 Hi, I'm Lorna Wilkinson-Wile, and I'm Jacqueline Matthews, and this little clip is to tell you about our paper that's about to come out in protein science. The paper's entitled Gata 1 directly mediates interactions with closely spaced pseudo-palantromic, but not distantly spaced double gata sites on DNA. And our co-authors are Crystal Lester, Nina Rippen, David Jakes, and Metro Gus. A few basics about gata 1 is that it's a transcription factor that's very important for red blood cell development. It binds to gata sites on DNA, predominantly via the C finger, and it combines double gata sites using both the N finger and the C finger, which in this presentation are coloured blue and green respectively. Now just note here that each of these fingers has a table region, which also comes into the story. One of the things that we know about gata 1 is that it mediates chromatin looping, and that means it can bring together distantly spaced sites on DNA, so promoters and enhancer elements, bring them physically together, and through that mechanism somehow allow transcription to take place. What we did in the paper was to solve the crystal structure of gata 1 bound to DNA. In this case the protein involved is just the N and C fingers, including the tail regions of both domains, and the DNA contains a pseudo-palindromic double gata site, and that's the site that's most frequently bound by, or one of the sites that's most frequently bound by gata in the genome. If you look at this pictorially, what you can see is that the tail region of the N finger is not very well defined. It appears to be quite flexible. It's not in our structure, whereas the tail region of the C finger binds into the minor group of DNA, and it allows that finger to bind quite tightly to DNA. So that part of the story is quite straightforward, and in fact the structure is very similar to a structure of gata 3 bound to an essentially identical site that was determined a few years ago. One of the things that came out of that particular study was the possibility that gata 3 might be directly mediating chromatin looping by allowing the C finger to bind one site and the N finger to bind another site on DNA. And we wanted to set about to see if this was also the case for gata 1. Now when we did binding studies with this double zinc finger region, we could see that the N finger did contribute to binding, but only fairly weakly. It was only a little bit stronger than binding to DNA than just the C finger alone. But things became a little bit more interesting when we started looking at the contributions of just the N finger or just the C finger. Under low salt conditions, both of these fingers bound quite well to DNA. The C finger bound much more strongly, but the N finger bound moderately well as well. The difference came in when we started adding in salt, and we were bringing in salt up to much more physiological conditions. We suddenly say that although the C finger binds quite tightly, but the N finger basically loses the ability to bind DNA at all. What we also see is that the C finger combines some of those variate gata sites as well. So what we think is happening in terms of chromatin looping is not that the, for gata 1 anyway, that not that the N and the C finger can independently mediate interactions with DNA. What we think is going on is that the N finger is binding to protein binding partners. Some of those can actually self-associate to mediate the long-range interactions and that the gata components are more responsible for locating those other protein complexes to DNA at the distantly placed sites. So to summarise, we think that the N and the C finger of gata 1 can be responsible for mediating close-range interactions on DNA at things like these pseudo-palindromic sites. But when it comes to long-distance interactions, so chromatin looping, the C finger seems to be responsible for binding to DNA, but the N finger is much more important for binding to other proteins. And it's those protein-protein interactions that will then mediate the actual chromatin looping event.