lipid

plasma membrane

lipid raft

hydrogen bond

London forces

bilayer

micelle

packing parameter

lipid polymorphism

hexagonal phase

The famous Francis Crick’s adage, “If you want to understand function, study structure,” does not apply to lipids. Observe and dissect the structure of phosphatidylcholine (PC) as long as you want, you will not understand how plasma membranes are organized and function. The reason is that lipids do not act alone but cooperate to form ordered molecular assemblies that have acquired specific functional properties. Here the whole is not only more than the sum of its parts, the whole creates the functions the parts totally lack. In first approximation, this unique feature can be compared to a brick (a single lipid molecule) in a wall (here representing the membrane, interestingly historically referred to as the cell wall). To begin, we will keep in mind this rough analogy while explaining why lipid molecules can self-assemble into various architectural motifs.

In biochemistry, when one molecule interacts with another one to form a complex, great attention is given to the chemical groups that are physically involved in the interaction. These chemical groups constitute the so-called “binding site,” a three-dimensional domain of a molecule that is specifically devoted to the binding of another one. Such binding sites should be readily accessible to the molecular partner with which it is promised to interact. Let us consider the interaction between two proteins circulating in a biological fluid. Each of these proteins displays a binding site at its periphery. In the water environment in which these proteins circulate, these peripheral binding sites are first occupied by water molecules. When both proteins come in close contact, bound water molecules leave the binding domains so that the proteins can interact through noncovalent bonds (Fig. 2.1).

In this respect, the binding process can be decomposed in two phenomena: (1) the destructuration of numerous water molecules that were initially associated with each binding site through a network of hydrogen bonds, and (2) the formation of a new set of noncovalent interactions between the binding sites of each protein partner.1 Both mechanisms contribute to the energy of association required to form the complex. We have learned from thermodynamics that the Gibbs free energy change associated with a chemical reaction (i.e., ∆G) takes into account both mechanisms. On one hand, the departure of numerous water molecules from each binding site generates a significant molecular disorder, quantified as an entropy increase (∆S). Disorder is energy, because going back to the order state requires energy (e.g., teenagers who let the disorder progressively develop in their bedroom know how hard it is to clean it). On the other hand, the noncovalent bonds that are formed between both binding sites contribute to the whole energy of interaction, because you would need energy to break them. This enthalpic contribution of the binding is expressed by the ∆H term of the classical equation ∆G = ∆H – T∆S where T is the absolute temperature (expressed in Kelvin, allowing ∆G to be expressed in joules per moles or more representatively in kJ mol–1). It is clearly beyond the scope of this book to dwell further on this famous thermodynamic equation. Let us just remark that the free energy change ∆G of any spontaneous reaction must have a negative value. In this respect, the sign of ∆H indicates whether the binding process is chiefly enthalpy-driven or entropy-driven. Schematically, enthalpic binding results from specific molecular interactions, such as hydrogen bonds that require a perfect adjustment of both ligands (∆H < 0). In contrast, entropic binding is far less specific because it relies primarily on the disorder created in water molecules surrounding the ligands. In practice, “pure” enthalpic or entropic binding does not exist because both contribute to the binding reaction. Indeed, whatever the sign of ∆H (positive or negative), it is clear that a significant value of the entropy change ∆S ensures that ∆G is negative and high, which is particularly critical when ∆H is positive. In this case ∆G will be negative only if the entropic term ∆S is sufficiently high, which will occur if there is enough disorganization of water molecules (Fig. 2.1). Moreover, even for binding reactions with a negative ∆H, it is important that the binding process is associated with the displacement of a maximal number of bound water molecules, which will further increase ∆S and thus ∆G. This information gives a realistic idea of the entropic contribution to binding reactions. Lipid–lipid interactions will not escape this important contribution of water molecules, and as we will see next, water will even do more.

Because lipids are mostly apolar molecules, they are not particularly inclined to accommodate water molecules. However, this statement can be viewed as a tautology because lipids are precisely defined on the basis of their insolubility in water (see Chapter 1). In fact the situation is more complex because in addition to their large apolar, water-insoluble domain, lipids also display a polar head group. Hence, lipid behavior is typically amphipathic, their apolar part exhibiting hydrophobic properties that contrast with their hydrophilic polar head group. Before going further, one should understand that the terms apolar and hydrophobic, and similarly polar and hydrophilic, cannot be used as synonyms. Apolar refers to an intrinsic property of chemical group due to the weak electronegativity of its atoms (chiefly C and H). Because an apolar group cannot form hydrogen bonds with water molecules, it does not interact with water and hence it will have a hydrophobic behavior. On the opposite, a polar group contains electronegative atoms (O, N) often linked to hydrogen (–O–H and –N–H groups), that, due to the...