1 00:00:00,920 --> 00:00:04,440 Can you start by saying your name and your various hats? 2 00:00:04,680 --> 00:00:14,149 Okay. So I'm David Stewart and I am joint head of the Division of Structural Biology in the Nuffield Department of Medicine and Dentistry. 3 00:00:14,150 --> 00:00:26,340 Be known as tribute. Yes, that's right. And I'm also so I'm for quite a long time, I've been significantly funded by the Medical Research Council. 4 00:00:26,340 --> 00:00:34,890 So they actually fund a large part of my salary as a MLC professor and that's 50% of my time. 5 00:00:35,580 --> 00:00:45,570 And then the other 50% of my time is life science director at the Diamond Synchrotron, which is half an hour down the road from me. 6 00:00:46,080 --> 00:00:51,430 Um, and being the life science director at Diamond for quite a long time, 7 00:00:51,840 --> 00:01:01,499 over ten years and until the end of last year, until the end of 2021, I was also the director of Instructor Eric, 8 00:01:01,500 --> 00:01:07,500 which is a pan-European structured biology research infrastructure, which I wasn't formally paid for, 9 00:01:07,500 --> 00:01:16,379 but because because the UK is a member country, my contribution was part of the MLC contribution to to instruct. 10 00:01:16,380 --> 00:01:19,410 Eric I stopped doing that at the end of 2014. 11 00:01:19,710 --> 00:01:22,890 Hm hmm. Well, I think we're going to talk about all of those things. Okay. 12 00:01:23,100 --> 00:01:28,589 But can we start by just very briefly telling me how you got to where you are now is 13 00:01:28,590 --> 00:01:33,660 starting from your earliest interest in biological sciences and then structural biology? 14 00:01:34,980 --> 00:01:38,760 Okay. Well, that's sort of quite long, but we need to keep it short. 15 00:01:39,990 --> 00:01:53,220 So I started off my first year in biophysics at King's College in London, and I became interested at that point in methods for looking at structures. 16 00:01:54,660 --> 00:01:59,190 And I, uh, Maurice Wilkins was the head of department. 17 00:01:59,280 --> 00:02:03,450 Um, and, you know, I discuss things with him as well. 18 00:02:03,450 --> 00:02:07,110 I was interested in the sort of psychology science as well, but, 19 00:02:07,500 --> 00:02:17,010 but I decided to do a PhD and did that at Bristol University and that was in X-ray crystallography. 20 00:02:17,100 --> 00:02:23,940 So looking at the structure of proteins at that point, there weren't a huge number of proteins, structures that had been done. 21 00:02:24,390 --> 00:02:28,680 Um, and I worked on the structure of an enzyme glycol, 22 00:02:28,680 --> 00:02:37,740 lytic enzyme and the basic sort of enzymes that manages the metabolism of cells and tissues throughout biology. 23 00:02:39,270 --> 00:02:44,790 So I did that for my Ph.D. Then I was at a loss really what to do. 24 00:02:45,210 --> 00:02:54,840 Um, I went to, came here to Oxford for a short period, but during that period I was organising to go and work in China, 25 00:02:55,860 --> 00:03:00,050 and I was supported in that by um, Dorothy Hodgkin and David Phillips. 26 00:03:00,570 --> 00:03:08,400 Um, and then so I went out to work in Beijing on getting structural biology in that case, 27 00:03:08,760 --> 00:03:16,320 mainly looking at two versions of insulin insulin with the insulin group in China that had very strong links with Dorothy. 28 00:03:16,740 --> 00:03:23,309 Um, in Oxford being that was in the early 1980s. 29 00:03:23,310 --> 00:03:29,250 So there wasn't that there wasn't the same sort of social media that there is now. 30 00:03:29,760 --> 00:03:37,530 So I was lucky that the end of that time I was offered a postdoc to return to Oxford by Liz Johnson. 31 00:03:38,160 --> 00:03:46,650 Um, so I came back and started with Louise and then, uh, after a relatively short time, 32 00:03:47,040 --> 00:03:54,180 I applied for a, uh, a lectureship that came up in Oxford and got a permanent position. 33 00:03:55,110 --> 00:04:02,550 And when I was starting my group, I, I sort of had to decide what to do. 34 00:04:03,570 --> 00:04:09,960 And it was a time when there was some developments in virology. 35 00:04:10,650 --> 00:04:19,860 There was an excellent virology group at the institute, which was, yeah, it's not that far from Oxford. 36 00:04:20,220 --> 00:04:23,520 And the person leading that group contacted David Phillips and said, Look, 37 00:04:23,520 --> 00:04:30,180 we're trying to make a better vaccine, will make a useful vaccine for for next disease virus. 38 00:04:30,500 --> 00:04:32,309 And we think the structure would be interesting. 39 00:04:32,310 --> 00:04:41,400 And David Phillips directed me to work with them, which was, uh, very sort of generous and he was very supportive of that. 40 00:04:42,090 --> 00:04:47,610 So that moved me and gave me the opportunity to do something which at the time was quite ambitious, 41 00:04:47,610 --> 00:04:51,750 which was to do the structure of a complete and complete virus. 42 00:04:52,560 --> 00:04:59,790 Um, so that, that, that put me into the sort of direction of doing. 43 00:05:00,360 --> 00:05:01,850 Structural virology. 44 00:05:03,120 --> 00:05:15,330 And in the early years, I worked on foot and mouth disease virus and also worked on it because at the time also there was the HIV epidemic came about. 45 00:05:15,390 --> 00:05:27,420 So the MLC set up an H directed programme of research and we collaborated with the pharmaceutical sort of branch of Wellcome because they, 46 00:05:29,160 --> 00:05:33,870 they were doing some, they produced the first of the antiviral drugs. 47 00:05:33,870 --> 00:05:37,259 So we were collaborating with them on the structure of the reverse transcriptase, 48 00:05:37,260 --> 00:05:41,730 which was the primary drug talked to that time along with the protease. 49 00:05:42,810 --> 00:05:59,129 So, so I got into trying to, to, to help people working in industry on drug design and got into working with the people that also in industry. 50 00:05:59,130 --> 00:06:05,250 Frank Brown and Dave Robbins, welcome. With the idea of using structure to help in vaccine design. 51 00:06:05,920 --> 00:06:10,260 So let's just unpack that a little bit. I mean, first of all, very basic question. 52 00:06:11,100 --> 00:06:16,680 Why is it important to know about the three dimensional structure of something like a virus or a protein? 53 00:06:21,550 --> 00:06:28,630 Well, there's there's a fundamental there's a fundamental biology and the science to this. 54 00:06:29,350 --> 00:06:37,690 There's two things there. And the. David Baltimore got the Nobel Prize for discovering retroviruses many years ago. 55 00:06:37,690 --> 00:06:46,540 And what his what he said was that the fantastic thing about viruses is that they are biology sort of stripped down to the minimum. 56 00:06:47,050 --> 00:06:53,650 So you can argue about whether they're alive or not, but they what they do have is the they have an evolutionary history. 57 00:06:53,650 --> 00:06:54,580 So in that sense, 58 00:06:54,580 --> 00:07:04,030 they're really the sort of simplest embodiment of life so that you can see Darwinian sort of evolution work and in the case of the pandemic, 59 00:07:04,030 --> 00:07:12,400 in sort of real time, you know, so that the thing is that the genome is very small. 