Domains In Biological Membranes Research Articles

Lipid domains in biological membranes: Current Opinion in Structural Biology 1993, 3:475–481

There is increasing appreciation for the heterogeneity that exists in the distribution of lipid and protein components within a biological membrane. This review summarizes recent contributions concerning this concept and highlights some current ideas on the existence, formation and function of lipid domains in membranes.

Interdigitated gel phase bilayers formed by unsaturated synthetic and bacterial glycerolipids in the presence of polymyxin B and glycerol

The ability of synthetic phosphoglycerolipids with a cis mono-unsaturated acyl chain in the 2-position and a saturated chain in the 1-position of glycerol to form interdigitated gel phase bilayers in the presence of amphipathic substances was monitored using a fatty acid spin label, 16-doxylstearic acid, and phosphatidylglycerol spin label containing 16-doxylstcaric acid. These spin labels become significantly more motionally restricted in an interdigitated gel phase bilayer than in a non-interdigitated gel phase bilayer. The results indicated that polymyxin B and polymyxin B nonapeptide caused interdigitation of 1-palmitoyl,2-oleoyl-phosphatidylglycerol (POPG) and glycerol caused interdigitation of 1-stearoyl,2-olcoyl-phosphatidylcholine (SOPC), similar to their effects on disaturated lipids. The fluidity gradient present in non-interdigitated gel phase bilayers was abolished. However, glycerol did not cause POPG to become interdigitated, in contrast to SOPC. We reported earlier that there is a kinetic barrier to interdigitation of saturated PG in the presence of glycerol, in contrast to saturated PC. This barrier is even greater for the unsaturated species of PG. Furthermore, these compounds lowered the gel to liquid-crystalline phase transition temperatures of the unsaturated lipids more than of saturated lipids suggesting that the interdigitated bilayer of the former may be less ordered or less stable than that of the latter. Since polymyxin B is an antibiotic we also examined its effect on a lipid extract from the Gram-negative bacteria Pseudomonas aeruginosa in order to assess whether interdigitation might be involved in its mechanism of bactericidal or bacteriostatic effect. Polymyxin B and polymyxin B nonapeptide also caused motional restriction of a small percentage (about 13% at −2°C and 25% at −14°C for polymyxin B) of the spin label in the lipid extract at low temperatures, where the lipid is in the gel phase, consistent with formation of a small domain of interdigitated bilayer lipid. However, the degree of immobilization was less than that in the interdigitated bilayers of the synthetic unsaturated lipids. This may be a result of the heterogeneous nature of the lipids in the extract. However, it cannot be ruled out that the motional restriction of the spin label in this extract may be caused by something other than interdigitation. Thus the results with the lipid extract are less conclusive of interdigitation than for the synthetic lipids. A motionally restricted population was not detectable at higher temperatures. This study shows that a cis-double bond in the acyl chain of phosphoglycerolipids does not prevent formation of an interdigitated gel phase bilayer. However, such lipids are in the liquid-crystalline phase at physiological temperature and thus will not form interdigitated domains in biological membranes.

Intermixing of dipalmitoylphosphatidylcholine with phospho- and sphingolipids bearing highly asymmetric hydrocarbon chains

We have used high-sensitivity differential scanning calorimetry to investigate the mixing of dipalmitoylphosphatidylcholine (DPPC) with N-lignoceroylgalactocerebroside, N-lignoceroylsulfogalactocerebroside and 1-lauroyl-2-lignoceroylphosphatidylcholine. These three lignoceroyl species, whose two hydrocarbon chains are quite discrepant in length, are completely miscible with DPPC in the liquid-crystalline state. Mixtures of all three lignoceroyl lipids with DPPC show phase separation in the gel state, which is observed over a limited range of compositions (from < 10 mol% to just over 40 mol% sulfatide) in the case of N-lignoceroylsulfatide and over a wide range of compositions in the cases of N-lignoceroylcerebroside ( < 10 mol% to> 90 mol% cerebroside) and 1-lauroyl-2-lignoceroyl-PC (roughly 10 mol% to 90 mol% lauroyl/lignoceroyl PC). The extensive solid-solid phase separation observed in mixtures of DPPC and 1-lauroyl-2-lignoceroyl-PC, which show eutectic behavior, is somewhat unexpected given the similar transition temperatures of the two components but appears to reflect the ability of the lignoceroyl species to form an interdigitated gel phase. However, we find no evidence that the N-lignoceroylsphingolipids are markedly more prone to segregate laterally in PC-rich bilayers than are previously studied sphingolipid species with shorter N-acyl chains. We suggest on the basis of these results that the primary biological importance of the very long N-acyl chains found in many sphingolipids may lie in some function other than the promotion of lateral segregation of sphingolipid-enriched domains in biological membranes.