| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Biophys J, December 1999, p. 3108-3119, Vol. 77, No. 6
*Department of Physics, Texas Tech University, Lubbock, Texas 79409; #Department of Radiology, University of California at Irvine, Irvine, California 92697; and §Department of Medical Chemistry, University of Helsinki, Helsinki 00014, Finland
 |
ABSTRACT |
|---|
 TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
Our previous fluorescence study has provided indirect
evidence that lipid headgroup components tend to adopt regular,
superlattice-like lateral distribution in fluid
phosphatidylethanolamine/phosphatidylcholine (PE/PC) bilayers (Cheng et
al., 1997
, Biophys. J. 73:1967-1976). Here we have
further studied this intriguing phenomenon by making use of the
fluorescence properties of a sterol probe, dehydroergosterol (DHE).
Fluorescence emission spectra, fluorescence anisotropy (r), and time-resolved fluorescence intensity decays of
DHE in 1-palmitoyl-2-oleoyl-PC (POPC)/1-palmitoyl-2-oleoyl-PE (POPE) mixtures were measured as a function of POPE mole fraction
(XPE) at 23°C. Deviations, including dips
or kinks, in the ratio of fluorescence peak intensity at 375 nm/fluorescence peak intensity at 390 nm
(I375/I390),
fluorescence decay lifetime (
), or rotational correlation time (
)
of DHE versus PE composition plots were found at
XPE 
0.10, 0.25, 0.33, 0.65, 0.75, and 0.88. The critical values at XPE 0.33 and 0.65 were consistently observed for all measured
parameters. In addition, the locations, but not the depth, of the dips
for XPE < 0.50 did not vary
significantly over 10 days of annealing at 23°C. The observed
critical values of XPE coincide (within
±0.03) with some of the critical mole fractions predicted by a
headgroup superlattice model proposing that the PE and PC headgroups
tend to be regularly distributed in the plane of the bilayer. These
results agree favorably with those obtained in our previous
fluorescence study using dipyrenylPC and Laurdan probes and thus
support the proposition that 1) regular arrangement within a domain
exists in fluid PE/PC bilayers, and 2) superlattice formation may play
a significant role in controlling the lipid composition of cellular
membranes (Virtanen et al., 1998, Proc. Natl. Acad. Sci.
USA. 95:4964-4969). The present data provide new information
on the physical properties of such superlattice domains, i.e., the
dielectric environment and rotational motion of membrane sterols appear
to change abruptly as the lipid headgroups exhibit regular
superlattice-like distributions in fluid bilayers.
| |
INTRODUCTION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
A recent study (Cheng et al., 1997) suggested
that phospholipid headgroups of fluid binary
phosphatidylcholine/phosphatidylethanolamine (PC/PE) membranes may
exhibit regular superlattice-like domains at certain critical PE molar
fractions. Using noninvasive Fourier transform infrared spectroscopy
and steady-state fluorescence measurements based on site-specific
fluorescent probes, Laurdan and dipyrenylPC, evidence that
1-palmitoyl-2-oleoyl-PE (POPE) and 1-palmitoyl-2-oleoyl-PC (POPC)
molecules adopt superlattice arrangements at several PE mole fractions
(XPE), 0.04, 0.11, 0.16, 0.26, 0.33, 0.51, 0.66, 0.75, 0.82, 0.91, and 0.94, has been presented (Cheng et al., 1997).
Another study indicated that such superlattice arrangements could also
exist in natural membranes, and it was proposed that they might play a
crucial role in the compositional regulation of such membranes
(Virtanen et al., 1998; Somerharju et al., 1999). Notably, sterols also
seem to adopt superlattice-like arrangements in membranes (Chong, 1994;
Parasassi et al., 1995; Virtanen et al., 1995). However, more
experimental evidence on superlattice formation needs to be obtained.
The effects of superlattice formation on the location and dynamics of
membrane components also need to be determined.
