J Biol Chem, Vol. 275, Issue 8, 5355-5360, February 25, 2000
Effect of Ethanol and Osmotic Stress on Receptor Conformation
REDUCED WATER ACTIVITY AMPLIFIES THE EFFECT OF ETHANOL ON
METARHODOPSIN II FORMATION*
Drake C.
Mitchell and
Burton J.
Litman
From the Section of Fluorescence Studies, Laboratory of Membrane
Biophysics and Biochemistry, National Institute on Alcohol Abuse
and Alcoholism, National Institutes of Health,
Rockville, Maryland 20853
 |
ABSTRACT |
The combined effects of ethanol and osmolytes on
both the extent of formation of metarhodopsin II (MII), which binds and
activates transducin, and on acyl chain packing were examined in rod
outer segment disc membranes. The ethanol-induced increase in MII
formation was amplified by the addition of neutral osmolytes. This
enhancement was linear with osmolality. At 360 milliosmolal, the
osmolality of human plasma, 50 mM ethanol was 2.7 times more potent than at 0 osmolality, demonstrating the importance of
water activity in in vitro experiments dealing with ethanol
potency. Ethanol disordered acyl chain packing, and increasing
osmolality enhanced this acyl chain disordering. Prior osmotic stress
data showed a release of 35 ± 2 water molecules upon MII
formation. Ethanol increases this number to 49 water molecules,
suggesting that ethanol replaces 15 additional water molecules upon MII
formation. Amplification of ethanol effects on MII formation and acyl
chain packing by osmolytes suggests that ethanol increases the
equilibrium concentration of MII both by disordering acyl chain packing
and by disrupting rhodopsin-water hydrogen bonds, demonstrating a
direct effect of ethanol on rhodopsin. At physiologically relevant
levels of osmolality and ethanol, about 90% of ethanol's effect is
due to disordered acyl chain packing.
| |
INTRODUCTION |
One of the proposed modes of action of ethanol on biological
macromolecules and membranes involves the replacement of
hydrogen-bonded water by ethanol (1, 2). Based on the hydrogen bonding
capability of both water and ethanol, it is postulated that they
compete for hydrogen bonding sites on the surface of lipids and
proteins. Thus, this action of ethanol is equally applicable to
mechanisms of ethanol action involving both lipid-mediated and direct
protein interactions. The general importance of changes in hydration in enzyme activity is widely acknowledged (3, 4), but there is little
detailed knowledge regarding how replacement of water by ethanol in
specific protein-water hydrogen bonds might alter protein
conformational equilibria or function. In addition, the physical
properties of the surrounding phospholipid bilayer are known to
modulate membrane function. Ethanol-induced changes in phospholipid
hydrogen bonding have been observed to change acyl chain packing in
pure lipid bilayers (5-7). Thus, the function of integral membrane
receptors could be especially susceptible to the disruption of
hydrogen-bonded water by ethanol. Thorough examination of the interplay
between ethanol and hydration in modulating membrane receptor function
requires the ability to separately analyze the effects of ethanol on
both protein structure and bilayer physical properties and to be able
to correlate the observed changes. Previously, we examined the effects
of ethanol (8) and a series of n-alkanols (9) on both the
MI1-MII conformational
equilibrium and acyl chain packing in the surrounding bilayer. In the
present work, the effects of ethanol on the MI-MII equilibrium and acyl
chain packing are reexamined as a function of osmotic stress.
The osmotic stress protocol (10) is an established method for
determining changes in the number of bound water molecules associated
with a specific enzymatic process. This experimental strategy is based
on the fact that the effect of a neutral osmolyte on the water activity
of aqueous compartments in equilibrium with a protein depends on the
degree to which it is excluded from the protein-associated water.
Osmolyte inaccessibility is determined by the protein topology and the
size of the osmolyte (11). The result is that by reducing the activity
of water, the presence of an osmotically active osmolyte inhibits
protein conformational changes in which there is a net uptake of water
and favors processes in which a protein releases water to the bulk
aqueous medium. Osmolytes, which are so small that they are not
excluded from any clefts or pockets on the protein surface, will not be
osmotically active (12). Recently, we employed this technique to show
that the MI to MII transition of photoactivated rhodopsin is dependent upon water activity, with the MII conformation binding 35 ± 2 fewer water molecules than MI (13).
