Biology Reference
In-Depth Information
It is not an easy task to understand the variety of forces driving and maintaining partial
lipid asymmetry
[15
17]
. For example, lipid asymmetry must be severely altered at
membrane segments that are tightly curved. In order to fit the enormous quantity of internal
membranes into the tiny intra-cellular space (discussed in Chapter 1), there must be countless
places in internal membranes that are under tremendous curvature strain. These locations are
difficult to study since the strain is instantly released upon breaking open the cell. Therefore
membrane curvature is best studied by making spherical lipid vesicles of known diameter
and hence strain. The tightest vesicle that is achievable with phospholipids is ~200
˚
in diam-
eter. These vesicles, referred to as SUVs (small unilamellar vesicles, see Chapter 13), are char-
acterized by having two-thirds of their lipids in the outer leaflet and one-third crunched into
the inner leaflet. In mixed-lipid SUVs, the larger head group lipids (PC, SM, glycolipids) tend
to accumulate in the less densely packed outer leaflet while lipids with smaller head groups
(e.g. PE, cholesterol) are forced to accumulate in the more compact inner leaflet. If SUVs
undergo a series of fusions, the curvature strain is released and the membrane becomes
essentially planar in the new large unilamellar vesicles (LUVs, see Chapter 13). In the
LUVs, lipid asymmetry is lost.
As one example, if a mixed lipid PC/PS SUV is made at pH 7.0, PC is predominantly
found in the outer leaflet, while PS is more prevalent in the inner leaflet. At pH 7.0, PC is
larger than PS. However, if the pH is increased, PS flips to the outside and PC moves to
the inside. At high pH PS becomes larger than PC, due to negative charge repulsion of the
PS head group. At pH 7.0, PS has a net charge of
e
1, while at high pH the charge of PS is
2. The problem with phospholipid size is far more complex than this. One must keep in
mind that the size of a phospholipid is also dependent on the nature of the attached acyl
chains. The phospholipid base diameter increases with the number of double bonds (see
Chapter 11). One would therefore expect that lipids with more double bonds would be forced
into the outer leaflet in membrane segments with high curvature. But there is an additional
complication, as lipids with more double bonds are more 'compressible' than lipids with
saturated chains (discussed in Chapter 11). In sharp contrast, cholesterol is almost totally
non-compressible. The net effect of all of
these forces is unknown and at present
unpredictable.
Lipid partial asymmetry was first worked out in the erythrocyte and the results are
depicted in
Figure 9.11
. If the erythrocyte is transformed into a spherical 'ghost' upon emul-
sion in hypotonic media (the erythrocyte is 'hemolyzed'), much of the intracellular cytoskel-
eton is lost. The ghost still has lipid asymmetry, but it is much reduced compared to the intact
erythrocyte. If the erythrocyte lipids are extracted and used to make an LUV, no lipid asym-
metry is observed. If the LUVs are sonicated to produce SUVs, lipid asymmetry is once again
established. However, this lipid asymmetry does not resemble either that observed for the
intact erythrocyte or the erythrocyte ghost, but instead reflects curvature strain.
D. LATERAL DIFFUSION
A key aspect of the FluidMosaic model is 'fluidity'. The model predicts that the membrane
is in constant flux (it is fluid) with components always in the process of exchanging lateral
positions with other components. The first proof of membrane lateral mobility was the iconic