In the present study, a small fixed amount of fluorescent sterol,
dehydrergosterol (DHE), was incorporated into the binary POPE/POPC
mixtures. An extensive study, including both steady-state and
time-resolved fluorescence measurements, was performed for XPE = 0.08-0.92. The aim of this study was
to determine whether these new fluorescent sterol measurements would
reveal the presence of critical compositions and whether those
compositions would agree with those predicted by the headgroup
superlattice model (Virtanen et al., 1998). In addition, the effects of
the putative superlattice formation on the location and dynamics of
sterols (as reported by DHE) in the membrane were investigated.
| |
MATERIALS AND METHODS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
Lipid membrane preparations
Lipids were purchased from Avanti Polar Lipids (Alabaster, AL) and were found to be more than 99% pure, based on thin-layer chromatography analysis. DHE was obtained from Sigma (St. Louis, MO). The stock solutions containing POPC and POPE in varying ratios were mixed in chloroform, and a small amount of DHE (1.0 mol%) was added. The samples were placed in a 42°C water bath, and the solvent was evaporated under a nitrogen stream. Any residual solvent was removed by keeping the samples under vacuum at 42°C for at least 5 h. The dry lipid films were subsequently dispersed in buffer (100 mM NaCl/10 mM N-tris-(hydroxymethyl) methyl-2-aminoethanesulfonic acid/2 mM EDTA, pH 7.4) at 35°C and under rigorous vortexing for 15 min. After more than 24 h of incubation at 4°C, the lipid samples were kept at 35°C for 30 min and then at 4°C. This temperature cycling was repeated at least three times. The samples were kept in the dark at 4°C before any spectroscopic measurements.
Laser-induced fluorescence measurements
Fluorescence spectral measurements were performed on a home-built optical multichannel analyzer equipped with a UV-enhanced proximity focused intensified photodiode array IRY-700S detector (Princeton Instrument, Trenton, NJ), which was attached to a 1/3 m SPEX Minimate 1681 spectrograph (SPEX Industries, Edison, NJ). A Liconix 4240NB cw UV He-Cd laser (Santa Clara, CA) operating at 325 nm was used for excitation. With this instrument, a single fluorescence spectrum was obtained in less than 50 ms with a wavelength resolution of 0.24 nm. Normally 20-50 spectra were accumulated, from which the peak intensity ratio was calculated. Steady-state fluorescence emission anisotropy measurements were performed on an ISS GREG 200 (ISS, Champaign, IL) fluorometer, using an L-format arrangement with excitation at 325 nm and the same He-Cd laser as above. Fluorescence emission was collected through a 380-nm low-cutoff filter.
Fluorescence lifetime measurements were performed on the same ISS GREG
200 fluorometer equipped with digital multifrequency cross-correlation
phase and modulation acquisition electronics. The same He-Cd laser was
also used for excitation. An excitation polarizer with its transmission
axis set at 35° with respect to the vertical was placed in the
excitation beam to eliminate the contribution of the rotational
diffusion effect of the sample to the measurements (Spencer and Weber,
1970). No polarizer was placed on the emission side. Phase delay and
demodulation values of the DHE fluorescence signal were compared with
that of a standard solution
(p-bis[2-(5-phenyloxzaolyl0]benzene (POPOP) in ethanol, fluorescence lifetime = 1.35 ns) and measured at different
modulation frequencies ranging from 5 to 200 MHz. All fluorescence
measurements were carried out at 23°C. The samples were equilibrated
at 23°C for at least 45 min before the measurements.
Fluorescence data analysis
In general, the time-resolved fluorescence intensity decay
[I(t)] of a fluorescent sample can be expressed
by a sum of exponential decays, i.e.,
|
(1) |
i and 
i are the fluorescence
decay lifetime and the preexponential factor, or mole fraction, of the
ith resolved fluorescence component, and n is the
total number of components. In the frequency-domain measurements, the
values of i and i can be recovered using
a nonlinear least-squares procedure. The average fluorescence decay
lifetime 
can be calculated from the equation given by
|
(2) |
|
(3) |
, a parameter
inversely proportional to the average rate of rotation, of the
fluorescent probe can be obtained from the measured average lifetime
and steady-state anisotropy r, using the Perrin
equation (Perrin, 1936),
|
(4) |
).
