1 00:00:00,420 --> 00:00:13,460 I think it's sort of. Well, welcome to the ground around today. 2 00:00:13,460 --> 00:00:21,670 It's nice to see a good turnout in her term week, so it's supposed to introduce Dominic Furness and Burke. 3 00:00:21,670 --> 00:00:25,400 Dominic, many of you know, is the only consultant, 4 00:00:25,400 --> 00:00:35,510 plastic surgeon and associate professor graduating at Trinity College Cambridge and then moved to Christchurch, Oxford, studying clinical medicine. 5 00:00:35,510 --> 00:00:47,750 Did DM research degree and in 2012 was awarded the Wellcome Trust Intermediate Fellowship, which was the first to a plastic surgeon. 6 00:00:47,750 --> 00:00:56,480 His research interests were find out of the room to investigate the genetic and non-genetic causes of common hand surgery conditions, 7 00:00:56,480 --> 00:01:03,780 and his interest is in hand surgery and the super microsurgery for the treatment of lymphedema. 8 00:01:03,780 --> 00:01:17,860 So thank very much. This is. So it was the fourth century BCE when Aristotle called the hand the tool of tools. 9 00:01:17,860 --> 00:01:25,090 And of course, we still use the hand to manipulate tools and create the environment around us. 10 00:01:25,090 --> 00:01:31,180 Hands are use for opening doors, both literal and metaphorical. 11 00:01:31,180 --> 00:01:37,090 And we use them increasingly on keyboards like this and more commonly like this. 12 00:01:37,090 --> 00:01:45,430 And almost everyone in this room uses their hands to help heal people who are suffering from illness. 13 00:01:45,430 --> 00:01:51,610 But of course, the hands have a symbolic role as well that can be used to swear an oath. 14 00:01:51,610 --> 00:02:03,610 What's promised to tell the truth? Hand on heart that can be a sign of respect, but also salutes can be used for other purposes. 15 00:02:03,610 --> 00:02:13,680 The right hands can produce moments of sporting genius, or they can help us cling on to life itself. 16 00:02:13,680 --> 00:02:22,220 When applied by the correct brain, they can produce moments of musical genius. 17 00:02:22,220 --> 00:02:35,770 And very intricate carvings and not only a hands use for painting, but they can actually be the subjects of great works of art themselves. 18 00:02:35,770 --> 00:02:43,450 And when we look back to our ancestors, the earliest cave paintings, what do we find that they painted, they painted hands? 19 00:02:43,450 --> 00:02:47,440 And of note here five out of the six hands are left hands, 20 00:02:47,440 --> 00:02:55,750 which suggests that the artist here was right handed and we'll come back to that in a carers talk later. 21 00:02:55,750 --> 00:03:01,840 That could be used to ask for more that can symbolise love and protection. 22 00:03:01,840 --> 00:03:10,930 Hans can give blessings. They can say that things are OK and that things are not OK. 23 00:03:10,930 --> 00:03:23,110 They can symbolise threats or try to placate be a symbol of peace of friendship, and they can be used to end wars. 24 00:03:23,110 --> 00:03:32,440 Fans can show our appreciation for the enormous achievements of others, symbolise our hope for the future. 25 00:03:32,440 --> 00:03:38,450 They might give us the thumbs up or sometimes the thumbs down. 26 00:03:38,450 --> 00:03:55,920 That can be a symbol of victory and also of surrender. Rebellion, but we must always remember when we're imaging the hand to take two views. 27 00:03:55,920 --> 00:04:02,400 And finally, the most remarkable thing that the hands can do is give sight to the blind. 28 00:04:02,400 --> 00:04:11,910 So this reading of braille really relies upon the tremendous connexion that we have between the hand and the brain. 29 00:04:11,910 --> 00:04:18,510 And the brain, of course, is the most complex and wonderful other organ in our body. 30 00:04:18,510 --> 00:04:26,730 And when we look at the sensory homunculus, we can see how overrepresented the hand is compared to its physical size. 31 00:04:26,730 --> 00:04:34,530 And then when we change to the motor homunculus, whereas other organs have shrunk away, the hand still retains its importance here. 32 00:04:34,530 --> 00:04:41,550 A new imaging techniques such as this functional MRI image using tracks holography are showing us the 33 00:04:41,550 --> 00:04:48,030 enormous number of intricate connexions within our brain that then lead down the spinal cord to our hand. 34 00:04:48,030 --> 00:04:56,790 And again, we'll come back to this later in a carers talk. So when things go wrong, you need a hand surgeon. 