Rhodopsin resides in the retinal ROS disc membrane, where it represents
about 95% of the integral membrane protein. About half of its mass is
within the lipid bilayer, while the other half makes up the hydrophilic
loops, connecting the 
-helices. The detailed structural information
obtained recently for both the loops (14) and the transmembrane helices
(15, 16) makes rhodopsin the most well characterized member of the G
protein-coupled receptor superfamily. Within a few milliseconds of
light absorption, a metastable equilibrium is established between MII,
the conformation which binds and activates transducin (17), and its
inactive precursor, MI (18). The MI-to-MII conformational change is the principal conformation change of photolyzed rhodopsin, and many details
of this structural change have been determined (19), including changes
in the hydrogen bonding of internal water molecules (20, 21). Previous
studies have determined the sensitivity of MII formation with respect
to both bilayer composition and phospholipid bilayer acyl chain packing
(22-24). The available structural information for rhodopsin coupled
with its central role in visual transduction and its relationship to
neurotransmitter receptors make rhodopsin an ideal receptor on which to
study the interplay between hydration and ethanol.
Ethanol and osmotic stress have opposite effects on acyl chain packing.
The ability of ethanol to disorder phospholipid acyl chain packing has
been measured in a large number of natural and artificial membranes by
several techniques (8, 9, 25). In contrast, osmotic stress, or lowered
water activity, leads to increased phospholipid acyl chain packing with
a concomitant reduction in bilayer free volume (13, 26). A complicating factor in most ethanol experiments, which obscures their relevance to
in vivo effects, is that they are generally performed on
dilute aqueous suspensions of membranes or artificial vesicles. In such experiments, the potential for in vivo osmotic conditions to
alter water activity and thereby change the effect of ethanol on
membranes is left in question. In this study, we examined the combined
effects of ethanol and osmotic stress on both the activating
conformational change of the G protein-coupled receptor rhodopsin
(i.e. the formation of MII) and on acyl chain packing of the
ROS disc membrane.
| |
EXPERIMENTAL PROCEDURES |
Sample Preparation--
Intact ROS disks were prepared from
frozen bovine retinas as described previously (27). Stachyose was
purchased from Sigma, and DPH was purchased from Molecular Probes, Inc.
(Eugene, OR). Samples for all studies (8 µM rhodopsin for
all absorbance measurements, 1.2 µM rhodopsin for
fluorescence measurements) were prepared in a low ionic strength buffer
(10 mM HEPES, 50 µM
diethylenetriaminepentaacetic acid, pH 7.5) with the required osmolyte.
Osmolyte-containing samples were then put through 10 freeze-thaw cycles
to ensure that the osmolyte had equilibrated across the membrane and
then were extruded through a 0.2-µm pore filter 10 times to reduce scattered light. All procedures were carried out under argon in a glove
box to minimize oxidation of the polyunsaturated acyl chains of the ROS
disc membrane. The osmolality of all solutions was determined with a
Wescor Vapro 5520 vapor pressure osmometer.
Equilibrium Absorbance Measurements--
Absorbance spectra of
MI-MII equilibrium mixtures ~3 s following a flash that bleached
15-20% of the rhodopsin were acquired with a Hewlett-Packard 8452A
diode array spectrophotometer (0.2-s measurements yielded <0.3%
bleach by measuring beam) (28). Individual MI and MII bands were
resolved by using nonlinear least squares to fit the sum of two
asymmetric Gaussian absorbance bands to difference spectra that had
been corrected for the presence of unbleached rhodopsin (28). The
equilibrium constant for the MI to MII conversion is defined by
Keq = [MII]/[MI]. In determining the effect
of neutral osmolytes, the ethanol-free osmolyte solution served as the
control; thus, when examining effects of the osmolytes on the
ethanol-induced MII formation, the solution osmolality does not include
the contribution due to ethanol. The change in free energy,
(G), for the MI-MII equilibrium due to the presence of
ethanol was calculated according to the following.
|
(Eq. 1)
|
The effect of osmotic stress on the MI-MII equilibrium was
analyzed using established thermodynamic relationships (10). The change
in the number of water molecules in osmolyte-inaccessible regions,
Nw, is given by the slope of line relating ln(Keq) and the osmolyte concentration as
follows.
