The need for fat

John Whitfield

It sounds like an architect's fantasy: bricks that build themselves into walls. Walls that vary from place to place, and moment to moment — in some parts nigh-on impregnable, in others little more than a curtain. The bricks move the wall's windows and doors about, and open and close them to make sure the building is correctly lit and ventilated, deliveries received and waste expelled.

Living cells are built out of these magic bricks, in the form of the lipid molecules that make up the bulk of the cell membrane's surface area. The lipids defend and define the cell's borders, keeping the things that belong inside the cell in place, and excluding those that don't. But the membrane is a border, as well as a wall — what should be inside and outside the cell is always changing, and the lipids help to regulate this traffic.

The principle behind lipid membranes' powers of self-assembly is so simple, it has become proverbial — oil and water do not mix. The most common lipid molecules in cell membranes are like two-tailed tadpoles. The tails are fatty acids, poorly soluble in water. The head is a water-loving group. A small mammalian cell will have about a billion of these molecules. The majority are phospholipids, the head of which contains a phosphate group. If you put a bunch of these lipids into water, they spontaneously cluster their water-hating tails together, pointing towards each other, while the water-loving heads point outwards. The result is a layer of two molecules that is nanometres thick — tails in the middle, heads pointing inwards and outwards (FIG. 1). It is relatively easy for the molecules to move about within their layer, but it requires much more energy for them to flip-flop between layers; in the cell there are enzymes to assist in this job.

		Figure 1 | Basic structure of the lipid bilayer. Phospholipids, the type of lipid that makes up the majority of lipids found in the cell membrane, are made up from a phosphate head (circles) that likes water and a fatty-acid, or lipid, tail (lines) that hates it. In an aqueous environment, such as that found in cells, these lipids line up so as to limit the exposure of the hydrophobic portions to water, thus forming a membrane layer.

Put one type of lipid in a test tube of water and it will make a perfectly good cell-like compartment, separating the inside from the outside with an oily bilayer membrane. Yet the membrane of a single living cell can contain several hundred types of lipid, all with differing head and tail groups. Why such diversity?

One reason is that different lipids have differing physical properties. The membrane's average consistency is similar to olive oil, but this can vary. If it gets cold, for example, the membrane can 'freeze', changing from an oil to a gel. Different lipids have different freezing points. This is partly determined by the number of double bonds in the fatty-acid tail which creates kinks in the shape of the molecule (FIG. 2). This means that they pack together less closely than straighter, saturated fatty-acid chains, and so have a lower freezing point. In cold conditions, cells such as yeast and bacteria will make more of these kinked lipids to retain fluidity in their membranes. In general, the inner layer of lipids, called the inner leaflet, contains more of these unsaturated fatty acid chains, and so is more fluid than the outer layer — making intuitive sense, as one would expect the cell's outer coat to be more robust than its inner lining.

		Figure 2 | Saturated and unsaturated lipids. The terms saturated and unsaturated lipids refer to the number of bonds on each carbon atom that makes up the fatty-acid tail of the molecule. Saturated lipids are so-called because they have single bonds between all the carbon atoms, and therefore all the carbons are bonded to the maximum number of hydrogen atoms. These chains are fairly straight and can pack closely together, making these fats solid at room temperature. Other fats have some double bonds between some of the carbons in the tail, causing the molecule to bend. As carbon atoms with double bonds are not bonded to as many hydrogens as possible, they are called unsaturated fats. The kinks in the tails mean that unsaturated fats can't pack as closely together, making them liquid at room temperature.

Other cells use lipids to stiffen their membrane. The myelin sheath that surrounds nerve cells needs to insulate the nerve axon for efficient impulse transmission. These cells' fatty acids have long chains that bond to each other, creating a tough coat. Another way to make the membrane more rigid, and less permeable, is to increase its cholesterol content. Our cell membranes are full of cholesterol — around one molecule for every phospholipid molecule. Cholesterol is smaller than phospholipids, and its structure contains carbon rings that form rigid plates; it slots in the gaps between the larger phospholipid molecules, so restricting their movement and stiffening the membrane.

As well as lipids, the cell membrane is studded with proteins — the windows and doors in our magic wall. Among other things, these function as channels that let chemicals in and out, and as sensors that trigger changes inside the cell when they latch onto molecules in the outside world. These proteins are another cause of the diversity in lipids in cell membranes; each cell has a lipid composition tailored to interact with its particular complement of proteins. Robert Cantor from Dartmouth College, Hanover, New Hampshire, has suggested that the lipid composition of membranes can control the shape of membrane proteins by creating sideways pressure. More rigid lipids will put a tighter squeeze on the proteins floating in it than a more fluid membrane. Variations in pressure could control protein shape, and therefore activity. This is still hypothetical — while it is possible to estimate the pressure across a membrane, understanding the effect that this has on proteins, at different points in the membrane, is much more difficult.

More certain is that specific types of lipid interact with specific proteins, controlling their assembly and folding, or acting as cofactors — switching the protein on by slotting into spaces on the larger molecule. A lipid called cardiolipin, for example, is found only in the membrane of mitochondria — structures found inside eukaryotic cells that have bilayer membranes of their own. Many of the chemical reactions of respiration happen in the mitochondrial membrane, and several of the enzymes that turn food into fuel need cardiolipin to function; the lipid helps to hold the components of protein complexes together, or helps the proteins interact with their environment. Some people with autoimmune diseases, such as systemic lupus erythematosus, form antibodies against cardiolipin; this can lead to severe cardiovascular and neurological symptoms.