60 00:07:12,490 --> 00:07:17,640 So in RNA virus, most RNA viruses are not very accurate. 61 00:07:17,680 --> 00:07:23,979 That polymerase that copies their genome makes errors so they can only support a relatively simple genome. 62 00:07:23,980 --> 00:07:30,969 So there's a small number of proteins that a lot of the complications of cell life they don't do. 63 00:07:30,970 --> 00:07:38,140 They don't make their own energy and stuff like that. So they they're so I know from my perspective, 64 00:07:38,140 --> 00:07:44,170 they're a really interesting system because you can have this sort of crazy idea that you can sort 65 00:07:44,170 --> 00:07:49,780 of understand them in a in more in a more complete way than you would a very complicated cell. 66 00:07:50,650 --> 00:07:57,969 So that's the sort of fundamental science. And then, of course, there's there's also the the sort of health impact. 67 00:07:57,970 --> 00:08:08,770 And and that's human and animal health. And the to understand the basic science of something like a virus structure. 68 00:08:09,490 --> 00:08:21,700 I think everybody recognises now that that that's a sort of key aspect to hang your understanding on and for the therapeutic side to say the things. 69 00:08:22,000 --> 00:08:31,479 Really there's sort of two aspects, major aspects of think of the vaccines and then the therapeutics are the small molecules or, or biologics. 70 00:08:31,480 --> 00:08:38,050 And in a sense, both of those structures should be important. 71 00:08:38,200 --> 00:08:47,950 And for small molecule drug design, the the argument is accepted really by the drug companies. 72 00:08:47,950 --> 00:08:53,410 All the major drug companies have a a large amount of activity in the area. 73 00:08:53,800 --> 00:09:04,120 And Steve Bury, the head of the protein database in Rutgers in the US, did an analysis and I can't remember the exact numbers, 74 00:09:04,120 --> 00:09:11,740 but well over half of the drugs coming to market now have used structure at some point in their development. 75 00:09:12,070 --> 00:09:15,879 And this is just because you need to you've got two things that have to interact. 76 00:09:15,880 --> 00:09:24,520 Yes. So you can try to try to design a small molecule that's going to bind to a protein, usually a protein target, and stop it working. 77 00:09:25,090 --> 00:09:29,530 So, you know, in the case of the HIV, 78 00:09:29,770 --> 00:09:39,070 we've got the the key enzyme reverse transcriptase that's responsible for allowing that the virus to incorporate into the host genome. 79 00:09:39,820 --> 00:09:50,649 So you have little chemical pieces that are like the chemical fragments that the virus uses to make replicate the genome, which bind, 80 00:09:50,650 --> 00:09:56,260 but then inactivate the enzyme so that the whole sort of basis of that success is 81 00:09:56,260 --> 00:10:03,430 based on the physical chemical recognition of of a large molecule by a small compound. 82 00:10:05,290 --> 00:10:10,959 And so you could you can imagine you would you could do it you can do it all by chance. 83 00:10:10,960 --> 00:10:12,340 But it is very inefficient. 84 00:10:12,850 --> 00:10:26,590 And beyond that, you know, in principle, what happens with viruses, because because as I say, the error rate of the polymerase is very high, 85 00:10:27,010 --> 00:10:36,550 the virus is able to mutate very quickly, an escape from antibodies or escape from small molecule therapeutics very often. 86 00:10:36,880 --> 00:10:42,460 And that's that's one of the things that with the HIV story, we find that extremely quickly. 87 00:10:42,850 --> 00:10:50,590 And and that that was something that was I was very impressed with was the way that even though welcome, you know, 88 00:10:50,620 --> 00:10:56,450 they they were they they produced a drug and they thought that they were hoping to sell this and and 89 00:10:56,460 --> 00:11:01,150 and hoping for it to be effective as soon as they discovered and they discovered it very quickly, 90 00:11:01,150 --> 00:11:08,139 because the resistance comes up very quickly in tissue culture, they were completely open about it and made all the information available. 91 00:11:08,140 --> 00:11:18,129 So but that escape, again, is explicable in terms of the simple, simple physicochemical processes. 92 00:11:18,130 --> 00:11:22,040 There's a. Station in the in the viral protein. 93 00:11:22,400 --> 00:11:23,150 And, you know, 94 00:11:23,150 --> 00:11:32,750 you might have a bigger protein SIDECHAIN that's that's put in there by the mutation that then blocks the binding of the small molecule drug. 95 00:11:33,140 --> 00:11:43,490 So one of the questions that we were looking at with with welcome and the one of the students that was looking at this was Andrew Hopkins, 96 00:11:43,490 --> 00:11:48,560 who's now the CEO of Ciencia, one of the local drug companies. 97 00:11:48,920 --> 00:11:59,659 Is can you can you build into your compound design a way of reducing the chance that the virus can escape from it? 98 00:11:59,660 --> 00:12:05,989 And it's not a trivial question. It's still an open question. But in principle, if you have the structural information, you've got the tools, 99 00:12:05,990 --> 00:12:10,760 then to start to think about those questions so that so the structure is, 100 00:12:10,970 --> 00:12:21,590 is important for small molecules and the for the same with antibodies, you know, biological therapeutics essentially. 101 00:12:21,590 --> 00:12:25,850 The but then, you know, we'll probably discuss this later. 102 00:12:25,850 --> 00:12:36,200 One of the things we want you to understand is what how the virus changes to avoid being avoid recognition 103 00:12:36,200 --> 00:12:44,870 by antibodies that the that the people generate and you know that again it's a very simple thing. 104 00:12:44,870 --> 00:12:50,779 There's a there's a there's a recognition of the of the viral protein by the antigen. 105 00:12:50,780 --> 00:12:59,420 And if you make a mutation in the viral protein, if it's well if it's well designed, prevent the antibody binding and then the vaccine side. 106 00:13:00,710 --> 00:13:14,330 Because my my feeling from the beginning was that you should be it should be possible to improve the properties of the the the antigen, 107 00:13:14,330 --> 00:13:19,700 which is the vaccine to to make more successful vaccines. 108 00:13:19,700 --> 00:13:25,790 And there's various ways that you can go about doing that. You can get rid of the genetic material from a vaccine. 109 00:13:26,630 --> 00:13:30,950 So you're not producing vast amounts of live virus, so it becomes safe. 110 00:13:32,360 --> 00:13:40,429 And that then also reduces the requirement that that the antigen is capable of supporting life in a sense. 111 00:13:40,430 --> 00:13:48,290 So you can make changes to its that stabiliser to making it a better vaccine but would be incompatible with it being a replication virus. 112 00:13:48,560 --> 00:13:56,000 So the the recognition sites tend to be on the sort of outside envelope with the virus, whereas the genetic material is on the inside. 113 00:13:56,000 --> 00:13:58,700 Yes. So you could that. Exactly. You can throw that away. 114 00:13:59,090 --> 00:14:06,080 And in principle, if you do it right, it doesn't affect the what is seen by the immune system at all. 115 00:14:06,800 --> 00:14:11,540 And we should just very quickly go with the tools that you use. 116 00:14:11,540 --> 00:14:20,510 So you started out an X-ray crystallography and also the also involved with electron microscopy. 