![<UP>ln</UP>(K <UP>eq</UP>)=<UP>−</UP>&Dgr;N<SUB>w</SUB>×<FR><NU>[<UP>osmolal</UP>]</NU><DE>55.6</DE></FR>](/content/vol275/issue8/fulltext/5355/img002.gif)
|
(Eq. 2)
|
Fluorescence Measurements and Analysis--
Fluorescence
lifetime and differential polarization measurements were performed with
an ISS K2 multifrequency cross-correlation phase fluorometer with
excitation provided by an Innova 307 Argon ion laser (Coherent).
Lifetime and differential polarization data were acquired using 15 modulation frequencies, logarithmically spaced from 5 to 250 MHz.
Scattered excitation light was removed from the emission beam by a
390-nm high pass filter. All lifetime measurements were made with the
emission polarizer at 54.7° relative to the vertically polarized
excitation beam and with 1,4-bis(5-phenyloxazol-2-yl)benzene in
absolute ethanol in the reference cuvette. For each differential polarization measurement, the instrumental polarization factors were
measured and found to be between 1 and 1.05, and the appropriate correction factor was applied. At each frequency, data were accumulated until the S.D. values of the phase and modulation ratio were below 0.2° and 0.004, respectively, and these values were used as the S.D.
for the measured phases and modulation ratios in all subsequent analysis. Both total intensity decay and differential polarization measurements were repeated a minimum of three times.
Measured polarization-dependent differential phases and
modulation ratios for each sample were combined with the measured total
intensity decay to yield the anisotropy decay,
r(t). All anisotropy decay data were analyzed
using the Brownian rotational diffusion model as described previously
(29). The results of the Brownian rotational diffusion model-based
analysis were interpreted in terms of an angular distribution function,
f(
), which is symmetric about = 
/2 (29). The
extent to which the equilibrium orientational freedom of DPH is
restricted by the phospholipid acyl chains was quantified using the
parameter fv (30), which is defined as follows.
|
(Eq. 3)
|
All measurements were performed at 20 °C, and all data
analysis were performed with NONLIN (31), with all subroutines
specifying the fitting functions written by the authors. Confidence
intervals corresponding to 1 S.D. were obtained by NONLIN for both
fitting variables and derived parameters.
Bilayer Correction Factor to (G)--
The bilayer
correction to account for the effect of altered phospholipid acyl chain
packing on (G) was calculated using the known
relationship between (G) and fv
induced by varying independently both ethanol concentration and osmotic
pressure. These relationships allowed the determination of
(
G/fvEtOH) and
(G/fvOs),
respectively. Specifically, the measured changes in
fv due to the presence of osmolytes were subtracted from the observed changes in fv due to the presence
of both osmolyte and ethanol to obtain the value of
fvEtOH. This allowed the inclusion of
the enhanced effect of ethanol induced by the presence of the osmolyte.
(G)bilayer was calculated according to
Equation 4.
|
(Eq. 4)
|
| |
RESULTS |
Effect of Ethanol on MI-MII Equilibrium in the Presence of
Osmolytes--
Previous studies showed that both neutral osmolytes
(13) and ethanol (8, 9) shift the MI-MII equilibrium toward MII. In
order to compare the combined effect of ethanol and osmolytes on the
MI-MII equilibrium, the observed shifts in Keq
were converted to differences in the change in free energy in the
presence and absence of ethanol, (G) (Equation 1). In
the absence of osmolyte, increasing the ethanol concentration made
(G) more negative (i.e. ethanol shifted the
MI-MII equilibrium toward MII), as shown by the control curve for 0 sucrose in Fig. 1. The effect of ethanol was amplified by the presence of neutral osmolyte, as shown by the
curves for sucrose and stachyose (Fig. 1). Increasing osmolyte concentration reduces water activity. The effect of decreasing water
activity on the efficacy of ethanol is demonstrated by plotting (G) due to ethanol as a function of increasing
osmolality for four different ethanol concentrations, Fig.
2. The effect of 50, 100, 300, and 500 mM ethanol on the MI-MII equilibrium increased approximately linearly with increasing osmolality. Solution osmolality was varied with sucrose, except for where shown by the open
symbols, which correspond to stachyose solutions. An earlier
investigation of the effects of osmotic stress showed that the
influence of sucrose and stachyose on the MI-MII equilibrium is
equivalent at the same osmolal concentration (13). The clear trend of
decreasing (G) with increasing osmolality for both
sucrose and stachyose (Fig. 2) demonstrates that at a given ethanol
concentration, reduced water activity enhances the ability of ethanol
to promote the formation of the MII conformation.