Lipid headgroup superlattice model
Assuming that the phospholipid headgroups are hexagonally (HX)
or rectangularly (R) arranged, the critical PE mole fractions, XHE,PE or XR,PE,
respectively, for a binary system of PE/PC with mole fraction of PE
(XPE) are obtained from the equations
|
|
(5) |
|
|
(6) |
; Virtanen et al., 1998). The
critical mole fractions thus distribute symmetrically around
XPE = 0.5. Some representative critical
mole fractions and their lattice constants (a, b) are shown
in Table 1.
|
| |
RESULTS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
Steady-state fluorescence measurements (intensity ratio from emission spectra and emission anisotropy) of DHE in POPE/POPC mixtures of varying PE contents (XPE = 0.08-0.92) were performed at 23°C. The separation between two successive PE mole fractions is 0.005 for all POPE/POPC samples. Because it is impractical to cover the entire composition region in a single experiment, three sets of samples covering the low-PE (XPE = 0.08-0.40), mid-PE (XPE = 0.40-0.60), and high-PE (XPE = 0.60-0.92) regions were prepared. In each case, three parallel (or independently prepared) samples were prepared and averaged. In addition, the data were smoothed by using a three-point running average. Time-resolved fluorescence decay measurements of DHE in POPE/POPC mixtures of varying PE contents were also performed. For each sample, more than 20 min was needed to complete one fluorescence decay measurement. Therefore, the PE mole fractions were less closely spaced (typically 0.01-0.02) than in the case of steady-state measurements. After equilibration of the samples at the measurement temperature of 23°C for at least 45 min, the steady-state and time-resolved fluorescence measurements were carried out within 2-3 h.
Low-PE region (XPE = 0.08-0.40)
Steady-state fluorescence measurements
Two major fluorescence peaks, at 375 and 390 nm, and a small shoulder at 418 nm were found in all DHE spectra. Fig. 1 A shows a few representative spectra of DHE for XPE = 0.29, 0.31, 0.33, 0.35, and 0.37. The spectra were normalized at the 375-nm major peak. As the PE content increased from 0.29 to 0.33, the intensity of the peak at 390 nm and that of the shoulder at 418 nm increased steadily with the PE composition. However, as the PE content increased further from 0.33 to 0.37, the intensities at 390 and 418 nm began to drop. The DHE spectrum for XPE = 0.37 was found to be almost identical to that for XPE = 0.31. Fig. 1 B shows the difference spectra, obtained by subtracting the normalized spectra for XPE = 0.31, 0.33, 0.35, and 0.37 from the normalized spectrum for XPE = 0.29. The rather subtle changes in the intensities at 390 and 418 nm are clearly demonstrated in these difference spectra. Although the absolute intensity of fluorescence varied among samples, the changes in the spectral features were found consistently at XPE 0.33. Because of
the well-resolved fluorescence peaks at 375 and 390 nm in all samples,
a useful spectral parameter, i.e., the fluorescence intensity at 375 nm
divided by the fluorescence intensity at 390 nm
(I375/I390), was
calculated from each spectrum measured for DHE in POPE/POPC mixtures.
Fig. 2 B shows the
composition-dependent changes of
I375/I390 in the low-PE
region. Apparent dips at XPE 0.16, 0.25, and 0.35 were observed. In addition, a deviation at
XPE 0.11 may be present but is defined
by only a single low data point and thus considered uncertain. Fig. 2
A shows the composition-dependent changes of the average
fluorescence intensity at 375 nm. Dips at
XPE 0.16, 0.25, and 0.35 were also
observed. Steady-state fluorescence anisotropy measurements revealed a
prominent dip at XPE 0.33 and,
possibly, a small dip or kink at XPE 0.10 and a broad dip at XPE 0.17 (Fig.
2 C).
|
|
Time-resolved fluorescence measurements
Fig. 3 shows the representative frequency domain data, phase delay, and modulation ratio for DHE when XPE = 0.29, 0.33, or 0.37. Both monoexponential and biexponential decay functions, corresponding to n = 1 and 2 in Eq. 1, respectively, were fitted to the frequency-domain data. Table 2 shows the recovered fluorescence lifetime parameters based on a nonlinear regression analysis. The values of the reduced
2 of the
biexponential fits were found to be smaller than those of the
monoexponential fits for all PE contents. This observation agrees with
the fact that the theoretical biexponential curves fit the frequency
domain data much better than do the theoretical monexponential curves
as shown in Fig. 3 for all PE contents. No significant improvements in
the values of reduced 2 were found when more complicated
decay functions, like triexponential or continuous distribution
(results not shown), were used. The values of the average lifetimes
of DHE were also calculated as given by Eq. 3. The value of
for XPE = 0.33 was found to be
lower than that for XPE = 0.29 or 0.37. A
similar trend was found for the single fluorescence lifetimes obtained
from the monoexponential fits and is shown in Table 2.
|
|
0.22 is indicated for the long and, perhaps, for the short , as shown in Fig. 4 A. For
the average fluorescence lifetime , a sharp kink at
XPE 0.11 and two broad dips at
XPE 0.22-0.25 and 0.32-0.36 seem to
be present (Fig. 4 C). Based on the Perrin equation (Eq. 4),
the rotational correlation time was calculated from the values of and steady-state anisotropy as given by Eq. 4. Fig.