117 00:14:20,810 --> 00:14:26,060 And these are ways of trying to see things that are smaller than the wavelength light. 118 00:14:26,450 --> 00:14:34,790 Yes, yes. And and a virus viruses were not seen at all until electron microscopy was invented in the 1930s. 119 00:14:34,790 --> 00:14:38,569 And I think it was only about eight years after the electron microscope was invented, 120 00:14:38,570 --> 00:14:44,360 that risk of the person that built the first microscope looked and saw a virus for the first time. 121 00:14:44,360 --> 00:14:49,370 So so electron microscopes were needed to see them at all. 122 00:14:49,730 --> 00:14:52,490 And that was a really important thing to do, 123 00:14:52,510 --> 00:14:58,159 because a lot of the classification of viruses was based on the morphological character seen in the in the end, 124 00:14:58,160 --> 00:15:06,889 but that that was at a much to poorer level of detail to be able to to help in drug 125 00:15:06,890 --> 00:15:10,940 design or vaccine design because you couldn't see the sort of chemical basis of them. 126 00:15:12,110 --> 00:15:19,429 So that's, that's why crystallography became the first method to really sort of open up virus structures. 127 00:15:19,430 --> 00:15:30,980 And the it did that because compared to so crystallography started with Bragg looking at salts and first version the whole history of course 128 00:15:31,430 --> 00:15:38,990 you know we're going to have enough to know now about to do as a modern as a modern quiz what way do you think that the point is that, 129 00:15:39,080 --> 00:15:41,290 Crystal, we started off with very simple things, yes. 130 00:15:41,750 --> 00:15:49,910 And a virus is probably of something like two orders of magnitude more complicated than a protein, 131 00:15:50,000 --> 00:15:55,910 which is where crystallography, you know, what what is bread and butter was a few years ago. 132 00:15:56,120 --> 00:16:01,339 So that I was interested because it was sort of quite challenging to do crystallography on it, 133 00:16:01,340 --> 00:16:10,129 but it's perfectly possible to do it if you can get the crystals that gives you a crystal structure with atomic detail. 134 00:16:10,130 --> 00:16:15,230 So that's very useful. But then the you still have to grow a crystal. 135 00:16:15,230 --> 00:16:18,290 So there's, there's that which can be problematic. 136 00:16:19,390 --> 00:16:24,220 And you get a single structure out of that, which is the crystal structure. 137 00:16:24,670 --> 00:16:37,120 Cryo electron microscopy really sort of completely changed from a sort of ability to recognise morphology to to see atomic detail with a series of, 138 00:16:37,510 --> 00:16:45,130 of improvements in the electron microscopes in the ways the electrons were detected and the ways the samples were presented. 139 00:16:45,580 --> 00:16:52,000 The electrons interact much more strongly with matter than X-rays, so the sample has to be extremely thin. 140 00:16:52,000 --> 00:16:55,740 Otherwise the electrons are completely absorbed and extremely cold. 141 00:16:55,750 --> 00:16:59,350 That's hot. And they have to be they have to be cold at cryo. 142 00:16:59,380 --> 00:17:03,280 And that that's they in general. 143 00:17:03,820 --> 00:17:14,530 What we what we learned from X-ray crystallography was that we get a 100 fold greater lifetime of the crystal of a biological sample, 144 00:17:14,530 --> 00:17:18,819 because electrons and X-rays are both damaging of biological samples. 145 00:17:18,820 --> 00:17:24,790 So they destroy the chemistry ultimately. And that's that that's the limitation compared to material science. 146 00:17:25,270 --> 00:17:33,760 So on the other hand, with with electron microscopy, if if your sample, the key thing there is the electron microscope has to be at very high vacuum. 147 00:17:34,390 --> 00:17:45,580 And if you just had a biological sample in liquid in high vacuum, the liquid evaporate so, so way and also the sample would move around. 148 00:17:45,760 --> 00:17:54,190 So so that's the real sort of advantage there is that you sort of freeze and make it make the sample stationary. 149 00:17:54,190 --> 00:17:59,049 You put it in a very thin film of water, so you buy it by proteins. 150 00:17:59,050 --> 00:18:06,070 And also viruses very often need to stay hydrated, otherwise they they become inactivated. 151 00:18:06,310 --> 00:18:15,070 So you can do that with this process of vitrification, which just means that instead of cooling the water down slowly and getting ice crystals, 152 00:18:15,070 --> 00:18:23,380 you could it very quickly and it forms a glass, which then doesn't disturb that the the the proteins and viruses that you want to look at. 153 00:18:23,800 --> 00:18:30,070 So so in the electron microscope, you get these very thin, vitrified cryo specimens. 154 00:18:30,460 --> 00:18:38,200 You have a very, very beautiful sort of beam of electrons and you have a very high quality detector. 155 00:18:38,560 --> 00:18:43,780 And over the last ten years, that's completely transformed the capability of the method. 156 00:18:43,780 --> 00:18:47,930 And it's not finished that transformation process. 157 00:18:48,130 --> 00:18:54,730 There's still more to go in the future, I think. And I think we just need a very quick one sentence description of what synchrotron 158 00:18:55,030 --> 00:18:58,569 enable you to do that you couldn't do with old fashioned X-ray methods. 159 00:18:58,570 --> 00:19:02,440 This same synchrotron gives diamond like. 160 00:19:02,440 --> 00:19:07,899 Yeah, I mean diamond synchrotron. 161 00:19:07,900 --> 00:19:15,070 A lot of this thing from develops actually sort of came about in the UK diamond it allows you 162 00:19:15,070 --> 00:19:21,639 to it gives you a source of X-rays that that is all many orders of magnitude brighter than 163 00:19:21,640 --> 00:19:28,240 a lamps also the sort of hospital X-ray source and that that means that that you can 164 00:19:29,140 --> 00:19:35,920 essentially tackle structures such as viruses that would be almost impossible by a normal lab. 165 00:19:35,920 --> 00:19:43,570 So so it's it's transformed. It's and in terms of the science response and the viral virology response, 166 00:19:44,080 --> 00:19:51,190 the that increased brightness of the beam means that you can collect a data set for, 167 00:19:51,190 --> 00:19:55,060 for instance, a viral protein in perhaps 2 minutes at the synchrotron. 168 00:19:55,480 --> 00:20:01,809 So it alters the the holes. It's a game changer in that rather than just look at a single structure, 169 00:20:01,810 --> 00:20:09,670 you can think about looking at many hundreds of structures of the targets with a number of different chemicals bound. 170 00:20:09,670 --> 00:20:18,530 And then you can help. Then you can much more quickly learn what the basis of the recognition is and improve the chemical design for it. 171 00:20:18,640 --> 00:20:25,180 So that's a wonderful introduction to virology and structural studies and your career so far, is it? 172 00:20:25,210 --> 00:20:33,580 Finally, we get to Kovac. Can you remember where you were, what you were doing, or how you first heard that something was going on in China? 