Protein Structure

Crystallography and Protein Structure

This is a quick tour in the world of crystallography and protein structure. It assumes basic knowledge of chemical principles and protein chemistry.

The crystallographic exercise may be summarised in this scenario:

A compound, e.g. the form of human bile acid shown below, is known to be made up of 24 carbon, 4 oxygen and 40 hydrogen atoms. In order to understand its properties, its shape, i.e. the arrangement of its atoms relative to each other, needs to be known.

Chemical principles can quickly suggest a number of feasible arrangements, that make chemical sense and satisfy the above composition. Which one of these is the correct one remains to be determined experimentally. Many techniques are useful, but only Nuclear Magnetic Resonance (NMR) and Single Crystal X-ray Diffraction (crystallography) can produce three dimensional models. X-ray crystallography, however, is 10 to 100 times more accurate than NMR in determining the coordinates, although it is limited by the availability of the sample in a crystalline form.

In the course of a diffraction experiment, X-rays are shone on a well ordered crystal of the compound to be studied. They interact with the electronic cloud to produce diffracted X-ray beams. Because there are no lenses that can recombine these beams into a comprehensible image of the object, 'mathematical lenses' are needed instead. These are computational algorithms that can reproduce the electronic cloud of the compound.


Space Filling Model	Ball & Stick Model	Stick Model

Although the space filling represntation is the true appearance of the compound, at the atomic level, it is not always easy to visualise when viewed in two diemnsions. Simplifications have to be introduced, like displaying a small sphere at each atomic centre, with sticks joining the bonded atomic positions. A further simplification uses only sticks to connect bonded atomic sites.

Protein crystals can be prepared and studied rather like any other chemical. The same techniques are used to reproduce the shape of the protein, at the atomic level. The large number of atoms that are typically found in proteins, makes even the visualisation of simple stick models rather incomprehensible, as can be seen in the case of the daffodil lectin, which has 858 atoms.


Space Filling Model	Stick Model	Ribbon Model

Proteins being a long continuous chain of peptides that are bonded head-to-tail, it is a reasonable simplification to ignore the 'side chains' of the individual amino acid residues, and concentrate on the main chain atoms in order to understand the pattern of non-bonded interactions. There are three main categories:

Alpha helices tend to coil round and round in a cylindrical fashion, where the oxygen atom on residue i makes a hydrogen bond with the nitrogen atom of residue (i+3) or (i+4). Either interaction produces an alpha helix, which can be tens of residues long. If a smooth line is drawn through the central atoms, the result would appear like a spring.

Space Filling Sticks & Labels Sticks Ribbon
Beta sheets are formed by sideways intearctions between the main-chain O and N atoms from different parts of a protein structure. They can cover consecutive stretches of the peptide chain or even very distant parts of the chain, as long as they come close together 'geographically'. Their interactions resemble web threads between the fingers of one hand. If smooth lines are drawn through the central atoms, the result would be a series of 'picket fence' shapes. They can be flat or curved, rather like sheets of paper.

Space Filling Sticks Ribbon
Random loops is the collective description of the parts of the chain that do not fall in either one of the above two categories. They are usually represented as a smooth rope connecting the various structural elements in the protein. In the above ribbon representation of the beta sheet, the stretches of peptide chain linking the different strands (coloured blue) are random loops.

Many proteins are quite large, comprising thousands of atoms, and many of them are also made up of several copies of the same or slightly diffrent peptide chains that are necessary for the biological function of that protein.


Bovine F1 ATP Synthase. This work won the Nobel Prize for Chemistry, 1997	Bacterial Light Harvesting Protein

A solid representation of such proteins would be meaningless, because the observer would only see one solid mass of electron density spheres. Even smooth stylised ribbon representations might be overwhelming, but that only reflects the complexity of the life processes involved. Thanks to the Synchrotron Radiation Source at Daresbury, and the highly intense and well colimated beams of X-rays it produces, the structural determination task is made much more easy.

Another Respresentaion of the F1 ATP Synthase Structure

Scientists from across the UK, and from overseas, have been using the Protein Crystallography facilities at Daresbury for many years, and they are regarded as highly efficient and user friendly. However, PX staff were astounded to see that their efforts have been noted in other parts of the universe, and that 'stars' were flocking to the Laboratory to witness the successful activities and the vibrant atmosphere.


The Moon and Comet Hale-Bopp Visiting Daresbury Laboratory	Daresbury Tower Receiving a Visit from Comet Hale-Bopp, March 1997


Space Filling	Sticks & Labels	Sticks	Ribbon