5 A shows the
composition-dependent changes of as a function of PE content in the
low-PE region. These data are compatible with but do not clearly show
the existence of a sharp dip at XPE 0.11 and broad dips at XPE 0.22-0.25
and 0.32-0.36.
|
|
Mid-PE region (XPE = 0.40-0.60)
Steady-state fluorescence measurements
Fig. 6 B shows the composition dependency of I375/I390 in the mid-PE region. No clear dips can be observed, but a deviation at XPE 0.43-0.50 seems to be present. The
fluorescence intensities at 375 nm are shown in Fig. 6 A. No
significant deviations were observed. The fluorescence anisotropy
measurement supports a deviation at XPE 0.45-0.50, as shown in Fig. 6 C.
|
Time-resolved fluorescence measurements
Fig. 7 summarizes the fluorescence lifetime data of DHE in this mid-PE region. No significant changes were observed for the values of the resolved long and short lifetimes and the short lifetime fractions (Fig. 7, A and B). Again, no significant deviations were found for the values of
as shown in Fig. 7 C. Fig. 5 B shows the
composition-dependent changes of as a function of PE content in the
mid-PE region. Here a broad deviation at
XPE 0.46-0.53 was observed.
|
High-PE region (XPE = 0.60-0.92)
Steady-state fluorescence measurements
Fig. 8 B shows the composition dependency of I375/I390 in the high-PE region. An obvious kink at XPE 0.75 and, possibly, smaller kinks at XPE 0.65 and 0.88 are present. The fluorescence intensities are shown in
Fig. 8 A. A kink at XPE 0.75 and, possibly, a dip at XPE 0.65 are
present. The anisotropy data are rather noisy but are compatible with
kinks being present at XPE 0.67-0.70 and 0.77-0.80 and close to 0.88 (Fig. 8 C).
|
Time-resolved fluorescence measurements
Fig. 9 summarizes the fluorescence lifetime data of DHE in the high-PE region. A broad dip at XPE 0.63-0.65 was observed in the
values of long as shown in Fig. 9 A. The fraction of short indicates a broad peak at XPE 0.65-0.70 (Fig. 9 B). The average lifetime is
compatible with a broad dip at XPE 0.62-0.65 (Fig. 9 C). Fig. 5 C shows the
composition-dependent changes of as a function of PE content in the
high-PE region. A dip close to XPE 0.65 and a possible deviation close to XPE 0.80 were found.
|
Stability of dips
To determine the stability of the observed dips, low-PE samples
were kept at 23°C, i.e., the temperature of fluorescence measurements in this study, for up to 10 days. Measurements were made on the first,
third, fourth, sixth, eighth, and tenth days after preparation and
temperature cycling (see Materials and Methods). The locations and
depth of the DHE spectral ratio and anisotropy dips were determined. The depth of a dip is defined as (max 
min)/max × 100, where max and min are the local maximum and minimum within a given
local region of XPE, i.e., 0.09-0.14,
0.14-0.21, 0.21-0.30, or 0.30-0.39. As shown in Fig.
10 A, the spectral ratio
dips at XPE 0.11, 0.17, 0.24, and 0.34 were present and remained invariant within ±0.02, and the depth of all
these dips became stabilized at 2% after the third day of annealing at
23°C.
|
Similar results were obtained for the anisotropy dips at
XPE 0.10, 0.16, and 0.32 (Fig.
11). However, the depth of the
anisotropy dips remained at the level of 10% for the two small dips at
XPE 0.10 and 0.16, but increased to
30% for the prominent dip at XPE 0.32 after 10 days of annealing.
|
| |
DISCUSSION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
The present study supports our previous findings by showing
significant dips/deviations at certain critical compositions predicted by a headgroup superlattice model (Cheng et al., 1997). As was noted in
the Introduction, the existence of such a critical composition could be
intimately involved in the regulation of lipid compositions of
biological membranes (Virtanen et al., 1998; Somerharju et al., 1999).
