Last modified 6th April '95 Birkbeck College 1995

Back to main PPS Index

Back to Tertiary Structure Index

Globular Proteins

Most proteins which occur in the aqueous, intracellular environment or in the plasma are of globular nature: they are very approximately spherical in shape, or consist of several different lobes (domains). Generally, proteins with more than about 200 amino acid residues are multi-domain (although there are exceptions eg 2CPP). The different domains of a single protein may be responsible for quite different functions. This is taken to extremes in a number of 'modular' proteins, which are mostly extracellular and glycosylated (see section on mosaic proteins).

Individual domains may occur as `signalling' proteins in the extracellular environment in higher organisms, more usually called hormones. Examples include :-
mini-gif insulin, the growth factors, glycoprotein hormones, and the cytokines (Glasgow & Oxford).

The tertiary structure of globular proteins reflects their interaction with their aqueous solvent. At a simple level, a globular protein may be considered to consist of a hydrophobic core surrounded by a hydrophilic external surface which interacts with water. The tertiary fold of the polypeptide is such that those residues with apolar side chains are buried in the centre, while the polar residues remain exposed. This principle is held by many to be the dominant driving force behind the folding of the polypeptide chain into the compact globular form: the aggregation and burial of the hydrophobic surface reduces the number of unfavourable interactions of these groups with water; thus the hydrophobic effect. (A section on this will soon be available as part of the "Overview of Molecular Forces" Chapter.)

Compare this distribution of hydrophilic and hydrophobic side chains with that in an integral membrane protein (see the previous section).

Bear in mind however that although the concept of "hydrophobic residues in, polar residues out" generally applies, it has its limitations. Residues with hydrophobic side chains have some polar surface area (the backbone atoms), and vice versa. There will necessarily be some buried polar groups in the hydrophobic interior (generally main chain atoms involved in hydrogen bonding in secondary structures). Likewise, the amount of solvent-exposed hydrophobic surface area is far more than minimal: in fact it constitutes approximately 50% of the total exposed surface area of a globular protein. A patchwork arrangement of hydrophobic and polar surface allows solvent molecules close to apolar regions to interact favourably with neighbouring polar surfaces.

Different types of globular proteins have been outlined in the Overview section of this chapter.

Structure determination

Globular proteins are more likely to crystallize than membrane proteins. X-Ray diffraction is the technique most used, and which we use at Birkbeck. That will be the subject of `another course'.

Another important technique for elucidating protein structure is Nuclear Magnetic Resonance (NMR) Spectroscopy (see also Imperial College's NMR stuff).


Many proteins have associated ligands which are an integral part of their structure and may be essential for their function. These ligands may be water molecules (use RasMol, with Display:Ball and stick option, to find the buried water molecules in this crystal structure of Bovine Pancreatic Trypsin Inhibitor, BPTI), metal or other inorganic ions, or organic molecules.

Click here for diagram of myoglobin, an oxygen-carrying protein which contains a haem porphyrin (ring) group which has an Fe ion ligand at its centre. It is at this site that one oxygen molecule is bound. Here is the PDB structure.

The enzyme carboxypeptidase A is a zinc metallo-protease, which has a single Zn++ ion bound to its single polypeptide chain. This ion is involved in the binding of substrates and inhibitors (see the section on enzymes).Click here for a diagram of bovine pancreatic carboxypeptidase A, showing the zinc ion at the active site. (This picture, and others referred to in this section, were generated by Manuel Peitsch at GLAXO Geneva.)

Insulin exists of a trimer of dimers in a hexagonal arrangement. Each monomer consists of two chains joined by disulphide bridges. There are two forms of insulin, in which either 2 or 4 Zn ions are bound. Click here for the insulin dimer showing zinc ion binding sites.

Several examples of protein-ligand binding are available, produced by Andrew Wallace and Roman Laskowski's LIGPLOT software.

Allosteric interactions

Haemoglobin exists of a tetramer of two types of chains, alpha and beta, each of which is homologous to myoglobin, complete with haem group. There are therefore 4 oxygen-binding sites. The binding of an oxygen molecule to one of these sites triggers a change in the tertiary structure of the chain to which it has bound, which results in symmetrical changes occurring in the other subunits, increasing their affinity for oxygen. This is so that there is a sharp change in the propensity of the protein to bind oxygen in environments with a high oxygen partial pressure and release it in oxygen-poor environments. This will be examined in more detail in a following chapter. Click here for a picture of oxygenated heamoglobin showing the heam groups and iron ions.


Back to the Top

Back to Main PPS Index
J. Walshaw