35 00:04:56,790 --> 00:05:03,870 And this is the kind of thing that we do on a daily basis. We treat trauma such as tendon injuries and fractures. 36 00:05:03,870 --> 00:05:11,250 We sometimes treat tumours of the hand and then in the bottom panel that we commonly treat arthritis. 37 00:05:11,250 --> 00:05:18,480 But I'm going to talk to you today about cheap trends, disease, which many of you will have heard me talk about before. 38 00:05:18,480 --> 00:05:22,470 So this is a fibro proliferative disease of the Palmer fascia, 39 00:05:22,470 --> 00:05:30,870 where excess collagen gets laid down and then contracts by my fibroblasts to form section contractures of the finger. 40 00:05:30,870 --> 00:05:36,170 It's most commonly treated by surgery, as in this case here, where the Egyptians cords been removed. 41 00:05:36,170 --> 00:05:47,510 This allows a finger to straighten surgeries often complicated by recurrence of the disease because it doesn't treat the biology of the condition. 42 00:05:47,510 --> 00:05:55,670 Jeep turns disease is a typical complex genetic disease where multiple genetic factors are multiple, 43 00:05:55,670 --> 00:06:02,480 non genetic factors come together in a single person to cause expression of the disease. 44 00:06:02,480 --> 00:06:09,620 And the way that we study the genetics of such diseases is by a genome wide association study. 45 00:06:09,620 --> 00:06:14,060 So we have about three billion base pairs in our genome, 46 00:06:14,060 --> 00:06:20,660 and each of you differs from the person sitting next to you at about 10 million of those places. 47 00:06:20,660 --> 00:06:26,750 And the vast majority of those variants are called single nucleotide polymorphisms or snips. 48 00:06:26,750 --> 00:06:36,360 So here you can see the person on the left has an a at a particular place and the person on the right has a T. 49 00:06:36,360 --> 00:06:38,550 So when we're thinking about complex disease, 50 00:06:38,550 --> 00:06:50,340 we compare those snips at up to a million places and sometimes more than that across the entire genome in cases and controls. 51 00:06:50,340 --> 00:06:55,500 And if we find that a particular variant is more common in cases than controls, 52 00:06:55,500 --> 00:07:08,430 significantly more common than we can say that that snip and the area around that snake, we call that a locus is associated with a particular disease. 53 00:07:08,430 --> 00:07:17,310 So it's important to say that it's not necessarily that variant that causes the disease because the variants are inherited in blocks together. 54 00:07:17,310 --> 00:07:22,950 And so there may be hundreds of variants in that area, all of which are inherited together, 55 00:07:22,950 --> 00:07:33,300 and you can't tell the difference between them genetically. So this is a marker of where there is some genetic predisposition to a particular disease. 56 00:07:33,300 --> 00:07:40,950 Now, because we're doing a million tests, we can't use a normal P value of nought point nought five to declare statistical significance. 57 00:07:40,950 --> 00:07:43,590 We have to correct that for multiple testing. 58 00:07:43,590 --> 00:07:52,500 So that gives us a generalised threshold for declaring genome wide significance of five times 10 to the minus eight. 59 00:07:52,500 --> 00:08:02,340 So in order to get the power to achieve those kind of P values, he needs thousands of cases and thousands of controls. 60 00:08:02,340 --> 00:08:08,370 So we've done this for cheap trans disease, and last time I spoke to grind around a couple of years ago, 61 00:08:08,370 --> 00:08:17,310 I reported some of the findings that we had from our second genome wide, our bigger genome wide association study, which is now published. 62 00:08:17,310 --> 00:08:25,320 And we found 26 snips 26 loci regions of the genome that predispose to dupe turns disease. 63 00:08:25,320 --> 00:08:30,690 And I'm just going to update you on some of the functional work that we've done on one of those snips, 64 00:08:30,690 --> 00:08:39,480 which we think is going to be a good therapeutic target, not only in Duke trends, but potentially in other diseases. 65 00:08:39,480 --> 00:08:42,060 And this is the snap. 66 00:08:42,060 --> 00:08:57,060 It's on chromosome 14, and it's within a gene called MMP 14, also known as MT1 MMP or memory membrane bound type one matrix Martello protease. 67 00:08:57,060 --> 00:09:04,770 It's the only snip out of those 26 that we found that actually causes a coding change with energy. 