We still had to find out what the structure was, there was no sequence, there was no amino acid sequence and no... it hadn't been cloned. Trying to recall exactly what happened. But it was a 40 kilodalton protein. Now it was known from Brown's work that it's bound to a region of about 50 base pairs. That's quite a long region DNA for 40 kilodaltons is 340 amino acids and this bound to 50 base pairs which is a really long stretch, it's about 100... more than a 150 angstroms long. So this had to be some kind of linear program. Now Brown, that's Don Brown, had shown that this 40 kilodalton protein could be broken down by adding proteolysis. Was quite common in those days to look for domains in proteins and you could chop off a 30 kilodalton domains which still had DNA binding and... so I got Jonathan to start doing systematic proteolysis experiments to see if we could reduce the length of the protein and still get binding. So he actually did it systematically, we used trypsin and chymotrypsin and the... I was looking for... what I was looking for was some kind of domain structure. I thought it might be made up of lumps of protein something of that sort, but one day I noticed on his desk, because he never cleared up his desk, a lot of gels as a result of those experiments, I could see that you could see that you could break things down to a rather small amount, which will be about three kilodaltons, but there were little tell-tale bands in-between on the way down. So he told me he'd spoken to Hans Christian Thorgerson, the protein chemist... he said you always get that when you do protein digest but I said they looked to me pretty regular so I said, 'Plot them, measure them and plot them out on the log scale' and it was clear that they were three kilodalton intermediates and a three kilodalton final product. So this meant the thing is made up of repeating domains, repeating domains and if you... if you look at the 30k then you see it was clear that there would be about ten domains of three kilodaltons each. And so I was pretty sure it had to be a repeating structure of some kind and also the repeating structure would explain how it could bind to a long stretch of DNA, there would be repeating units. So then we were stuck at this stage and I was waiting and Brown and Roeder were working on this, Brown particularly, and then the sequence was cloned by Roeder and the sequence was published. I didn't know this but Jonathan Miller and Hans Christian Thorgerson had advanced knowledge of this sequence and they played around. When the sequence was published I looked at the sequence and by eye I could see repeating units. And repeating units had histidine and cysteine in it. They're common ligands for zinc, remember I told you they had that. They're also ligands for iron and for copper. So... it was pretty easy to interpret, I went to Andrew McLachlan who was a computer theorist and put it onto the computer and did computer searches and showed that there indeed... of the 340 bases... about 300 of them, just under 300, actually about 280, could be put onto... into... sorted out into nine repeating units of about 30 amino acids each and each one contained two cysteine's and two histidine's. And that's exactly what you need to kill zinc because zinc is tetrahedrally coordinated. I mean I knew enough chemistry for that, it was fortunate knowing some chemistry as that had come into play, that's ordinary classical chemistry that I'd learnt, the ligands for zinc, so it fitted very well. So we wrote a paper, published in EMBO journal in 1985, the work was started in '82. Now while this was going on other people were working on all this. A group working in transcription, Hanas and Bogenhagen and so on published about there were two zincs they had founded. This was published just about the same time as us. But they... just a little bit before. We already knew that... there would be two zincs, you looked at their preparations and they were using EDTA so it was a... the reason they were using rather small quantities of EDTA and so... and we had nine different ones and Ray Brown also, who'd been in the lab, was now working in Heidelberg, also... had studied with a man called Patrick Argos did you know them when they were in Heidelberg? Beg your pardon, Patrick Argos?

[Q] Yes, Argos.

But who's group where they in?

[Q] Well, I think at that stage Chris Sander...

Chris Sander, yes. Well, they also, they were simply doing computer analysis so the... we sent our paper in to the EMBO Journal and I know for certain that John Tooze showed them the paper and they quickly wrote up a paper, their paper, which is just purely amino acid analysis and they had twelve repeating units, they hadn't understood... because they hadn't understood about the zinc binding, they took Hanas's result about two... because they hadn't understood so they assumed that the repeating units were like the florets, petals of the flower, and the zinc was putting all these at the edge, they hadn't any real idea that these things were linear at all. Anyway, they published their paper... the reason why I dwell on this because they sometimes regard themselves as co-discoverers of zinc fingers but they had no... they hadn't done really anything much, just computer analysis. And they indexed it, they indexed it on the 39-repeating unit but the abstract says 30. Published in FEBS Letters, so... so... I went to a meeting, 1985 that year, this was published about, just 20 years ago and I heard Roeder speaking when I was there, he said we've been sitting on this sequence for a whole year and this fellow Klug, that's what he said, he didn't know I was there, told us what it was. It's... I think it's very simple, it's because Roeder and Brown brought up as molecular... and biochemists don't know any chemistry. And later on, this was repeated, because when zinc was found another class, which were later called zinc finger proteins by Ron Evans, he actually made a model in which the zinc was accorded with six ligands, and of course zinc is not, it doesn't make six ligands... so there's some virtue in knowing some chemistry.