173 00:20:33,580 --> 00:20:38,080 And when you came to realise that it was something that you were going to have to engage with. 174 00:20:40,310 --> 00:20:52,570 It's a tricky one because there's often things happening around the world with viruses and there was the first hint that something was happening, 175 00:20:53,070 --> 00:20:58,510 I guess before the end of the year in 2019. 176 00:20:59,290 --> 00:21:07,330 That wasn't at all obvious to me that this was going to develop into a into a problem. 177 00:21:08,590 --> 00:21:17,200 It did in January 2020, started to become clear that this was this was a serious this was serious. 178 00:21:17,770 --> 00:21:26,150 And. I we i my group at that time weren't working on coronaviruses. 179 00:21:26,750 --> 00:21:33,830 We've done many years before. We've done some work on, uh, sales when, when that outbreak occurred. 180 00:21:34,380 --> 00:21:40,820 Um, but so we were not, it wasn't something that we were particularly attuned to. 181 00:21:40,830 --> 00:21:46,580 So we're probably, uh, you know, I guess perhaps I thought, well, you know, 182 00:21:46,580 --> 00:21:50,989 unless this is really going to be a big thing, we, we'll carry on doing what we're doing. 183 00:21:50,990 --> 00:21:59,540 But I got a call from, uh, a colleague of mine writes to her in China fairly late in January. 184 00:21:59,960 --> 00:22:04,940 Um, and they, by that point, it was clear that this this was serious. 185 00:22:05,810 --> 00:22:11,420 And we we so we were thinking, how should we engage with this? 186 00:22:11,420 --> 00:22:23,569 And how was his he with some other peers were the first people to do a structure of SARS-CoV-2 protein, 187 00:22:23,570 --> 00:22:31,670 which was the main protease, which is the target for the, the rush drug for, for instance, at the moment. 188 00:22:32,390 --> 00:22:45,020 Um, so he phoned up because, uh, well, we do stay in touch, but in particular they'd, they'd got the structure. 189 00:22:47,010 --> 00:22:52,760 And they, they worked extraordinarily quickly to, as soon as the sequences available, 190 00:22:52,760 --> 00:22:57,230 they made the synthetic gene for the protein they wanted to work on. 191 00:22:57,620 --> 00:23:00,469 They driven to the place where the gene was made and brought it back. 192 00:23:00,470 --> 00:23:07,820 And within a few days it got crystals because they'd worked on the samples and the mouse versions of this. 193 00:23:07,820 --> 00:23:11,810 That gave them a lot of sort of insight because it's a related protein. 194 00:23:12,110 --> 00:23:14,690 So they're able to get a structure very quickly. 195 00:23:14,870 --> 00:23:21,680 And they started to sort of think about how they might look at, uh, designing compounds that would inactivated. 196 00:23:22,760 --> 00:23:27,770 But then they, they were using the synchrotron in China, the Shanghai Synchrotron. 197 00:23:27,770 --> 00:23:35,599 And like any synchrotron that runs in, uh, sort of cycles, you have a run and you have a shutdown period, and they had a shutdown period coming up. 198 00:23:35,600 --> 00:23:41,630 So they, he got in touch to see if there was a way this collaboration with Diamond to sort of keep things moving along. 199 00:23:42,020 --> 00:23:48,800 And because of the difficulty of sort of exchanging things, materials, 200 00:23:49,580 --> 00:23:54,170 what actually happened was we ended up using information from them to synthesise 201 00:23:54,170 --> 00:24:00,350 everything over here and but essentially using all of their knowledge that they gained. 202 00:24:00,470 --> 00:24:11,090 And we set up the, uh, a program at Diamond because of, over the years we set up a system at Diamond called X Chem, 203 00:24:11,480 --> 00:24:20,360 which is basically adding a lab on to adjacent to the beamline so that people can come in and, 204 00:24:20,690 --> 00:24:26,390 uh, use pre-made libraries of small chemical fragments, 205 00:24:27,440 --> 00:24:35,419 which can then be dispensed to very high throughput and sort of quickly and easily and reliably onto protein crystals. 206 00:24:35,420 --> 00:24:40,480 And then that there's a system for simulation, well, 207 00:24:40,490 --> 00:24:47,780 helping people prepare those crystals and get them ready for the synchrotron and then those get shipped on to the very high throughput beamline. 208 00:24:48,950 --> 00:24:54,019 So that was that process was set in motion very early on. 209 00:24:54,020 --> 00:25:03,620 The the cloning and crystallisation work was was led by Martin Walsh at Diamond and my deputy at Diamond. 210 00:25:03,620 --> 00:25:10,940 And then the person that leads the project, Frank Delft, who's joint appointment here with Department of Medicine, 211 00:25:11,510 --> 00:25:17,060 was involved in the, uh, the making sure the X Games stuff worked efficiently. 212 00:25:17,390 --> 00:25:22,879 And this was, was this just looking at the main protein? So initially it was just the one thing. 213 00:25:22,880 --> 00:25:28,160 Yeah, because we just wanted to get something up and running and we, we didn't have any extra money for this, 214 00:25:28,160 --> 00:25:35,180 so we just had to sort of people just repurposed what they were doing and then worry about it afterwards, you know. 215 00:25:35,680 --> 00:25:44,989 Um, so that was set up initially. Yeah. And that went very quickly and the results just and right as soon as we got the structure actually 216 00:25:44,990 --> 00:25:50,840 made the structure available even before the protein databank had sort of been overwritten, 217 00:25:50,840 --> 00:25:54,530 which takes a week or so to, to, to review it and then make it available. 218 00:25:54,530 --> 00:25:59,480 So made it available immediately to people. And then it was put on the PDB very quickly. 219 00:25:59,780 --> 00:26:04,939 So in the same way with the with the fragment screen, as soon as the results came through, 220 00:26:04,940 --> 00:26:12,889 they were made available from through from diamond before they could be sort of 221 00:26:12,890 --> 00:26:16,570 put on the PDP because when you get large amounts of data for these things, 222 00:26:16,570 --> 00:26:19,879 it doesn't fit into the deposition method. The president spoke very well. 223 00:26:19,880 --> 00:26:24,380 So it was easier to to make a special sort of site that people could take the data from. 224 00:26:24,890 --> 00:26:30,350 And then the that was to sort of let people know about that. 225 00:26:31,370 --> 00:26:34,460 It was that that really took off through Twitter rather than anything else. 226 00:26:34,640 --> 00:26:37,250 Martin sort of told people on Twitter about this. 227 00:26:37,940 --> 00:26:44,990 And it was a time when probably a significant number of medicinal chemists that were not able to go into work. 228 00:26:45,110 --> 00:26:48,140 So they picked up on the information that was there. 