68 00:09:04,770 --> 00:09:15,720 So the vast majority that we find are regulatory variants, and they change gene expression and gene regulate and regulation of gene expression, 69 00:09:15,720 --> 00:09:24,270 whereas this one actually changes in amino acid in a protein, and that's pretty rare for genome wide association studies. 70 00:09:24,270 --> 00:09:30,150 The other thing is that, as I said to these variants tend to be inherited in blocks, 71 00:09:30,150 --> 00:09:43,780 and so there might be 20 or 40 or 100 or 200 different variants, which we can't tell which one of those is causative from using genetic means. 72 00:09:43,780 --> 00:09:54,100 Interestingly, with this snip, it was a sole snip. So there were no other snips in that block that could look like they could be causative. 73 00:09:54,100 --> 00:10:04,390 And it's very common. So if we look in the western European population, we find that almost 40 percent of people are heterozygous for this snap, 74 00:10:04,390 --> 00:10:13,010 an almost seven percent of people are homozygous for it. So have we followed this up? 75 00:10:13,010 --> 00:10:20,360 Well, it just so happens that in our department, in end dorms, we have probably the world's expert on empty one MP. 76 00:10:20,360 --> 00:10:26,360 So life is all about happy coincidences. And this is one of those happy coincidences. 77 00:10:26,360 --> 00:10:31,700 So I went to speak Tsuyoshi, and we've done some follow up work. 78 00:10:31,700 --> 00:10:42,440 A lot of the work has been done in his lab. So a little bit more about MTV on MMP, as I said, it's a transmembrane protease. 79 00:10:42,440 --> 00:10:52,610 The important function here to focus on is that it degrades extracellular matrix components, so particularly the fibular collisions. 80 00:10:52,610 --> 00:11:01,670 It also works as a home, a timer, so it needs to time arise in order to be able to digest those loans and habits. 81 00:11:01,670 --> 00:11:11,290 Proteolytic functions. It's important not only in trends, but in the multitude of other diseases. 82 00:11:11,290 --> 00:11:19,930 So it's vital for cellular invasion, and if you culture tumour cells with cancer associated fibroblasts, 83 00:11:19,930 --> 00:11:24,790 then they will invade through a matrix through college and matrix. 84 00:11:24,790 --> 00:11:30,910 If you knock out A. one MMP in those cancer associated fibroblasts, they can no longer invade. 85 00:11:30,910 --> 00:11:36,070 So it's really important for metastasis of cancer. 86 00:11:36,070 --> 00:11:40,930 It's important in the invasion of cartilage in rheumatoid arthritis, 87 00:11:40,930 --> 00:11:46,750 very inflammatory panis to invade through into the cartilage and degrade the cartilage. 88 00:11:46,750 --> 00:11:52,070 And as I said, it's also vital in extracellular matrix homeostasis. 89 00:11:52,070 --> 00:12:02,200 So in the MCU, on MMP knockout mice, they die at about 12 weeks of age and they have generalised musculoskeletal fibrosis. 90 00:12:02,200 --> 00:12:07,460 So it's kind of it's it's a seesaw, really when we think about disease. 91 00:12:07,460 --> 00:12:13,660 So if you have two little mouse, you want MMP function, then you're predisposed to fibrosis. 92 00:12:13,660 --> 00:12:16,930 And if you have too much and you're unlucky enough to get cancer, 93 00:12:16,930 --> 00:12:28,310 then it may well predispose you to having more invasion, more metastasis, etc. And that's an area that we're looking at. 94 00:12:28,310 --> 00:12:31,430 So what about this variant? 95 00:12:31,430 --> 00:12:42,080 So here, top left, you can see a structural model of the catalytic domain of M2 on MMP and where the zinc ion binds, that's the catalytic cleft. 96 00:12:42,080 --> 00:12:50,960 That's where the the protease activity happens and the coloured area on the bottom right. 97 00:12:50,960 --> 00:12:53,840 That's where our particular amino acid is. 98 00:12:53,840 --> 00:13:05,330 So the aspartic acid to aspire gene substitution, you can see on the top right doesn't really give us a major structural change at all. 99 00:13:05,330 --> 00:13:12,380 So we thought, I'm not quite sure what this is going to do to the catalytic activity of empty one MMP. 100 00:13:12,380 --> 00:13:20,210 So we transfected some cos nine cells, which don't express them to you on MMP and then looked at the general catalytic function. 101 00:13:20,210 --> 00:13:22,520 So this doesn't require a timer ization. 102 00:13:22,520 --> 00:13:32,240 This is just gelatine degradation, and you can see that the wild type and mutant form degrade the gelatine in exactly the same way. 103 00:13:32,240 --> 00:13:46,280 Not very promising. We then looked at the ability of the MT1 MMP to activate another Matrix metallic protease programme MP two MP to the active form. 104 00:13:46,280 --> 00:13:52,340 And we did find a reduction in the mutant form of about 50 percent. 