Table 1 provides a summary of the observed composition-dependent changes in the measured DHE fluorescence parameters (steady-state intensity peak ratio, steady-state anisotropy) and the calculated DHE
fluorescence and rotational parameter (average fluorescence lifetime
and rotational correlation time ). The changes are categorized based on the appearances of kinks or dips from the above
DHE parameters versus PE composition plots for the low-PE, mid-PE, and
high-PE regions. Our data points represent averages of fluorescence
measurements from three independently prepared samples. Notably, our
fluorescence measurement using dipyrenyl probes (results not shown)
indicated that the Tm decreases from 26°C at
XPE = 1.00 to ~21°C at
XPE = 0.95. Therefore the POPE/POPC bilayers are essentially in the fluid state at 23°C for
XPE = 0.92, which is the maximum PE content
of this study.
The major dips for XPE < 0.50 appeared
reproducibly at certain compositions upon repeated measurements during
a 10-day postpreparation period. This is important because positions of
the dips should not depend on time if they indeed mark the critical
compositions predicted by the superlattice model (Cheng et al., 1997;
Virtanen et al., 1998; Somerharju et al., 1999). On the other hand, the depth of many dips appeared to change with time. We believe that the
magnitude, or depth, of a dip depends on the relative amount of regular
distributed domains and the coexisting nonregular domains. In addition,
there may have been a significant nonequilibrium component immediately
after the sample preparation, and the formation of certain superlattice
domains may require stability or at least metastability.
Dehydroergosterol (DHE), a fluorescent sterol probe, has been used
extensively when the structure and function of lipid membranes have
been studied (Schroeder, 1984; Chong and Thompson, 1986; Bar et al.,
1989; Kao et al., 1990; Liu et al., 1997; Loura and Prieto, 1997). The
structure of DHE is similar to that of cholesterol (Schroeder, 1984).
The fluorescence intensity, spectral ratio, and lifetime of DHE are
useful indicators of the dielectric constant of the
microenvironment of membrane sterols in bilayers (Schroeder et al.,
1987; Liu et al., 1997). By combining the fluorescence lifetime and
steady-state fluorescence anisotropy data, the average rotational
correlation time, which is inversely related to the rotational rate, of
DHE in the membranes can also be estimated (Liu et al., 1997). In a
recent study of sterol/PC mixtures (Liu et al., 1997), the above
physical parameters of DHE exhibited dips/peaks at several mole
fractions of sterols. Those mole fractions agree with the critical mole
fractions predicted by the acyl chain/sterol superlattice model (Chong,
1994; Virtanen et al., 1995; Liu et al., 1997).
The steady-state fluorescence intensity ratio
(I375/I390) and the
fluorescence decay lifetime of DHE are useful parameters for the study
of the quenching properties of fluorescent sterols in membranes.
Quenching of DHE fluorescence is related to with the distribution of
DHE within the membrane during its fluorescence lifetime (around a few
nanoseconds). Spectral features of DHE in media of different polarities
and lipid membranes have been studied extensively (Schroeder, 1984;
Loura and Prieto, 1997). In an aqueous medium, DHE has two major
fluorescence peaks at 402 and 426 nm. However, when DHE is in a
nonpolar medium or in lipid membranes, those peaks shift to 375 and 390 nm. In this study, normalized fluorescence and difference spectra (Fig.
1) of DHE revealed only two major peaks at 375 and 390 nm for all PE
contents. The lack of 402-nm and 426-nm peaks indicated that essentially all DHE molecules are located within the membranes. The use
of an intensified multichannel optical detection system in this study
allowed us to acquire laser-induced fluorescence emission spectra of
DHE in PE/PC membranes with reasonably good sensitivity and spectral resolution.
In general, intensity ratio measurements provide an accurate way of
studying subtle spectral changes and require no absolute intensity
determinations. Near the critical mole fractions, e.g., XPE 0.33, a decrease in
I375/I390 or an increase
in the relative fluorescence intensity at the longer wavelength may be
related to an increase in the dielectric constant of the
microenvironment of DHE. This is supported by the time-resolved
fluorescence decay measurements in which a decrease in was
observed at the mole fractions where dips in
I375/I390 were found. In
addition, fluorescence intensity of the major peak at 375 nm also
showed a minimum at those critical fractions (Figs. 2 A, 6
A, and 8 A). A previous study (Chong and
Thompson, 1986) focusing on acrylamide quenching kinetics of DHE in
membranes indicated that the decrease in DHE intensity and lifetime can
be mainly attributed to an increase in the dielectric constant of the
DHE environment in lipid membranes. Other factors may also contribute
to the changes in intensity, spectral ratio, or lifetimes. We believe
that subtle changes in the dielectric environment of DHE are the major
contributions to the observed deviations in
I375/I390 and .
