000). In the present
study we compared the hydrogen bonding of cholesterol, ergosterol and
lanosterol molecules in DMPC bilayers. The hydrogen bond is defined
using the geometric criteria (Pasenkiewicz-Gierula et al.,
1997): the distance between water or cholesterol oxygen and the
DPPC oxygen is shorter than 3.25 Å and the angle between the vector
linking DPPC oxygen with water (or cholesterol) oxygen and H-O bond of
the water (or cholesterol) is less than 35° as proposed by
Raghavan et al. (1992). Cholesterol can either form a
direct hydrogen bond with DMPC oxygen atoms or bind to them through a
water bridge (Pasenkiewicz-Gierula et al., 2000). The water bridge between sterol and DMPC molecules is formed when a water
molecule forms hydrogen bonds with both sterol and DMPC molecules. The
average number of sterol molecules linked to different oxygen atoms of
DMPC via direct hydrogen bonds or water bridges are listed in Table
2. Although cholesterol and ergosterol
behave somewhat similarly, cholesterol has a slightly higher tendency to hydrogen bond to DMPC compared to ergosterol. Lanosterol forms hydrogen bonds differently compared to the other two sterols. First, it
forms fewer hydrogen bonds with phosphate oxygen O11, which
is probably due to the fact that lanosterol is positioned closer to the
bilayer center. Second, lanosterol forms more hydrogen bonds with the
carbonyl oxygen O32 (these oxygens are the closest to the
bilayer center) via water bridges.
|
Sterol dynamics
Different interactions between sterols and DMPC may be also linked to variations of dynamical properties of membrane lipids. Diffusion coefficients were obtained from molecular dynamics simulations using the following equations:
|
... indicates averaging over initial times
t0. In Fig. 10
we show the mean square displacements along the bilayer normal and in the plane of membrane for three different sterols in DMPC membranes. By
fitting straight lines to these curves (in the interval from 1 ns to
2.5 ns) we obtained the following values for the lateral diffusion
coefficients: Dlat = 1.5, 3.0, and 3.2 × 107 cm2/s for cholesterol, ergosterol, and
lanosterol respectively. These values are similar to the diffusion
coefficients of lipids in pure DPPC membranes determined from computer
simulations (Essmann and Berkowitz, 1999) and
experiments (Sackmann, 1995). The lanosterol diffusion
coefficient is higher than the one for cholesterol. This was also
observed in computer simulations based on the minimal model
(Polson et al., 2001). The motions of three sterols in
the direction along the normal to the bilayer are different. In the case of ergosterol the mean square displacement saturates to a constant
value of ~4 Å2 after approximately 500 ps. In the case
of lanosterol and cholesterol, the fit to a straight line produced the
values of Dt = 0.9 × 107 cm2/s and 1.25 × 107
cm2/s for the corresponding transverse diffusion
coefficients. One should understand that the values of the diffusion
coefficients given above are rather approximate due to a small number
of sterol molecules sampled and limited time of the runs.
|
Simulations at large sterol concentrations
In lieu of some of the observed differences in the properties of DMPC membranes with different sterols such as cholesterol and lanosterol at sterol:phospholipid ratio 1:8, it would be interesting to find out whether or not such differences exist when membranes are rich in sterol. For this purpose we performed two more simulations: first on a membrane containing cholesterol and DMPC molecules at 1:1 ratio, and the second, on a membrane containing lanosterol and DMPC molecules in the same 1:1 ratio. The simulation of cholesterol:DMPC was done for 2 ns, while the simulation of lanosterol:DMPC was performed for 1.5 ns. This choice of the time lengths was dictated by the time required for the membrane area to stabilize. As we can see from Fig. 11 the area of the cholesterol:DMPC dimer stabilized after ~1.0 ns, while the area of the lanosterol:DMPC dimer stabilized after ~0.5 ns. The data analysis for both systems was performed for the last 1.0 ns of the two simulations. We found that the area per cholesterol:DMPC heterodimer is 77.5 ± 0.6 Å2, while the area of the lanosterol:DMPC heterodimer is 82.6 ± 0.6 Å2. This may be due to the fact that lanosterol molecule is bulkier than cholesterol. In Fig. 12 we show the SCD order parameter averaged over two DMPC chains. As the figure shows, lanosterol has a stronger effect on the ordering of chain molecules towards the headgroups. This can be attributed to the fact that the lanosterol molecule is not as flat as cholesterol, especially toward its head due to the presence of two extra CH3 groups. The SCD order parameter for carbon atoms towards the middle of the bilayer is larger for phospholipids in membranes with cholesterol. Overall, the condensing effect of lanosterol is weaker compared to cholesterol. The membrane thickness in the presence of cholesterol is 1 Å larger than the one with lanosterol, and therefore the condensing effect of cholesterol is stronger. In Fig. 13 we show the distributions for the angle between the sterol and bilayer normal. No substantial difference is observed between the two curves on the figure. The average angle between cholesterol and the bilayer normal 10.6° is the same within the experimental error as the angle for lanosterol 10.0°. No indication that lanosterol turns perpendicular to the bilayer normal is given by the angular distribution curve and such configurations are not seen in any snapshots obtained from simulations with a high content of lanosterol.
