Delineation of a 16 amino acid sequence that forms a core DNA recognition motif in the Epstein-Barr virus EBNA-1 protein.

EBNA-1 is the sole virally specified protein required for replication of the Epstein-Barr virus latency origin and also modulates its own expression. Both of these functions are mediated through specific DNA binding. We created a series of amino acid substitutions across the previously identified DNA-binding domain of EBNA-1 to further define the amino acids required for DNA recognition. The results of electrophoretic mobility shift assays using in vitro-translated EBNA-1 (aa 408-641) polypeptides indicated that: (i) The predicted alpha helical segment between aa 477 and 487 is not directly involved in DNA recognition but appears to contribute to a critical local polypeptide conformation. (ii) The positively charged residues Arg459, Lys460, and Lys461 are dispensable for DNA binding. (iii) The region between Gly462 and Lys477 contains residues important for DNA recognition. This region is extremely sensitive to mutation. The behavior of a synthetic peptide representing EBNA-1 aa 458-478 supported the mutagenesis data. A dimer form of this peptide, which lacks the predicted alpha helical domain, was capable of binding DNA, but only nonspecifically. The effect of binding affinity on transactivation was examined in cotransfection assays. EBNA-1 mutants with reduced binding affinity also demonstrated reduced levels of transactivation.