229 00:26:48,530 --> 00:26:57,170 And essentially there was this sort of moonshot project that sort of fell together with people around the world, 230 00:26:57,230 --> 00:27:07,790 particularly sort of led by groups in America. But that the initiative driven by this work in Diamond and that that meant 231 00:27:07,840 --> 00:27:13,010 that the medicinal chemist could look at the results and come up with ideas, 232 00:27:13,010 --> 00:27:18,559 come up with the next generation of compounds that should be better because the first compound should get out of a procedure like that, 233 00:27:18,560 --> 00:27:25,070 a very poor, very weak binders. But you get something quickly and you get quite a lot of results to give you a heads and that. 234 00:27:25,250 --> 00:27:34,540 And then there was also a company that makes chemicals sort of engaged as well to help get the next generation of compounds made. 235 00:27:34,550 --> 00:27:40,760 And that that was the company and Amine, which is a company based in the north of Kiev. 236 00:27:41,360 --> 00:27:45,379 So that they've not not been in action recently. 237 00:27:45,380 --> 00:27:49,040 But they they they were extraordinary helpful in this process. 238 00:27:49,550 --> 00:27:57,410 And that that then sort of went on. And, of course, you know, at each stage, you kind of there's there's different things needed. 239 00:27:57,410 --> 00:28:02,959 And I think the process is sort of what it was done in a sort of open way. 240 00:28:02,960 --> 00:28:11,350 Not it not as a not in a company. So at each stage for the next stage, there had to be a sort of collaboration set up where you had, 241 00:28:11,960 --> 00:28:17,780 you know, how do you do the assays had to be set up and that was done, particularly people in Oxford. 242 00:28:18,740 --> 00:28:24,350 Chris Schofield was, was great in that and is great and Chris is still very heavily involved in lots of aspects of this. 243 00:28:41,800 --> 00:28:46,210 It's it's also spawned further work. 244 00:28:46,390 --> 00:28:56,920 So there's now some some in H projects that were launched a week or two ago which Oxford and Diamond are heavily involved in, where they engage, 245 00:28:56,920 --> 00:29:02,559 recognise that they want to expand this sort of activity to cover not just sociology too, 246 00:29:02,560 --> 00:29:06,160 but to think about other viruses that are potential pandemic threats. 247 00:29:06,670 --> 00:29:10,450 Um, so, so that, that was, that took off. 248 00:29:10,870 --> 00:29:19,389 But of course, because Diamond is a use facility that lots of people came in with expertise in other proteins. 249 00:29:19,390 --> 00:29:27,810 And over the period, I think most of the obvious targets within SARS-CoV-2 have been looked at now with those and those are going forward. 250 00:29:27,960 --> 00:29:31,450 But so the protein that most people seem to hear about is the spike protein. 251 00:29:31,850 --> 00:29:34,690 Now, that's yeah, that's that's that's not really a drug target. 252 00:29:34,690 --> 00:29:39,670 That's just something that's what with your interest, that's what we worked with as well. 253 00:29:39,870 --> 00:29:43,990 Yeah. Yeah. Now we've done we've, we've done a huge amount to this atomic spike protein. 254 00:29:44,050 --> 00:29:47,410 Yeah. Well, so yeah. 255 00:29:47,410 --> 00:29:56,229 Because I'm interested in structure of, of, of whole viruses and the spike is a key component on the virus. 256 00:29:56,230 --> 00:30:01,900 The spike is the protein that recognises receptors on the cell. 257 00:30:01,900 --> 00:30:08,320 So and each virus has lots of spikes. Yes, it's a good one, but no, no, there's quite, quite a lot. 258 00:30:08,860 --> 00:30:13,419 I mean, how many is it is probably still the sort of subject of some discussion, 259 00:30:13,420 --> 00:30:24,610 but there's a good number of spikes and each spike it's it's it's attached at one end into the virus. 260 00:30:24,610 --> 00:30:33,520 The virus has a membrane which surrounds it and the spike has a little tail that, that is anchored in the membrane. 261 00:30:33,790 --> 00:30:37,329 And then and then there is a substantial amount of protein. 262 00:30:37,330 --> 00:30:44,350 It's a big protein, it's a trimer. So there's three protein chains that are identical to each other, arranged in a cluster. 263 00:30:44,740 --> 00:30:56,890 Uh, and at the top of the spike there is one particular bit, but only about a sixth of the total protein, about 200 residues out to 1200. 264 00:30:57,970 --> 00:31:07,720 That is critical for binding to the sort of main, I guess or you know, really key internalisation receptor, which is called ace2. 265 00:31:08,020 --> 00:31:14,560 So this is a molecule on the on our cells that the virus has to recognise to get into the cell. 266 00:31:15,280 --> 00:31:20,560 Um, so that recognition, recognition event is, is critical for cell entry to the virus. 267 00:31:21,670 --> 00:31:30,910 Once that's happened, then the rest of the protein has a different job to do, which is to get the virus into the cell. 268 00:31:31,210 --> 00:31:41,350 And it does that by a series of conformational changes that essentially anchor the the virus into to the host 269 00:31:41,350 --> 00:31:49,480 membrane and then pull the two membranes together so that the viral genome is is released into the host cell. 270 00:31:49,540 --> 00:31:52,900 It's a bit like a search. Yeah, yeah, yeah, yeah, yeah. 271 00:31:53,350 --> 00:31:59,880 So, so, so that. But all of that depends on the initial recognition of the receptor via this the 272 00:32:00,040 --> 00:32:05,049 receptor binding domain RBD as it's known on the top of the top of the spike. 273 00:32:05,050 --> 00:32:10,900 So the spike a three of the sitting on the top and they're they're, they're, they're little, 274 00:32:11,530 --> 00:32:19,150 little domains that can move around when, when they move up, they expose the bit that binds to the ace2 receptor. 275 00:32:19,870 --> 00:32:26,440 When they're down, they don't they can't see the receptor. So it's it's a complicated, dynamic process. 276 00:32:27,760 --> 00:32:35,220 And clearly you, you might expect that, uh, 277 00:32:35,500 --> 00:32:44,290 antibodies that would bind to the place where the receptor normally binds would block the entry of the virus. 278 00:32:44,290 --> 00:32:47,170 And so those would be so-called neutralising antibodies. 279 00:32:47,320 --> 00:32:57,670 Um, so that's, and that's been entirely borne out a huge amount of work that's, that's followed and all the things you've just told me just now, 280 00:32:58,000 --> 00:33:01,960 did we know that already from studies of other coronaviruses or has that come about? 281 00:33:02,230 --> 00:33:14,470 We knew that we knew the basics of we as soon as the sequence came out, you could align it with the first solid virus. 282 00:33:14,770 --> 00:33:17,080 So that's why it was relatively simple to name it. 283 00:33:17,470 --> 00:33:25,990 And not all coronaviruses have the same receptor binding domain structure, but the cells have something that's pretty similar. 284 00:33:27,220 --> 00:33:36,460 And in fact, there are one or two antibodies that neutralise source that can also neutralise SARS-CoV-2, the very, very small number. 285 00:33:36,700 --> 00:33:41,020 But it's sufficiently similar that there are, you know, recognisable, structural. 286 00:33:41,350 --> 00:33:45,700 Things that some some antibodies could recognise between the two viruses. 