IDENTIFICATION OF A PAX PAIRED DOMAIN RECOGNITION SEQUENCE AND EVIDENCE FOR DNA-DEPENDENT CONFORMATIONAL-CHANGES

Pax genes encode a family of developmentally regulated transcription f actors that have been implicated in a number of human and murine conge nital disorders, as well as in tumorigenesis (Gruss, P., and Walther, C. (1992) Cell 69, 719-722; Hill, R., and van Heyningen, V. (1992) Tre nds Genet. 8, 119-120; Chalepakis, G., Tremblay, P., and Gruss, P. (19 92) J. Cell Sci. Suppl. 16, 61-67; Maulbecker, C. C., and Gruss, P. (1 993) EMBO J. 12, 2361-2367; Walther, C., Guenet, J. L., Simon, D., Deu tsch, U., Jostes, B., Goulding, M. D., Plachov, D., Balling, Il., and Gruss, P. (1991) Genomics 11, 424-434; Barr, R. G., Galili, N., Holick , J., Biegel, J. A., Rovera, G., and Emanuel, B. S. (1993) Nature Gene t. 3, 113-117). These genes are defined by the presence of an evolutio narily conserved DNA binding domain, termed the paired domain. The str ucture and the DNA binding characteristics of the paired domain remain largely unknown. We have utilized repetitive rounds of a polymerase c hain reaction-based selection method to identify the optimal DNA bindi ng sequences for the Pax-2 and Pax-6 paired domains. The results sugge st that the paired domain family of peptides bind similar DNA sequence s. Identification of this binding site has revealed an important struc tural clue regarding the mechanism of paired domain binding to DNA. CD and NMR structural analyses of the purified Pax-6 paired domain revea l it to be largely structureless in solution. Upon binding the recogni tion sequence, the complex becomes markedly less soluble and displays CD spectroscopic evidence of significant alpha-helical structure.