|
|
|
| |
CONCLUSIONS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
CONCLUSIONS
REFERENCES
|
|---|
We performed three molecular dynamics simulations of DMPC membranes with sterol at 8:1 ratio. Each simulation was done with a different sterol: the first was done with cholesterol, the second with ergosterol, and the third with lanosterol. Although we did not observe any major differences in the structure of the DMPC membranes with different sterols or in the energetics of the sterol-phospholipid interaction, still there were some differences between the behavior of lanosterol and other sterols. We observed that lanosterol on the average was closer to the center of the bilayer. We also found that the angle between the molecular axis and the bilayer normal was larger for lanosterol than for two other sterols, and we observed that one of the lanosterol molecules reorients and spends ~1 ns in a plane that is parallel to the bilayer surface. We also observed a difference in the hydrogen bonding pattern between lanosterol and DMPC. Thus, in the case of lanosterol, the hydrogen bonding between the sterol hydroxyl group and phosphate oxygen is diminished while the number of hydrogen bonds with carbonyl oxygens is increased in a manner consistent with the observation that lanosterol is (on the average) located closer to the membrane center. What can be the reason for this small difference in the location of lanosterol and other two sterols? As the structures of sterols show, lanosterol has two methyl groups attached to carbon 4, which makes the first ring of lanosterol larger in size. To fit more comfortably into the bilayer, lanosterol slides somewhat towards the bilayer center and as a result becomes more mobile (compared to cholesterol) in its orientational and lateral motion. How these differences in sterol location and mobility influence the phase diagram is a very interesting question that remains to be investigated.
We also performed two simulations with high concentration of sterol
where the ratio of sterol to phospholipid was 1:1. One of the
simulations was done with cholesterol and another with lanosterol. In
this case we observed that cholesterol had slightly stronger condensing
effect on the membrane compared to lanosterol, although the ordering of
phospholipid chains close to headgroups was larger for membranes with
lanosterol. We did not observe any major differences in orientational
properties of sterols when membranes contained large amounts of
sterols. Our simulations together with simulations performed with
simpler potential models (Nielsen et al., 2000) indicate
that indeed there are differences in the physical properties of
membranes with different sterols. Such differences may be very
important when additional molecules are present in the membranes, such
as proteins. Our present results also show that more experimental and
simulation work is needed in order to understand how the change in
sterol structure affects the properties of membranes.
| |
ACKNOWLEDGMENTS |
|---|
This work was supported by National Science Foundation Grant MCB9604585. Computational support from the National Partnership for Advanced Computing Infrastructure is gratefully acknowledged. Calculations were also performed on the IBM-SP at the North Carolina Supercomputer Center. Conversations with Prof. M. Zuckermann are greatly appreciated.
| |
FOOTNOTES |
|---|
Received for publication 31 May 2000 and in final form 13 December 2000.
Address reprint requests to Dr. Max L. Berkowitz, University of North Carolina, Dept. of Chemistry CB 3290, Chapel Hill, NC 27599-3290. Tel.: 919-962-1218; Fax: 919-962-2388; E-mail: maxb{at}unc.edu.