287 00:33:46,000 --> 00:33:51,760 So much of that was and that and that knowledge was was absolutely critical 288 00:33:51,760 --> 00:33:57,010 for the for a lot of things that follow that if we just started from scratch, 289 00:33:57,010 --> 00:33:59,890 it would have slowed everything down. But so, again, 290 00:34:00,310 --> 00:34:11,620 just as I've had experience with the the enzyme that was experienced from other groups on doing structural biology as spikes of cells and nerves, 291 00:34:12,760 --> 00:34:14,750 which they did, 292 00:34:15,070 --> 00:34:23,920 they knew that these were difficult proteins to work with and they knew that there were some tricks that you could do to stabilise the protein, 293 00:34:23,920 --> 00:34:32,200 make it easier to work with so those and, you know, fold it in the trimeric form when it's taken off the virus or expressed with that the virus. 294 00:34:32,770 --> 00:34:38,950 So there was a series of things that could immediately be applied from what we knew about the other viruses to making this. 295 00:34:38,950 --> 00:34:44,109 It was still it's still a big and, you know, slightly floppy protein. 296 00:34:44,110 --> 00:34:48,760 It was still not trivial to to make high quality reagents. 297 00:34:48,850 --> 00:34:55,030 And that was really quite important, being able it was very important to be able to do that. 298 00:34:55,030 --> 00:35:01,089 And in the early days there was a lot of stuff that, you know, they were they were low quality, simple, 299 00:35:01,090 --> 00:35:11,270 low quality reagents around, which somewhat muddied the waters with some of the work and the screw, be it. 300 00:35:11,770 --> 00:35:15,879 Yeah. The big strengths in structural biology is to be a sort of twofold. 301 00:35:15,880 --> 00:35:19,660 One is the structural virology and the other is cell surface proteins. 302 00:35:20,860 --> 00:35:26,739 And the cell surface proteins are the cell version of the spike. 303 00:35:26,740 --> 00:35:34,570 Essentially. No, they're not the same, but the problems are similar so that our proteins that have a lot of sugars on the outside, 304 00:35:35,200 --> 00:35:41,529 they are associated with membranes and they need to be made in special ways. 305 00:35:41,530 --> 00:35:48,190 So so we were in a good position to be to have expertise in making these proteins. 306 00:35:49,060 --> 00:35:56,200 So we had the very, very early days you going, who's a sort of post. 307 00:35:56,380 --> 00:35:59,470 This worked with me and also with Yvonne. 308 00:36:00,040 --> 00:36:05,470 He looked at this and came up with designs for the protein based on other people's experience. 309 00:36:06,430 --> 00:36:11,770 And people in America were doing the same thing and we sort of use their ideas 310 00:36:11,770 --> 00:36:19,180 and as well and were able to try to make the protein relatively quickly, 311 00:36:20,050 --> 00:36:28,390 get to get the protein in the right formulation and stable enough to do the to do structural work by electron microscopy. 312 00:36:28,390 --> 00:36:37,610 Took a bit of time, but we got there and as I say, there was a lot of groups working on it and the information was shared very rapidly because um, 313 00:36:39,460 --> 00:36:47,650 the, you know, the journals published have responded very quickly in terms of the peer review process. 314 00:36:48,100 --> 00:36:54,880 So there's been very compared to sometimes when it might take a year to get the paper through the whole thing. 315 00:36:55,210 --> 00:36:59,620 It now it can be tender is being turned around very quickly because everybody's recognising importance. 316 00:36:59,620 --> 00:37:04,360 But also the journals encourage people to make the papers available on preprint service. 317 00:37:06,040 --> 00:37:16,600 So it meant that there was an open exchange of information, which was, uh, very good, so that we were able to, to relatively quickly. 318 00:37:16,610 --> 00:37:23,620 I can't remember exactly when, but we were able to get some, some high quality spike and start on structural work. 319 00:37:24,840 --> 00:37:29,620 And and what were the main questions that he were seeking to answer about, about this spike? 320 00:37:36,690 --> 00:37:39,780 Well, I think the part of it was. Was. 321 00:37:42,000 --> 00:37:44,910 Getting the being able to have the reagents. 322 00:37:46,440 --> 00:37:55,440 The because although with structural biologists, high quality reagents like the spike are useful for a lot of other things. 323 00:37:55,800 --> 00:37:59,340 So they're useful for diagnostics that useful for serology testing. 324 00:38:00,080 --> 00:38:08,309 Um, and so actually the, when we were able to, to, to make this stuff, 325 00:38:08,310 --> 00:38:13,230 we then were asked for material by a number of companies that were trying to set up tests, 326 00:38:14,070 --> 00:38:19,200 whether it was sort of lateral flow tests or, or serology tests. 327 00:38:19,510 --> 00:38:29,669 Um, so, so that, that was, that was something that, it was, it was a useful reagent so we could make reasonable or not massive amounts. 328 00:38:29,670 --> 00:38:34,750 We can make enough, enough of the reagents to give to people to do a proof of principle experiment. 329 00:38:35,280 --> 00:38:41,990 So I finally understood. Sorry, I've noticed that your name appears on a lot of these enormous studies, right, 330 00:38:42,420 --> 00:38:46,620 that are looking at serology, that looking at the prevalence of infection in the population. 331 00:38:46,980 --> 00:38:50,040 And I, I hadn't understood why that was. And nowadays. Right. 332 00:38:51,300 --> 00:39:02,790 But and and it was sort of one of one of those one of those things was relatively early on. 333 00:39:02,790 --> 00:39:08,759 I was approached by, um, the person that leads biology, 334 00:39:08,760 --> 00:39:15,030 the biology activity at the Pool Scherrer Institute in, in Switzerland, who I've known for many years. 335 00:39:15,450 --> 00:39:24,090 And he was collaborating with somebody at a hospital in Zurich, and they had a very high throughput serology assay, 336 00:39:24,690 --> 00:39:30,219 robotic assay that they they that they were looking to repurpose, but they didn't have access to the reagents. 337 00:39:30,220 --> 00:39:32,910 So we, we, we, we sent them some reagents. 338 00:39:32,910 --> 00:39:42,270 And then I established contact with people that were developing the assay, you know, really, really nice people out there. 339 00:39:42,600 --> 00:39:49,080 And also established contact with Daniel Hepner in the top of Discovery Institute. 340 00:39:49,110 --> 00:40:00,689 And we sort of I, I, I don't, you know, I'm not, I don't do serology normally, but it just seemed that there was an opportunity to set something up. 341 00:40:00,690 --> 00:40:06,060 So Daniel worked really hard and then set aside some of his robotic equipment. 342 00:40:06,420 --> 00:40:13,470 And, you know, we the club, we work with the people in Switzerland and, and, 343 00:40:14,040 --> 00:40:20,639 and then because it starts to become a, you know, you've got to think about how you run this as a project. 344 00:40:20,640 --> 00:40:29,700 We Diamond donated one of their senior project managers to help sort of set up some of the structures of the thing in the early days. 345 00:40:29,730 --> 00:40:41,670 Um, and, and we set up, we set up that, uh, robotic platform to enable us to do a few thousand, uh, serology tests a day. 346 00:40:42,210 --> 00:40:47,580 Um, and that's, that's now run by, uh, Derek Crook. 347 00:40:47,640 --> 00:40:57,299 Yes. So, and it's, it's I, I think it's, it's, it's, it's been a successful leasing rule, 348 00:40:57,300 --> 00:41:04,680 which is that it's not nontrivial to do that, especially in a sort of academic environment. 