View larger version (23K):
[in this window]
[in a new window]
|
Fig. 1.
Neutral osmolyte enhancement of the effects
of ethanol on
(G) for the MI to
MII transition. (G) was calculated according to
(G) = Gethanol  Gethanol-free control (Equation 1). Solutions
are 0.3 M sucrose ( ), 0.4 M stachyose ( ),
1.0 M sucrose ( ), and control with no osmotic agent
( ).
|
|
View larger version (23K):
[in this window]
[in a new window]
|
Fig. 2.
Osmolality enhances the effect of ethanol
on (G) for the
MI to MII transition. Filled symbols
represent sucrose solutions; open symbols
represent stachyose solutions.  , 50 mM ethanol; , 100 mM ethanol;  , 300 mM ethanol; , 500 mM ethanol.
|
|
Effect of Osmolytes and Ethanol on DPH Fluorescence--
The
effect of solute osmolality on ethanol-induced phospholipid acyl chain
disordering was assessed by analyzing the fluorescence lifetime and
anisotropy decay of DPH under the same osmolyte and ethanol conditions
used to study the MI-MII equilibrium. The DPH fluorescence lifetime was
well characterized by a decay scheme consisting of a Lorentzian
distribution plus two discrete components, which produced values of
2 between 0.9 and 2.5. The lifetimes of the two discrete
components ranged from 2.2 to 1.4 ns and from 0.27 to 0.22 ns,
respectively, and together they accounted for 10-15% of the decay
under all conditions. The high concentration of rhodopsin in the ROS
disc membrane (1 rhodopsin per 75 phospholipids) results in substantial quenching of the DPH fluorescence due to energy transfer to the retinal
chromophore. This quenching of DPH fluorescence by retinal is the
probable source of the two short lifetime components. In the absence of
ethanol, the center of the Lorentzian distribution ranged from 9.1 to
9.5 ns under all osmolyte conditions, and the maximum change was seen
in 500 mM ethanol where this value was reduced by about 0.3 ns for all osmolyte conditions.
The effect of ethanol on the decay of fluorescence anisotropy,
r(t), was also increased by the addition of
osmolyte. The Brownian rotational diffusion model characterizes
r(t) in terms of the equilibrium orientational
distribution of the probe, f(), and the rotational
diffusion coefficient, D
(29). For a free-tumbling probe
like DPH, f() is especially informative regarding acyl chain packing, because f() consists of the set of angular
orientations that DPH is allowed to adopt within the confines of the
ensemble packing of the phospholipid acyl chains. The shape of the
orientation probability distribution, f() sin,
provides information about relative acyl chain packing density at
different depths in the bilayer, and the extent of overlap of
f() sin with a random orientational distribution
indicates the degree of overall ordering of DPH by the acyl chains.
Individually, neutral osmolytes and ethanol had large and opposite
effects on f() sin, as shown by comparing the curve
for 0 osmolyte, 0 ethanol with those for 0 osmolyte, 500 mM
ethanol, and 1440 mOsm sucrose, 0 ethanol in Fig.
3A. Osmolytes narrowed the
unimodal distribution centered about the membrane normal, while ethanol
broadened the distribution, and this broadening was enhanced by the
presence of osmolyte, as shown by the curve for 1440 mOsm, 500 mM ethanol. However, the disordering effect of ethanol was
not strong enough to fully reverse the ordering effect of the osmolyte.
An example of the manner in which reduced water activity enhanced the
ethanol-induced changes in f() sin is shown by the
difference curves in Fig. 3B, which demonstrate the net
effect on f() sin of 100 mM ethanol for
four different solute osmolalities. The curves in Fig. 3B were obtained by subtracting f() sin obtained with the
specified solute osmolality in the absence of ethanol from the
distribution obtained in the presence of 100 mM ethanol at
the corresponding solute osmolality. Areas below the
zero line in Fig. 3B correspond to
angular orientations from which the DPH is excluded in the presence of
ethanol, and areas above the zero line
denote angular orientations that are preferentially allowed by the
bilayer in the presence of ethanol. Ethanol enhanced DPH orientations
greater than ~20° from the membrane normal, and an increase in the
concentration of added osmolyte amplified this effect. As the osmolyte
concentration was increased, 100 mM ethanol progressively
broadened the probability distribution centered about the membrane
normal.