The observations of dips for both
I375/I390 and
at XPE 0.25, 0.35, and 0.65 lead us to
suggest that the dielectric constant of the environment of DHE
increases when the host PE/PC membranes adopt regular distributions at
the predicted critical mole fractions (XPE = 0.25, 0.33, and 0.67).
Usually, the deviations in the
I375/I390 and
plots, i.e., at XPE 0.25 and 0.35, are
more obvious in the low-PE region than those in the high-PE region,
e.g., XPE 0.67. Some (e.g., that at
XPE 0.73 in Fig. 10 C) are
defined by only a single composition point and may thus not be taken as
an indication of the presence of critical PE composition (e.g., at
XPE = 0.75), as predicted by the headgroup
superlattice model (Virtanen et al., 1998). However, a clear kink in
I375/I390 at
XPE 0.73 was found (Fig. 9
A). Unfortunately, the spectral ratio is not as
straightforward an indicator in revealing the environment of DHE as
. Based on the
I375/I390 and
data, we propose that membrane sterols, as reported by DHE, are sensing
a more polar environment in the superlattice domains as compared
to the random domains, within the nanosecond time average. In a recent
study on sterol/PC membranes (Liu et al., 1997), a decrease in the
lifetime and intensity of DHE was found at several compositions that
agreed with the critical compositions predicted by the acyl
chain/sterol superlattice model (Liu et al., 1997).
Steady-state fluorescence anisotropy (r) measurements
provide a qualitative parameter for both the rotational motion and
fluorescence decay rate of DHE in the membranes (Perrin, 1936; Chong
and Thompson, 1986; Liu et al., 1997). Dips and kinks in the
r versus composition plots were found at several predicted
critical PE mole fractions. Similar to
I375/I390 and ,
most of those critical mole fractions, e.g.,
XPE 0.10, 0.33, 0.70, 0.77, and 0.88, were near (±0.03) the mole fractions predicted by the headgroup
superlattice model. By using the calculated data, information
on the rotational behavior of DHE can be extracted from r
based on a simple model (Perrin, 1936). The calculated rotational
correlation time is inversely proportional to the average
rotational rate of DHE in the membranes. Dips in the values of for XPE 0.11, 0.22, 0.33, 0.50, 0.65, and 0.87 were observed, suggesting that DHE rotates faster in PE/PC
membranes at these PE mole fractions, which are close to the critical
mole fractions predicted by the headgroup superlattice model, i.e.,
XPE = 0.11, 0.25, 0.33, 0.50, 0.67, and
0.86 (Virtanen et al., 1998). In most cases, the dips in the calculated
and those in the calculated occurred at PE mole fractions
close to these predicted ones (see Table 1). However, some of the kinks
and dips revealed by steady-state parameters, e.g.,
XPE 0.16 for
I375/I390 and
XPE 0.17 and 0.49 for r, were not found in the calculated parameters, and . These discrepancies may be related to the fact that the calculated parameters are less accurate than or less sensitive to the subtle structural changes in the bilayer structure than the directly measured
steady-state parameters. Moreover, the observed changes in some of the
measured steady-state parameters may be associated with factors other
than the average fluorescence lifetimes or rotational flexibility of DHE. Again, the observed changes in the values of either r
or in the low-PE region are much better defined than those in the high-PE region.
Theoretically, the critical mole fractions distribute uniformly around
XPE = 0.5 (Cheng et al., 1997; Virtanen et
al., 1998). In other words, for each critical mole fraction
XHX,PE or XR,PE in the
low-PE region, there is a critical concentration at 1 XHX,PE or 1 XR,PE
in the high-PE region. PE is the guest molecule, whereas PC is the host
molecule in the low-PE region. The roles are interchanged, i.e., PE
becomes host and PC becomes guest, in the high-PE region. This symmetry
of critical compositions appears to be found in our fluorescence
measurements, particularly for the detection of the concentration pair
at XPE 0.35 and 0.65. The difference in
the intensities of the composition deviations, i.e., dips and kinks, in
the low-PE and high-PE regions could be associated with the two
different headgroups participating in the superlattice formation as
guest or host. More theoretical and experimental investigations are
required to study this issue.