| |
REFERENCES |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
CONCLUSIONS
REFERENCES
|
|---|
Biophys J, April 2001, p. 1649-1658, Vol. 80, No. 4
© 2001 by the Biophysical Society 0006-3495/01/04/1649/10 $2.00
This article has been cited by other articles:
 |
|
 |
|
  N. Yoshida, T. Mita, and M. Onda Susceptibilities of Phospholipid Membranes Containing Cholesterol or Ergosterol to Gramicidin and its Derivative Incorporated in Lysophospholipid Micelles J. Biochem., August 1, 2008; 144(2): 167 - 176. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Krivanek, L. Okoro, and R. Winter Effect of Cholesterol and Ergosterol on the Compressibility and Volume Fluctuations of Phospholipid-Sterol Bilayers in the Critical Point Region: A Molecular Acoustic and Calorimetric Study Biophys. J., May 1, 2008; 94(9): 3538 - 3548. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Rog, M. Pasenkiewicz-Gierula, I. Vattulainen, and M. Karttunen What Happens if Cholesterol Is Made Smoother:: Importance of Methyl Substituents in Cholesterol Ring Structure on Phosphatidylcholine-Sterol Interaction Biophys. J., May 15, 2007; 92(10): 3346 - 3357. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Rog and M. Pasenkiewicz-Gierula Cholesterol-Sphingomyelin Interactions: A Molecular Dynamics Simulation Study Biophys. J., November 15, 2006; 91(10): 3756 - 3767. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. A. Mannock, R. N. A. H. Lewis, and R. N. McElhaney Comparative Calorimetric and Spectroscopic Studies of the Effects of Lanosterol and Cholesterol on the Thermotropic Phase Behavior and Organization of Dipalmitoylphosphatidylcholine Bilayer Membranes Biophys. J., November 1, 2006; 91(9): 3327 - 3340. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. Shahedi, G. Oradd, and G. Lindblom Domain-Formation in DOPC/SM Bilayers Studied by pfg-NMR: Effect of Sterol Structure Biophys. J., October 1, 2006; 91(7): 2501 - 2507. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. L. Stottrup and S. L. Keller Phase Behavior of Lipid Monolayers Containing DPPC and Cholesterol Analogs Biophys. J., May 1, 2006; 90(9): 3176 - 3183. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Czub and M. Baginski Comparative Molecular Dynamics Study of Lipid Membranes Containing Cholesterol and Ergosterol Biophys. J., April 1, 2006; 90(7): 2368 - 2382. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Henriksen, A. C. Rowat, E. Brief, Y. W. Hsueh, J. L. Thewalt, M. J. Zuckermann, and J. H. Ipsen Universal Behavior of Membranes with Sterols Biophys. J., March 1, 2006; 90(5): 1639 - 1649. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. B. Kim, N. Lockwood, M. Chopra, O. Guzman, N. L. Abbott, and J. J. de Pablo Interactions of Liquid Crystal-Forming Molecules with Phospholipid Bilayers Studied by Molecular Dynamics Simulations Biophys. J., November 1, 2005; 89(5): 3141 - 3158. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Naumowicz and Z. A. Figaszewski Impedance Analysis of Lipid Domains in Phosphatidylcholine Bilayer Membranes Containing Ergosterol Biophys. J., November 1, 2005; 89(5): 3174 - 3182. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. J. Tierney, D. E. Block, and M. L. Longo Elasticity and Phase Behavior of DPPC Membrane Modulated by Cholesterol, Ergosterol, and Ethanol Biophys. J., October 1, 2005; 89(4): 2481 - 2493. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. E. Beattie, S. L. Veatch, B. L. Stottrup, and S. L. Keller Sterol Structure Determines Miscibility versus Melting Transitions in Lipid Vesicles Biophys. J., September 1, 2005; 89(3): 1760 - 1768. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
O. Edholm and J. F. Nagle Areas of Molecules in Membranes Consisting of Mixtures Biophys. J., September 1, 2005; 89(3): 1827 - 1832. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
O. Soubias, F. Jolibois, S. Massou, A. Milon, and V. Reat Determination of the Orientation and Dynamics of Ergosterol in Model Membranes Using Uniform 13C Labeling and Dynamically Averaged 13C Chemical Shift Anisotropies as Experimental Restraints Biophys. J., August 1, 2005; 89(2): 1120 - 1131. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y.-W. Hsueh, K. Gilbert, C. Trandum, M. Zuckermann, and J. Thewalt The Effect of Ergosterol on Dipalmitoylphosphatidylcholine Bilayers: A Deuterium NMR and Calorimetric Study Biophys. J., March 1, 2005; 88(3): 1799 - 1808. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. Pata and N. Dan Effect of Membrane Characteristics on Phase Separation and Domain Formation in Cholesterol-Lipid Mixtures Biophys. J., February 1, 2005; 88(2): 916 - 924. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. L. McWhirter, G. Ayton, and G. A. Voth Coupling Field Theory with Mesoscopic Dynamical Simulations of Multicomponent Lipid Bilayers Biophys. J., November 1, 2004; 87(5): 3242 - 3263. [Abstract] [Full Text] [PDF]
|
|
|
|