View larger version (18K):
[in this window]
[in a new window]
|
Fig. 3.
A, the effects of neutral solutes and
ethanol, alone or in combination, on the DPH orientational probability
distribution, f() sin. Dashed
and dotted line, 0 osmolyte, 0 ethanol
control; dashed line, 0 osmolyte, 500 mM ethanol; dotted line, 1440 mOsm
sucrose, 0 ethanol; solid line, 1440 mOsm
sucrose, 500 mM ethanol. B, difference curves
showing how the change in f() sin due to 100 mM ethanol is increased by osmolality. Curves were obtained
by subtracting f() sin obtained with the specified
solute osmolality in the absence of ethanol from the distribution
obtained in the presence of 100 mM ethanol at the
corresponding solute osmolality. Regions below the
zero line correspond to the range of angular
orientations where ethanol reduces the DPH orientational probability
distribution, while regions above the zero
line correspond to angular orientations where ethanol
increases the DPH orientational probability distribution.
Dashed and dotted line, 0 mOsm; dashed line, 330 mOsm sucrose;
dotted line, 740 mOsm sucrose; solid
line, 1440 mOsm sucrose.
|
|
A convenient, quantitative means of comparing different orientational
probability distributions is the parameter fv (Equation 3), which is proportional to the overlap of f()
sin for a given sample and f() sin of a random
distribution (29), and thus is a measure of ensemble acyl chain order.
A higher value of fv indicates an orientational
distribution that is more like a random distribution. The separate
effects of ethanol and the reduced water activity caused by increased
osmolality on fv are antagonistic, with ethanol
causing an increase in fv (8, 9), while in the
absence of ethanol, reduced water activity lowers fv
(13). However, the net result of these two factors acting
simultaneously is that the effect of ethanol is amplified by reduced
water activity, as shown in Fig. 4. At
330 mOsm, the effect of 50 and 100 mM ethanol is increased
by about a factor of 2.5, relative to dilute buffer conditions.
View larger version (40K):
[in this window]
[in a new window]
|
Fig. 4.
Combined effects of ethanol and solute
osmolality on the parameter fv in terms of
percentage increase induced by ethanol at each osmolality.
Cross-hatched bars, 0 mOsm; gray bars,
330 mOsm sucrose; diagonally striped
bars, 730 mOsm sucrose; white bars,
1440 mOsm sucrose.
|
|
Previous investigations demonstrated a direct correlation between the
MI-MII equilibrium constant and fv, when both parameters are altered by changes in temperature or bilayer cholesterol content (22-24). In order to examine the relationship between
ethanol-induced changes in acyl chain packing and in the MI-MII
equilibrium, values of (G) and
fv due to ethanol were plotted against each other
for each solute osmolality (Fig. 5). As
the concentration of ethanol increases at each osmolality, the
(G) values become more negative and the
fv values become larger. For 0 mOsm, the
correlation is linear; however, as the osmolality increases, this
relationship becomes increasingly nonlinear. This nonlinearity suggests
that (G) is being influenced by some factor in
addition to the ethanol-induced changes in bilayer acyl chain order,
reflected in the increase in fv.
View larger version (16K):
[in this window]
[in a new window]
|
Fig. 5.
Correlation between ethanol-induced changes
in G and fv at
four different osmolalities. fv = fv(with ethanol) fv(ethanol-free control). A,
observed values of (G). B, values of
(G) corrected for the effects of ethanol and osmolyte
on acyl chain packing, as described under "Experimental
Procedures." In order of decreasing value of (G),
the symbols within each group correspond to 50, 100, 300, and 500 mM ethanol. , 0 mOsm; , 330 mOsm sucrose;
, 740 mOsm sucrose; , 1440 mOsm sucrose.