105 00:13:52,340 --> 00:13:58,310 And with the heterozygous so the double transfected cells, it was about halfway between. 106 00:13:58,310 --> 00:14:04,700 So quite interesting, but not so amazing. 107 00:14:04,700 --> 00:14:07,790 But what we did find, which was really interesting, 108 00:14:07,790 --> 00:14:15,050 was when we did the collagen degradation assay a big difference between wild type and recent forms. 109 00:14:15,050 --> 00:14:25,010 So on the top left panel, the mock transfected cells don't degrade the college, and it's also the field remains dark. 110 00:14:25,010 --> 00:14:32,840 The wild type cells when you express wild type MT1 MMP, they punch multiple holes in the collagen matrix. 111 00:14:32,840 --> 00:14:39,200 And so the light comes through and you can see it's bright when we look at the mutant form. 112 00:14:39,200 --> 00:14:45,200 You can see that it's hardly able to degrade the collagen at all, which is really interesting. 113 00:14:45,200 --> 00:14:51,080 And the there's the heterozygous form, if you like the double transfected form. 114 00:14:51,080 --> 00:14:59,450 Again, the amount of collagen degradation is severely reduced, and you can see that in graphical form on the right hand side. 115 00:14:59,450 --> 00:15:04,190 So we wondered what happens in real patients. So these are just cells in the lab. 116 00:15:04,190 --> 00:15:09,280 What about cells from patients? 117 00:15:09,280 --> 00:15:21,130 So these are cells derived from trans patients who are operated on and we've grown their my fibreglass in the lab that group by genotype, 118 00:15:21,130 --> 00:15:27,430 and you can see that this is a collagen degradation assay in exactly the same way as the previous slide. 119 00:15:27,430 --> 00:15:31,660 The wild type that goes to great collagen normally, 120 00:15:31,660 --> 00:15:43,510 but both the heterozygous and the homozygous mutant forms are down to about 13 percent of collagen degradation, 121 00:15:43,510 --> 00:15:47,350 and it seems to be working in a dominant negative manner here. 122 00:15:47,350 --> 00:15:54,680 So this fits with the dimerisation theory as well. 123 00:15:54,680 --> 00:15:57,890 So that's roughly where we're at. We've done a few more things now. 124 00:15:57,890 --> 00:16:04,820 What we're interested in next is looking at this relationship between cancer and fibrosis, 125 00:16:04,820 --> 00:16:16,070 and we've approached several several groups who've done genetics studies of big genetic studies of various tumour types. 126 00:16:16,070 --> 00:16:22,400 What we're really interested in is not does MT1 want MMP predispose you to getting cancer in the first place? 127 00:16:22,400 --> 00:16:28,430 But if you got it? What does it do? What does this variant do to your chances of progression? 128 00:16:28,430 --> 00:16:34,370 Your chances of lymph, vascular invasion, metastasis, etc.? 129 00:16:34,370 --> 00:16:44,730 We're planning to perform a chemical library screening in collaboration with a pharma company looking for small molecules that can modify either 130 00:16:44,730 --> 00:16:57,650 to the wild type form to to mimic what happens in the mutant form or to restore collagen journalistic activity to the mutant form itself. 131 00:16:57,650 --> 00:17:01,580 And we're interested in what the structural basis of this is because, as I said, 132 00:17:01,580 --> 00:17:06,830 it's not a big structural change, but it's having quite a big biochemical change. 133 00:17:06,830 --> 00:17:13,520 And so we're collaborating with the Dunn School to look at criterium of that. 134 00:17:13,520 --> 00:17:20,480 So that's where we're at. With just one of those variants we would like to do now is introduce a carer. 135 00:17:20,480 --> 00:17:25,130 So I care why. Berg is DPhil student in my group. 136 00:17:25,130 --> 00:17:36,320 He's an MRC clinical research fellow, and he's going to talk to you about more of the nerve and brain side of the title. 137 00:17:36,320 --> 00:17:52,020 So I'm going to hand over the. I'm going to talk about carpal tunnel syndrome initially, 138 00:17:52,020 --> 00:17:57,090 everyone in this room for me certainly knows the carpal tunnel syndrome is caused by compression 139 00:17:57,090 --> 00:18:03,690 of the median as it enters the hand through an anatomical tunnel and left untreated, 140 00:18:03,690 --> 00:18:10,320 it leads functional impairments in the hand. It affects between five to 10 percent of the population. 