Sealed membrane systems are a defining feature of cellular life. Membranes provide a barrier between the cell and its external environment and, in many organisms, divide the interior of the cell into functionally distinct compartments. The barrier, comprising lipids that are impenetrable to electrically polarized molecules, has proteins inserted within it that allow the selective transport of ions and molecules. These proteins enable cells to ingest nutrients, excrete metabolic waste, sample the environment for the sake of the immune system, and store energy by means of ion electrochemical gradients. They mediate molecular signalling across the barrier. And they are the very devil to study.

Genome sequencing projects have highlighted the central role of membrane-linked processes in cells. They have revealed that membrane proteins represent about a third of the gene products in most organisms. Unfortunately, our molecular knowledge of these membrane proteins lags far behind that of proteins found in the cell cytoplasm and in external environments. This is primarily due to the difficulty in obtaining high-resolution structural information on which to build a mechanistic understanding. For example, the purification of membrane proteins for structure determination requires them to be removed from their native membrane environment using detergents. This renders the proteins less stable.

Notwithstanding this and other technical obstacles, isolated successes in the determination of membrane protein structures were reported as early as 1977 for bacteriorhodopsin, the light-powered ion pump in the membranes of archaebacteria, and 1983 for the photosynthetic ‘bacterial reaction centre’. However, it is only in the past five years that significant numbers of membrane protein structures have been determined, including structures of ion channels, most components of the mitochondrial and photosynthetic electron transfer chains, and proteins that mediate the transport of small molecules across membranes.

The prospects for membrane biology are bright, not only thanks to technical breakthroughs but also because of a sense of adventure.

Given this progress, we can now be said to be entering the golden age of membrane protein structure. A flavour of the excitement of membrane proteins can be obtained in many of the articles in the Insight on membrane biology in this issue of Nature (see page 577) and by reading the landmark paper on the aquaporin structure embedded in a lipid bilayer (see pages 633 and 569).

But what are the reasons for this recent explosion in membrane protein structures? It is partly driven by technological innovation — the availability of microfocus synchrotron beamlines suitable for data collection from small crystals, advances in the ability to express membrane proteins to high levels and, increasingly, the use of high-throughput screening methods.

In order to obtain crystals one needs to stimulate cells, often of a different organism, to generate, or ‘express’, large amounts of the required protein. Difficulties in expressing eukaryotic membrane proteins remain the most significant bottleneck in the field. Expressing such proteins in bacterial cells has not been achieved, for reasons that are unclear, and researchers are turning to alternative systems such as insect and yeast cells to obtain their proteins of interest. But this remains a challenge: milligram quantities are typically required. These challenges in expression, coupled with the difficulty in obtaining diffraction-quality crystals, mean that there is no guarantee of success. It can take several years, in many cases longer than the standard three-year postdoctoral contract, so embarking on such a quest is a risky business.

Despite the difficulties, there is one area of structural biology where membrane proteins are at the cutting edge: protein structure prediction. Membrane protein structures should be easier to predict than those of water-soluble proteins because the structural possibilities are constrained by the membrane environment.

The prospects for membrane biology are bright, not only thanks to technical breakthroughs but also because of a sense of adventure. The field's success is due in part to the willingness of scientists to dedicate their careers to this challenging endeavour.