|
|
| |
DISCUSSION |
One of the important issues in in vitro studies of the
mechanism(s) of ethanol's action on biological systems is the
biologically relevant range of concentrations of ethanol (32). The
present data can be used to address this issue by considering the
osmolality-induced increase in ethanol potency observed when
measurements made in the presence of a neutral osmolyte, at the
osmolality of human plasma, are compared with the results obtained for
ethanol at 0 osmolality. For example, Fig. 1 shows that 250 mM ethanol in 400 mM stachyose produces a
(G) of 300 cal/mol, while in the absence of osmolyte
500 mM ethanol is required to produce this value of
(G). The relationship between osmolality and the
ethanol dose that is required for an equivalent effect at 0 osmolality is very linear for all of the osmolyte-containing solutions. Therefore, all of the data can be used to determine the effects of 50 and 100 mM ethanol on the MI to MII conformation change at the
osmolality of human plasma, 360 mOsm. At this osmolality, 50 mM ethanol would be equivalent to 134 ± 20 mM at 0 osmolality, and 100 mM would be
equivalent to 291 ± 12 mM. Thus, at the osmolality of
human plasma, the potency of ethanol in the 50-100 mM
range is increased by a factor of 2.7-2.9. Using this potency factor,
the substantial changes observed at 0 osmolality with 50 and 100 mM ethanol would be observed at about 18 and 36 mM ethanol, respectively, at the osmolality of plasma.
The sensitivity of rhodopsin to changes in membrane composition and
phospholipid acyl chain packing is well documented. Hence, it is
reasonable to question whether the osmolyte enhancement of the effect
of ethanol on MII formation arises through some direct interaction of
ethanol on rhodopsin or via a lipid-mediated process induced by
increased binding of ethanol to the bilayer. Earlier studies
demonstrate that factors that increase Keq
(e.g. higher temperature (22, 23, 24), increased
phospholipid acyl chain unsaturation (23, 24), decreased bilayer
cholesterol (22, 23)) also increase fv, and changes
in Keq and fv are linearly
related. Ethanol-induced changes in (G) and
fv are also linearly related in the absence of
osmolyte, as shown in Fig. 5A. Fig. 4 shows that osmotic
stress increases the ability of ethanol to loosen acyl chain packing,
as shown by the greater percentage increases in fv
with increasing osmolality for all ethanol concentrations. If all of
the ethanol-induced increase in MII formation is due to ethanol-induced
changes in acyl chain packing, then the correlation lines in Fig.
5A would be coincident, and increasing osmolality would
simply shift each group of points to higher values of
fv and more negative values of
(G). However, marked nonlinearity and
deviations from the 0 osmolyte (G) versus
fv correlation line are observed.
A possible interpretation of the nonlinearity seen in Fig.
5A is that it reflects an osmotic
stress-dependent component to the enhancement of MII
formation by ethanol, which acts in addition to the effect of ethanol
on acyl chain packing. This interpretation is consistent with the idea
that the basic mechanism whereby ethanol reduces acyl chain packing
order in the core of the bilayer is unaltered by osmotic stress. This
basic mechanism involves hydrogen bonding to the phospholipid carbonyl
oxygen, resulting in increased average phospholipid head group spacing,
which induces acyl chain disorder (33). The increased ability of
ethanol to disorder acyl chain packing with reduced water activity
(Figs. 4 and 5A) is probably due to its enhanced competition
with water for this hydrogen bonding site at reduced water activity.
The data in Fig. 5A acquired in the absence of osmotic
stress, the 0 osmolyte data, represent the correlation between
ethanol-induced changes in the MI-MII equilibrium and ethanol-induced
changes in acyl chain packing. It is similar to the linear correlation
between these two parameters when both are altered by variation in
temperature or bilayer cholesterol content (22-24). The 0 osmolyte,
(G) versus fv
correlation line can be interpreted as representing the susceptibility
of the MI-MII equilibrium to changes in acyl chain packing due to ethanol. This allows the (G) versus
fv correlation to be used to identify the
fraction of the magnified effects of ethanol on
Keq, which are due to changes in acyl chain
packing. The slope of the 0 osmolyte, (G)
versus fv correlation line indicates
that an ethanol-induced increment in fv of 0.01 will
decrease G by 510 cal/mol. By using this relationship, we
can examine to what extent the ethanol-induced decreases in G under osmotic stress are attributable to changes in
acyl chain packing and thus mediated by the membrane. The extra
(G) component under various osmotic conditions
corresponds to the deviation of those points on the
(G) axis from the 0 osmolyte correlation line in Fig.