Another class of membrane lipids are key mediators of the cell's interactions with its environment. Every cell has small quantities of inositol lipids in its membrane. These molecules participate in a hugely complex web of signalling responses, involving hundreds of different proteins, which controls just about every cellular activity, including the signals that control cell growth and movement, programmed cell death, and the transport of chemicals into and out of cells. When activated, protein receptors in the cell membrane break inositol lipids into pieces. The phosphate-containing head group might then be released away from the membrane into the cell's interior, where it binds with and activates other proteins. The lipid tail, remaining in the membrane might then be broken down further, or might itself bind to and activate proteins.

Other types of lipids have sugars instead of phosphate as their head groups, either as mono- or oligosaccharides - hence they are called glycolipids. Glycolipids are found exclusively in the outward-facing part of the bilayer. In red blood cells, the type of glycolipids present determines which combination of ABO blood groups a person has. They were discovered owing to their role in so-called storage diseases, such as Tay-Sachs disease, in which enzymes needed to break glycolipids down are lacking, and they accumulate to harmful, often fatal, levels.

Like their chemical cousins, glycolipids can perform structural roles, insulating or protecting the membrane. And they can also function, in concert with membrane proteins, as receptors — glycolipids are involved in the insulin pathway, for example. But it's not just the body's own proteins that latch onto glycolipids, some of the cell's enemies also home in on them. The HIV virus recognizes glycolipids when it locks on to immune cells. The cholera and tetanus toxins also attach to membrane glycolipids. And some antibiotics disrupt the membranes of their target cells in bacteria by targeting glycolipids. Researchers are keen to exploit this property in designing new antibacterial drugs; bacteria are less likely to mutate membrane lipids than changes in, say, a single protein in order to confer resistance to antibiotics, as membrane changes are more likely to undermine many aspects of cell function.

Once it was thought that the different protein and lipid molecules floated around each other more or less at random — what was known as the fluid mosaic model of membranes (see 'Mind the membrane'). But now most researchers believe that there is local structure within different areas of the same membrane leaflet, in the form of relatively dense patches called lipid rafts (BOX 1).

The existence of lipid rafts has not been proved beyond doubt, but there is growing evidence accumulating from microscopic and chemical studies of cells and lipids in vivo and in vitro. In particular, a type of phospholipid called sphingolipids and cholesterol tend to pack together to create dense patches. Again, proteins and lipids influence one another: some can turn the rafts into flask-like invaginations of the membrane, called caveolae. And rafts are not spread evenly around the membrane. In epithelial calls, for example, which form the body's surfaces at places such as the mucus membranes and the gut, there are more rafts in the cell surface exposed to the outside world.

Lipid rafts have been implicated in a variety of tasks, including moving cholesterol, other lipids and membrane proteins around the cell, and as points for molecules to leave and enter. Rafts are also involved in processes such as the entry of the HIV virus into the cell, as the glycolipids they attach to are raft-forming. And the rafts can serve as anchoring points for proteins, controlling their distribution in the membrane. This last feature is linked to what is probably the rafts' most important role: intercellular signalling. Many of the proteins tethered to rafts are involved in signalling, including hormonal pathways and the immune system.

The structure and function of lipid rafts are still debated, and a deeper insight into the dynamics of the membrane in general will have to await new experimental techniques that allow us to view the membranes of living cells at high temporal and spatial resolutions. But it is already abundantly clear that the lipid membrane is not just a passive bag to stop the cell's innards leaking away, but a complex and dynamic cellular organ in its own right.

		Box 1 | What are lipid rafts? Lipid rafts are sphingolipid- and cholesterol-rich membrane microdomains in the outer leaflet of the plasma membrane. The plasma membrane is composed primarily of sphingolipids, phospholipids and cholesterol. Sphingolipids differ from most phospholipids in that they have long, largely saturated acyl chains that allow them to pack tightly in a bilayer, forming a gel phase in which there is very little lateral movement or diffusion. The gel phase of the sphingolipids is altered by the association of cholesterol, which condenses the packing of the sphingolipids by occupying the spaces between the acyl chains. So, cholesterol-containing sphingolipid microdomains exist in a liquid-ordered phase that is significantly more fluid than the gel phase. By contrast, phospholipids are rich in unsaturated acyl chains that tend to be kinked and consequently to pack loosely into a liquid-disordered phase that is considerably more fluid, allowing rapid lateral movement within the bilayer. The different packing of the sphingolipids and phospholipids probably leads to their phase separation in membrane bilayers. Sphingolipid microdomains float in a phospholipid bilayer, leading to the coining of the term 'lipid rafts'. Cholesterol preferentially partitions into the liquid-ordered phase rather than the liquid-disordered phospholipid bilayer and is essential for the maintenance of the two phases. The membrane outer leaflet rafts are believed to be linked to an inner leaflet that is probably rich in phospholipids with saturated fatty acids and cholesterol. The size of rafts and their lifetimes in the membranes of resting cells are uncertain. Current evidence indicates that the elemental rafts might be small (26-70 nm in diameter), containing only several thousand molecules and therefore accommodating only a few proteins. Rafts were shown selectively to include some proteins and to exclude others, so rafts provide a mechanism for the lateral sorting of proteins in the membrane. Modified from Pierce, S. K. Lipid rafts and B-cell activation. Nature Rev. Immunol. 2, 96-105 (2002) ? Macmillan Magazines Ltd

Further reading

Edidin, M. Lipids on the frontier: a century of cell-membrane bilayers. Nature Reviews Mol. Cell Biol.. 4, 414-418 (2003).

Irvine. R. F. Nuclear lipid signalling. Nature Rev. Mol. Cell Biol. 4, 349-361 (2003).

Pierce, S. K. Lipid rafts and B-cell activation. Nature Rev. Immunol. 2, 96-105 (2002).

Simons, K. & Toomre, D. Lipid rafts and signal transduction. Nature Rev. Mol. Cell Biol. 1, 31-39 (2000).