The fluorescence decay analysis may provide new information about
the heterogeneity of the DHE environment in membranes. The fluorescence
decay of DHE is known to be complex in lipid bilayers, especially in
binary ones (Liu et al., 1997). Heterogeneous decay was also observed
in the present PE/PC system. The complex decay may be related to the
intrinsically heterogeneous environment of any lipid membrane. It is
interesting to note that close to some critical mole fractions (e.g.,
0.33) this heterogeneity appears to be reduced. For example, as shown
in Table 2, the reduced 2 value of monoexponential fit
for XPE = 0.33 is significantly lower than
that for XPE = 0.29 or 0.37. This behavior
was observed consistently at most critical compositions. Furthermore,
the values of mole fraction of the short-lifetime component (Figs. 4
B and 9 B) appear to reach local maxima near some
of the predicted critical mole fractions. These observations indicate
that DHE generally senses a more homogeneous environment in the
putative superlattice domains than in the random ones, as predicted by
theory (Sugar et al., 1994).
Recently, the formation of DHE transbilayer dimers in lipid membranes
has been suggested (Loura and Prieto, 1997). It was proposed that DHE
dimer formation increases with the DHE mole fraction and results in a
decrease in steady-state anisotropy (Loura and Prieto, 1997). The
formation of DHE dimers might offer an alternative explanation of the
anisotropy dips as observed in this study. However, using a much
smaller amount of DHE (0.2 mol%), we also observed an anisotropy dip
near the major critical mole fraction at
XPE = 0.33 (result not shown). This result
indicates that DHE dimer formation may not be the key mechanism
explaining the anisotropy dips. However, further studies are needed to
determine the role of dimer formation of DHE in PE/PC bilayers.
In conclusion, this fluorescence study provides further evidence that a headgroup superlattice domain can exist in a PE/PC bilayer, as predicted by the headgroup superlattice model. In addition, the observed modulations in the various fluorescence parameters suggest that the existence of the putative superlattice domains may affect the water quenching and rotational motion of membrane sterols in fluid membranes. The annealing study further indicated that several putative superlattice domains are stable or at least metastable for several days at the temperature of measurements.
| |
ACKNOWLEDGMENTS |
|---|
This work was supported by a grant from the Robert A. Welch Research Foundation (D-1158) to KHC.
| |
FOOTNOTES |
|---|
Received for publication 1 March 1999 and in final form 8 September 1999.
Address reprint requests to Dr. Kwan Hon Cheng, Biophysics Lab, Department of Physics, Science 109, Texas Tech University, Lubbock, TX 79409-1051. Tel.: 806-742-2992; Fax: 806-742-1182; E-mail: vckhc{at}ttacs.ttu.edu.
| |
REFERENCES |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
Biophys J, December 1999, p. 3108-3119, Vol. 77, No. 6
© 1999 by the Biophysical Society 0006-3495/99/12/3108/12 $2.00
This article has been cited by other articles:
 |
|
 |
|
  H. A. Scheidt, P. Muller, A. Herrmann, and D. Huster The Potential of Fluorescent and Spin-labeled Steroid Analogs to Mimic Natural Cholesterol J. Biol. Chem., November 14, 2003; 278(46): 45563 - 45569. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Cannon, G. Heath, J. Huang, P. Somerharju, J. A. Virtanen, and K. H. Cheng Time-Resolved Fluorescence and Fourier Transform Infrared Spectroscopic Investigations of Lateral Packing Defects and Superlattice Domains in Compositionally Uniform Cholesterol/Phosphatidylcholine Bilayers Biophys. J., June 1, 2003; 84(6): 3777 - 3791. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. John, J. Kubelt, P. Muller, D. Wustner, and A. Herrmann Rapid Transbilayer Movement of the Fluorescent Sterol Dehydroergosterol in Lipid Membranes Biophys. J., September 1, 2002; 83(3): 1525 - 1534. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Wustner, A. Herrmann, M. Hao, and F. R. Maxfield Rapid Nonvesicular Transport of Sterol between the Plasma Membrane Domains of Polarized Hepatic Cells J. Biol. Chem., August 9, 2002; 277(33): 30325 - 30336. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Huang Exploration of Molecular Interactions in Cholesterol Superlattices: Effect of Multibody Interactions Biophys. J., August 1, 2002; 83(2): 1014 - 1025. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) of
the data points are also shown. Bars indicate the standard deviations
of measurements.
), and modulation ratio (
)
and short 
) (A), mole fraction of short 
), 21-30 (