349 00:41:04,680 --> 00:41:12,210 But, um, but I think it contributed some useful information is provided a lot of data for the Office for National Statistics. 350 00:41:13,740 --> 00:41:19,830 So that was one example of the value of the reagent. 351 00:41:21,210 --> 00:41:30,690 But then yeah, the sort of questions that you do, the thing the structural balances don't quite know what the questions it's not, 352 00:41:30,930 --> 00:41:34,470 it's often not entirely hypothesis driven but discovery science. 353 00:41:34,960 --> 00:41:41,740 So one of the things was, you know, well, what will what will we learn by doing it? 354 00:41:41,770 --> 00:41:53,670 And that's one of the things that we did wonder was whether there was the possibility for designing small molecule therapeutics against a spike. 355 00:41:54,270 --> 00:41:57,599 And that's proved really difficult. 356 00:41:57,600 --> 00:41:59,610 And we've not been successful in that. 357 00:41:59,610 --> 00:42:13,140 We've got you know, we we came I thought we were getting close at one point because we we found that within the receptor binding domain, 358 00:42:13,200 --> 00:42:20,760 there was a little lipid molecule, a little sort of fatty acid that was bound within it. 359 00:42:20,800 --> 00:42:24,630 And and when when the fatty acid was there, 360 00:42:24,900 --> 00:42:33,030 the receptor binding mates were locked in a down conformations would not be accessible to the two to find the cell. 361 00:42:35,520 --> 00:42:40,600 So I thought that might be a way to, to, to get some. 362 00:42:41,340 --> 00:42:48,290 Uh, if you could replace the fatty acid with something that bind much more tightly, then it could inactivate the virus. 363 00:42:48,300 --> 00:42:53,000 And there are precedents for that type of mechanism from other viruses that work. 364 00:42:53,020 --> 00:43:01,860 Um, but we were never able to reproducibly sort of, uh, get the spike in the right conformation, 365 00:43:01,860 --> 00:43:06,600 were never able to reproducibly bind things into the receptor binding domain. 366 00:43:06,930 --> 00:43:11,950 Once the lipid comes out it by thought and it's this suspect, 367 00:43:11,950 --> 00:43:18,299 all the people probably agree is that the lipid is put in there as the thing comes out of the 368 00:43:18,300 --> 00:43:24,090 cell through a membrane and then probably slowly falls out and that would then activate it, 369 00:43:24,720 --> 00:43:30,270 for instance, might help the virus get out of the cell because it was minimising chance of reattaching, for instance. 370 00:43:30,890 --> 00:43:40,980 Um, so. So that didn't go anywhere. But, but we did, we did try and we collaborated with, uh, with, with a company to try and find something. 371 00:43:41,480 --> 00:43:47,700 Um, but having said that, at the same time, as we stumbled upon this grouping, 372 00:43:47,700 --> 00:43:52,679 Bristol, observe the same thing and, and they, they've taken that much further. 373 00:43:52,680 --> 00:44:01,110 And, and I think they've demonstrated that the principle that you can inhibit the virus by using that binding site. 374 00:44:01,110 --> 00:44:09,239 So so in a sense because they were, they picked this up and they were in a much better position to take it forward. 375 00:44:09,240 --> 00:44:13,770 We just sort of let them do it and good luck. I hope we can do something. 376 00:44:13,770 --> 00:44:17,549 But, um, but you also looked at the interaction of antibodies as well. 377 00:44:17,550 --> 00:44:26,070 That was that those are the sort of major, major thing. Yeah. And that was that that as you hinted, sort of brought together an awful lot of people. 378 00:44:26,620 --> 00:44:40,979 Um, so Gavin Scruton and, and Gavin Ju have, have their group in this building that had moved back to Oxford from Imperial College for a while before. 379 00:44:40,980 --> 00:44:44,100 And we'd sat around and then said, Oh, you know what? 380 00:44:44,100 --> 00:44:49,829 Can we collaborate on that? And he was mainly working on flaviviruses things. 381 00:44:49,830 --> 00:44:57,000 I think the virus that at that time we had ideas, but we didn't, you know, somehow we were busy with other things. 382 00:44:57,000 --> 00:45:08,430 And then when this happened, Gavin's group sort of repurposed and the the spike production and all RBD production was, 383 00:45:09,120 --> 00:45:13,889 well, I didn't say you quangos involved, but from the beginning it was a joint effort with Gavin's group. 384 00:45:13,890 --> 00:45:20,940 So we've worked, we've worked extraordinarily closely together on this and. 385 00:45:22,420 --> 00:45:33,790 So that it was a case of making those reagents and then that enabled and Gavin's group were involved in really doing the gold standard 386 00:45:33,790 --> 00:45:41,350 serology not at such high throughput but slightly lower throughput which which was then used to calibrate the robotic stuff. 387 00:45:41,620 --> 00:45:48,560 But also that underpinned all of the work with the the variants and the, you know, 388 00:45:48,580 --> 00:45:57,310 thing that the whole different cohorts of people that had different sort of vaccination histories and infection histories and stuff like that. 389 00:45:57,820 --> 00:46:10,990 Um, so that was serology, but also because of the expertise of this group, they could, where they had access to, to, to serum samples, 390 00:46:10,990 --> 00:46:25,030 they could then do a book that they could do single cell sequencing from, from blood and get the sequence of antibodies. 391 00:46:25,600 --> 00:46:31,329 And then again using the the spike as a sort of a screen could filtrate. 392 00:46:31,330 --> 00:46:37,930 Those antibodies interact with Spike and then the live virus work was set up 393 00:46:37,930 --> 00:46:48,520 in a pic green suites because the soon after we set up TRUBEE we got money 394 00:46:48,520 --> 00:47:00,760 from welcome to set up a pic of the Oxford Particle Imaging Centre as a place where we could look at viruses in by electron microscopy in containment. 395 00:47:01,210 --> 00:47:09,560 So that's there's areas where we can use electron microscope containment, but there's also tissue culture labs in other parts of. 396 00:47:09,910 --> 00:47:17,950 And this just means a high level of biological safety. Yeah. So so that so so Gavin's group were using those contained labs. 397 00:47:18,310 --> 00:47:24,800 So they they then again repurposed and, and set up, set up the work with live virus. 398 00:47:24,910 --> 00:47:33,040 And then they also set up work with Pseudovirus, which is a sort of safe version where you just take a bit of the, 399 00:47:33,250 --> 00:47:42,909 the spike bits and put it on another, another biological system so that all of those pieces were set up again. 400 00:47:42,910 --> 00:47:48,459 You know, some of these things took longer than you'd hope, but but they came together. 401 00:47:48,460 --> 00:47:57,580 And, um, so the, so we were, we were also working with, we've also worked with other people. 