Most gene therapy vectors in current use drive the expression of the therapeutic gene using viral promoters which are active in all cell types. Such unregulated expression could represent a significant safety concern for some transgenes. In the future, gene therapy vectors will become more sophisticated incorporating regulatory DNA with the therapeutic gene with the goal of producing protein only in the correct cell type at the right time in development. In order for this goal to be achieved, we will need a detailed knowledge of how genes and transgenes are regulated. Drosophila is an excellent system to study gene expression in vivo because transgenic lines are easily made, the generation types are quick and most of the principles learned in these experiments will be applicable to other organisms. We have discovered a fragment of regulatory DNA from the Drosophila engrailed gene which has two unusual activities when included in transgenes: 1) it causes transposons to insert in the genome in a non-random manner 2) it exerts an unexpected silencing activity on the mini-white transgene: it represses or even eliminates mini-white expression when the animals are homozygous for the transgene insert (called """"""""pairing-sensitive silencing""""""""). We do not know whether the same protein(s) mediate both phenomenon. We are beginning to thoroughly characterize a pairing-sensitive silencer, to identify the proteins which interact with it, to try to determine which protein(s) are responsible for the silencing activity and to determine whether the same protein(s) mediate selective insertion of transposons. We have characterized one of the proteins which binds to the pairing-sensitive silencer. This protein is homologous to the mammalian transcription factor YY1, a protein which acts as both a negative and positive regulator of transcription, dependent on the context. YY1 plays a role in the expression of many different mammalian genes. We have identified the gene in Drosophila which encodes a YY1 homolog as pleiohomeotic (PHO), a member of the Polycomb-group of genes. Polycomb-group genes are known to be important in both Drosophila and mammalian development for keeping genes silenced in places where they should not be expressed. To date, no Polycomb group genes were known to encode sequence-specific DNA binding proteins. Thus our findings represent a significant advance in this field. In recent studies we have extended our finding to show that PHO binding sites are required for the function of a Polycomb response element from the Ultrabithorax gene. In addition to PHO, our laboratory has identified 2 other proteins required for pairing-sensitive silencing. We believe that these three proteins act together to recruit a repressive Polycomb-group protein complex. Gene silencing is still poorly understood, and may be important in gene therapy of the future. Our studies on the pairing-sensitive silencers of engrailed should lead to an increase in understanding of gene silencing. .