5A. The corrected (G) values are shown in
Fig. 5B. The results of this analysis show that under all
osmotic conditions and ethanol concentrations examined, the effect of
ethanol on G is greater than what would be expected from
the ethanol-induced increases in fv, as summarized
in Fig. 6. The bars in Fig. 6
represent the fraction of ethanol's effect on the MI-MII equilibrium
that exceeds the membrane-mediated effect of ethanol. We would assign
this component of (G) to a direct effect of ethanol on
rhodopsin. Near physiological osmolality, at 330 mOsm, the effect of 50 mM ethanol on the MI-MII equilibrium is ~90% due to
changes in acyl chain packing and ~10% due to a direct effect of
ethanol on rhodopsin (Fig. 6). At all concentrations of ethanol, the
fractional contribution of the direct effect of ethanol on rhodopsin
increases with osmotic stress. This suggests that osmotic stress
enhances the ability of ethanol to act directly on rhodopsin in a way
that favors MII formation. It is also observed that at each osmolality
the nonbilayer contribution generally decreases with increasing ethanol
concentration.
View larger version (44K):
[in this window]
[in a new window]
|
Fig. 6.
Percentage of ethanol-induced
(G) that is in
excess of the decrease in G predicted by
the (G) fv correlation line in Fig.
5B for 50 mM ethanol (gray
bars), 100 mM ethanol (diagonally
striped bars), 300 mM ethanol
(white bars), and 500 mM
ethanol (cross-hatched bars).
|
|
One of the proposed mechanisms for the effect of ethanol on receptors
is a change in hydrogen-bonded water (1, 2). In the absence of ethanol,
at 20 °C, an examination of the MI to MII conversion as a function
of osmotic stress showed that MII formation results in the net release
of 20 ± 1 water molecules. This number increased to 35 water
molecules when corrected for bilayer effects (13). If ethanol perturbs
the hydrogen bonding of water to either MI or MII, it should alter the
number of water molecules released upon MII formation. The effect of
ethanol on Nw, the change in the number of water
molecules in osmolyte-inaccessible regions, was assessed by analyzing
ln(Keq) as a function of osmolality for each
concentration of ethanol according to Equation 2 (10). The positive,
linear correlations in Fig. 7 demonstrate
that ethanol has not altered the basic relationship between increased
osmolality and increased MII formation that was observed in the absence
of ethanol. The values of Nw derived from the
data in Fig. 7 are 25 ± 3 for 50 mM ethanol, 30 ± 3 mM for 100 mM ethanol, 31 ± 3 for
300 mM ethanol, and 33 ± 4 for 500 mM
ethanol. In our earlier studies (13), adjusting (G)
for bilayer contributions yielded a corrected number of water molecules
released upon MII formation. Since both ethanol and osmotic pressure
are known to have opposing effects on bilayer acyl chain packing,
adjusting (G) for the effect of these agents on the
bilayer will again yield corrected numbers of released water molecules.
This correction was carried out using Equation 4, as described under
"Experimental Procedures," and results in a net bilayer correction
that makes (G) more negative for all osmolalities and
ethanol concentrations. This demonstrates that the acyl chain ordering
due to increased osmolality outweighs the disordering effect of
ethanol, resulting in a net inhibitory contribution of the bilayer to
MII formation. Plotting the corrected values of
ln(Keq) versus osmolality yields about 49 molecules released upon MII formation for all ethanol concentrations studied. Here, we see that the apparent variation in
numbers of released water molecules as a function of ethanol concentration is a reflection of the mixed effect of osmotic pressure and ethanol on bilayer acyl packing and not an inherent property of the
system. The increase in Nw with added ethanol suggests that ethanol can influence a second population of loosely bound water molecules in MII. Ethanol may replace all or part of this
second population of waters in the rhodopsin hydration sphere. This
represents a direct effect of ethanol on MII formation, independent of
its effect on the phospholipid bilayer. The direct effect of ethanol
does not appear to be present in the absence of osmolyte.
View larger version (22K):
[in this window]
[in a new window]
|
Fig. 7.
Relationship between
ln(Keq) and mOsm used to
determine change hydration between the MI and MII conformations,
according to Equation 2, for solutions containing 50 mM
ethanol (), 100 mM ethanol (), 300 mM
ethanol (), and 500 mM ethanol (). Values of
osmolality do not include the osmotic contribution of ethanol. However,
including the osmolality of ethanol would move each set of
symbols uniformly to the right and would have no
effect on the slopes of the correlation lines and hence would not alter
the derived values of Nw.
|
|
TOP
ABSTRACT
INTRODUCTION
To whom correspondence should be addressed: Park Bldg., Room 158, 12420 Parklawn Dr., Rockville, MD 20852. Tel.: 301-594-3608; Fax:
301-594-0035; E-mail: litman@helix.nih.gov.
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
