The transcription factor TEF-1 binds DNA via its TEA domain. The TEA domain is approximately 77 amino acids long, and has been proposed to bind to DNA via a twelve residue structure called the probe helix. A site-directional mutational analysis of this structure disproves aspects of the probe helix theory, including which residues within the probe helix interact with DNA. Crystallised models of other transcription factors, which also contain a probe helix that interacts with DNA, support these findings and expose the possible role of the probe helix as being involved in protein-protein interactions. A purified wildtype TEA domain is shown to be able to interfere with the DNA binding ability of full length TEF-1. However, proline substitutions at specific residues within the probe helix prevent this interference. Further, certain residues appear to contribute both to DNA binding and to TEA domain self-interaction. This indicates that the probe helix may be an important structural feature in TEA domains, but that this role involves both protein-DNA and protein-protein interactions. TEF-1 is known to be present in many types of vertebrate cells, and it has been shown to be a cellular target of SV40 T antigen. Nuclear extracts from TA-transformed scid and beige mouse pancreatic cells derived from primary or metastatic tumour nodules, were found to contain similar amounts of TEA domain proteins. Scid cell nuclear extracts, however, lacked a factor(s) which produced a high molecular weight DNA-protein complex in the beige mouse derived and human HeLa cell nuclear extracts. The possibility that this may be due to disruption of a DNA-dependent protein kinase complex in scid cells is discussed.