Tissue-specific transcription is generally thought to require cooperative interactions between multiple transcription factors, as well as co-activators. But the nature of these interactions, and the mechanism of cooperativity, is typically not well understood. This project focuses on interactions between two large transcription factor families: homdomains and basic-helix-loop-helix (bHLH) factors, which are essential for beta cell development and function. The beta cell-specific insulin promoter binds the homeodomain Pdx1 and the bHLH factor E47/NeuroD1 which activate the rat insulin I promoter synergistically.

Our immediate goal is to determine the nature of the cooperativity between these two factors. Because it has been shown that the homeodomain and bHLH domain of Pdx1 and E47/NeuroD1 interact directly in the absence of DNA, we are investigating whether this interaction contributes to cooperative DNA binding by these factors.

As a step towards achieving this goal, we have determined the crystal structures of the homeodomain of Pdx1 and the bHLH domain of E47/NeuroD1 bound to DNA (below). Both structures provide insights into the flexibility inherent in the DNA binding of these protein domains.

Further crystallographic and small angle scattering studies , in conjunction with DNA binding and transient transfection assays, will be used to characterize the importance of these protein-protein interactions for determining DNA binding specificity.

The importance of understanding cooperative interactions between these transcription factors is highlighted by the effort to generate artificial beta cells to treat diabetes by manipulating the expression of Pdx1 and NeuroD1. In addition, mutations in Pdx1 and NeuroD1 cause a familial form of diabetes Maturity-onset diabetes of the young (MODY).