[Next] [Previous] [Top]

No Title

2.0 Introduction to Terminology

To appreciate how much information is contained in a three-dimensional structure of a protein at atomic resolution, one needs only to display one where all atoms are displayed. While such detail is essential for some applications, a simplified representation is often more useful in grasping the essential features under investigation. Consider the example of triosephosphate isomerase. The protein has the following primary sequence:

APRKFFVGGNWKMNGDKKSLGELIHTLNGAKLSADTEVVCGAPSIYLDFAR

QKLDAKIGVAAQNCYKVPKGAFTGEISPAMIKDIGAAWVILGHSERRHVFGES

DELIGQKVAHALAEGLGVIACIGEKLDEREAGITEKVVFEQTKAIADNVKDW

SKVVLAYEPVWAIGTGKTATPQQAQEVHEKLRGWLKSHVSDAVAQSTRI

IYGGSVTGGNCKELASQHDVDGFLVGGASLKPEFVDIINAKH

The three dimensional structure has been determined by X-ray crystallography and is shown below:



Figure 1. Various representations of a protein three-dimensional structure. The X-ray crystal structure of triosephosphate isomerase (1TIM) shown in the same orientation (clockwise from upper left) using all 3711 atoms (including hydrogen) in CPK format, using all 988 backbone heavy atoms N, CA, C, O in wire frame format, using all 247 CA atoms, and in cartoon format produced with Molscript (Kraulis, 1991). Download 1TIM.PDB

In the top left corner, all 3711 atoms (including hydrogens added to the model) are displayed using the Corey, Pauling, Koltun (CPK) representation. Atoms are displayed as spheres with a diameter proportional to the atomic radius of the element. The coloring is as follows: hydrogen, white); carbon, grey; nitrogen, blue; oxygen, red; and sulphur, yellow. Considerable detail of the protein fold is revealed by removing the sidechains and displaying only the backbone heavy atoms N, CA, C, and O', retaining the CPK color scheme (top right). The secondary structure elements - helix and sheet can be readily identified using standard algorithms (4.1). The secondary structure and fold are very often further highlighted by displaying only CA atoms and connecting them with virtual bonds (lower left). Certain common structural features such as helical and extended chain conformations become apparent (highlighted here in color). These are the basic secondary structural elements identified alpha helices and beta strands (colored green and red, respectively). Schematic drawings (by hand) or cartoons were made popular by Jane Richardson (Richardson, 1981) before the widespread use of graphic workstations. Now very good computer software programs like Molscript (Kraulis, 1991), RasMol (R. Sayle, 1995), MidasPlus (Ferrin et al., 1988), and Ribbons (Carson, 1991) to name a few, are available to produce cartoons (as well as most other graphic representations) of a protein given the three-dimensional coordinates.

Most authors today (e.g., see Cantor & Schimmel, 1980; Creighton, 1993) consider disulfide bonds as well as any post-translational modifications as part of the covalent structure of the polypeptide and classify them as part of the primary structure. To put us all on the same footing, I will propose the following working definitions of structural levels. Perhaps a consensus definition can be hammered out in discussion sections.


No Title - 31 MAY 96
written by Kurt D. Berndt, Karolinska Institute, Stockholm, Sweden
[Next] [Previous] [Top]

VSNS Main Index

Generated with CERN WebMaker