402 00:47:57,850 --> 00:48:05,440 Jim Naismith Um, not, this wasn't his area of science, but he, he, he was, 403 00:48:06,280 --> 00:48:12,640 he's been developing the camera antibodies, the nanavati side for, for some years, 404 00:48:13,420 --> 00:48:21,340 particularly with Ray Owen's, who was initially we recruited to run the ultra protein production facility, 405 00:48:21,340 --> 00:48:25,959 which then sort of migrated down to how well did you say come alive? Yeah, this is to do with most. 406 00:48:25,960 --> 00:48:34,600 And he said, No, no, no. It's just that it's camels, llamas, camelids have special antibodies. 407 00:48:35,050 --> 00:48:45,040 Do you know about the Nanobodies? Because a normal antibody has two chains or like chain and have a chain and both. 408 00:48:45,040 --> 00:48:54,849 Both of those have variable domains, which are all the things that are swapped around genetically to give the variation and 409 00:48:54,850 --> 00:48:59,590 also subjected to somatic mutation in the general centres to optimise the interaction. 410 00:48:59,980 --> 00:49:05,590 So, so that that sort of the interaction area is, it's essentially two protein domains. 411 00:49:06,010 --> 00:49:13,600 The commonest antibodies just have a subset of their antibodies, just have one of those domains which is called the V H. 412 00:49:13,900 --> 00:49:16,510 So that means that there's a very small bit of protein. 413 00:49:16,940 --> 00:49:24,190 It's only 100 residues and it's not that much bigger than the insulin, um, that is responsible for recognition. 414 00:49:24,430 --> 00:49:28,300 So that is a useful reagent because it's so tiny. 415 00:49:29,140 --> 00:49:39,160 It might, you know, it's the sort of thing that you can imagine given giving a guy an inhaler or something to get, to get it into the airways. 416 00:49:39,520 --> 00:49:50,410 So so this was something that people had been working on for some time and the gentleman just got it right, just got money to set this up properly. 417 00:49:50,860 --> 00:49:54,130 So so they they they switched initially. 418 00:49:54,910 --> 00:49:59,530 So we were closer and closer with them to, to, to help initially. 419 00:49:59,530 --> 00:50:04,930 And then they, they got all their own stuff so they could quickly sort of do everything themselves. 420 00:50:05,290 --> 00:50:14,080 Um, and then they, they, they, they worked to develop very high affinity nanobodies as potential therapeutics. 421 00:50:14,610 --> 00:50:17,670 And I don't know whether I'm sure that's at Jim. 422 00:50:17,680 --> 00:50:26,540 Jim's hoping to, to get that. I think in to in to use but it's not so easy. 423 00:50:26,870 --> 00:50:30,770 They the be the big sort of pharmaceutical companies. 424 00:50:31,360 --> 00:50:34,850 You know, they tend to have very well-defined pipelines of what they do. 425 00:50:34,880 --> 00:50:41,360 So if you fit into there, then that normal sort of expectation of what products are, that is much easier. 426 00:50:41,980 --> 00:50:45,410 This is anything a little bit off track is quite a bit slower. 427 00:50:46,790 --> 00:50:52,610 That's but yeah. So. So there was, there was there was that that side. 428 00:50:54,110 --> 00:51:04,040 But yeah. So, so. So then so Gavin's group could then start to pull out antibodies and because they got the live virus, 429 00:51:04,040 --> 00:51:08,630 these thought all these assays, they could select those that would have high affinity. 430 00:51:09,470 --> 00:51:13,340 And then we, we had the ability to do crystallography, 431 00:51:13,610 --> 00:51:19,310 to look at the binding of those to receptor binding domain and do electron microscopy to see and interact with the spike. 432 00:51:19,580 --> 00:51:25,610 So it was an extremely sort of good collaboration because we just went back and forwards and, 433 00:51:25,610 --> 00:51:30,950 you know, came up with ideas and tried to sort of understand these things. 434 00:51:33,010 --> 00:51:44,720 And, you know, we started off with the original sort of early pandemic virus and we've somehow tracked through a lot of work since then. 435 00:51:46,010 --> 00:51:50,870 I noticed that it's 11:00. It's almost got. So I'm just going to one final question. 436 00:51:50,880 --> 00:52:05,990 I just have to skip most it, which is really just a summary of what you've learned really about how to how to do science from working in the pandemic. 437 00:52:05,990 --> 00:52:11,030 I mean, everything you've told me about has involved massive collaboration and sharing and openness. 438 00:52:11,450 --> 00:52:17,390 Was that typical of how you were working before, or do you think there's been a sea change and will it continue? 439 00:52:18,710 --> 00:52:31,220 I, I, I've been really lucky to have had fantastic collaborations and to work with consortia for a long, long time. 440 00:52:31,790 --> 00:52:35,000 So, so that sort of thing is not new to me. 441 00:52:35,210 --> 00:52:39,770 The vaccine work that we're doing for matches with a consortium of the vaccine work in partnership 442 00:52:39,770 --> 00:52:46,490 with a consortium of the European work has always been very consortia like but but what was this was. 443 00:52:48,380 --> 00:52:59,530 But what was really good here was that the consortium was sort of so joined up and and broader than what you used to. 444 00:52:59,800 --> 00:53:10,210 And Richard Cornell was really sort of made, I think I think did some fantastic stuff because as soon as it was clear what was happening, 445 00:53:10,450 --> 00:53:13,780 he set up a sort of group within the Department of Medicine. 446 00:53:13,810 --> 00:53:18,790 We would meet very regularly, you know, and have completely sort of open discussions. 447 00:53:19,270 --> 00:53:28,630 And the interaction that I've had with medics and all of these people around has been completely transformed by this. 448 00:53:29,560 --> 00:53:40,210 And it made it made a huge difference because, as you say, you know, if you look at the people involved, the people that work with, you know, 449 00:53:40,780 --> 00:53:43,059 working with the patients with getting their consent, 450 00:53:43,060 --> 00:53:49,720 making sure that the right samples were obtained and putting those things together, it normally is is terribly hard. 451 00:53:49,730 --> 00:53:57,220 So it's really effective. And and, you know, that extends to the to to the sort of vaccine work as well. 452 00:53:57,260 --> 00:54:02,319 You know, so we'd have these regular meetings when we find out what was happening from from Sarah. 453 00:54:02,320 --> 00:54:13,910 And, you know, and I think that there was that that was really sort of refreshing to have that and and, 454 00:54:14,080 --> 00:54:17,629 you know, something like Derek Crook, I mean, I knew about it, but I've never worked before. 455 00:54:17,630 --> 00:54:24,550 So, you know, it's brilliant. You get to know these fantastic people and, you know, 456 00:54:24,550 --> 00:54:34,570 it was so so that so that that that was I think that that it's that's been one I think that's been a really wonderful thing. 457 00:54:35,770 --> 00:54:39,190 What you know, one of the questions is what have what happens next? 458 00:54:40,150 --> 00:54:47,440 And I think I think I hope it'll be possible to to hold onto the realisation that by 459 00:54:47,440 --> 00:54:53,050 putting all these pieces together you can actually sort of progress things through to a, 460 00:54:53,140 --> 00:54:57,100 to a much sort of richer outcome than you would if you just looking through an area. 461 00:54:57,910 --> 00:54:58,630 But let's see.