141 00:18:10,320 --> 00:18:16,410 So it's a common disease, and it's by far and away the most common entrapment neuropathy. 142 00:18:16,410 --> 00:18:23,010 Over 50000 carpal tunnel operations are performed in England every year, so it's an economically potent disease, 143 00:18:23,010 --> 00:18:28,350 both to society and to the individual and to the health service as well. 144 00:18:28,350 --> 00:18:30,480 What causes carpal tunnel syndrome? 145 00:18:30,480 --> 00:18:35,820 Well, we know that there are associations with other diseases that make you more likely to get carpal tunnel syndrome, 146 00:18:35,820 --> 00:18:41,280 such as diabetes, rheumatoid arthritis, hypothyroidism and gout. 147 00:18:41,280 --> 00:18:48,480 We also know that pregnant women are more likely to get carpal tunnel syndrome, although this tends to be transient and the majority of cases, 148 00:18:48,480 --> 00:18:58,190 and there are some weaker evidence for occupational risk factors such as performing repetitive tasks and the use of racing tools. 149 00:18:58,190 --> 00:19:03,290 But that doesn't tell us why Person A will go on to develop carpal tunnel syndrome. 150 00:19:03,290 --> 00:19:09,020 All things being equal risk Person B will not get carpal tunnel syndrome. 151 00:19:09,020 --> 00:19:15,170 And that's because carpal tunnel syndrome is your typical complex disease with a significant genetic component. 152 00:19:15,170 --> 00:19:22,370 It has a relatively high heritability of 46 percent, and about a third of patients report a family history. 153 00:19:22,370 --> 00:19:29,330 Yet we know next to nothing about the genes that are important in the pathogenesis of carpal tunnel syndrome. 154 00:19:29,330 --> 00:19:35,900 Before we started this research, we asked ourselves, How can your genes confer risk to carpal tunnel syndrome? 155 00:19:35,900 --> 00:19:41,090 And we came up with two competing hypotheses. Either your genes, 156 00:19:41,090 --> 00:19:51,200 also the environment through which the median nerve transits or your genes somehow render the median nerves more vulnerable to increase pressures. 157 00:19:51,200 --> 00:20:00,200 So the aim of this research was to try and answer that question by discovering genes that predispose to carpal tunnel syndrome by performing a glass. 158 00:20:00,200 --> 00:20:04,220 And for us, we use carpal tunnel patients in UK Biobank. 159 00:20:04,220 --> 00:20:11,750 This is a prospective cohort study of about half a million individuals aged between 40 to 69 at the time of recruitment. 160 00:20:11,750 --> 00:20:18,620 And all of these people have had whole genome genotyping undertaken, so we know that genotypes at about 800000 steps. 161 00:20:18,620 --> 00:20:23,030 And we also have access to them medical records in the form of diagnostic codes. 162 00:20:23,030 --> 00:20:26,990 And we have lots of other lifestyle data and also some imaging as well. 163 00:20:26,990 --> 00:20:32,130 And the great thing about UK Biobank is that it's freely available to all researchers around the world, 164 00:20:32,130 --> 00:20:38,660 so it's a profoundly powerful resource for performing this sort of genetic association study. 165 00:20:38,660 --> 00:20:43,470 So we started off with about half a million UK Biobank individuals in about just under 100000 166 00:20:43,470 --> 00:20:49,550 steps on the first and arguably most important step in a glass perform quality control. 167 00:20:49,550 --> 00:20:57,620 And this is a form of curating your data by getting rid of the snips that are either not particularly informative or misleading, 168 00:20:57,620 --> 00:21:04,940 and also getting rid of individuals whose DNA might be contaminated or mixed up, or people who are related to each other. 169 00:21:04,940 --> 00:21:11,630 So after QC, we were left with just under 400000 individuals and just over half a million snips. 170 00:21:11,630 --> 00:21:16,670 We then performed case ascertainment and we phenotype these patients and we decided who is a case or not, 171 00:21:16,670 --> 00:21:25,400 depending on whether they had diagnostic codes for carpal tunnel syndrome. So we used ICD 10 codes, OPEC's codes, the carpal tunnel decompression. 172 00:21:25,400 --> 00:21:30,290 And we looked at self-reported codes for carpal tunnel syndrome and for carpal tunnel surgery. 173 00:21:30,290 --> 00:21:38,240 And this gave us about 12000 cases and about three hundred and ninety thousand controls going into the association study. 174 00:21:38,240 --> 00:21:45,680 I'm going to quickly summarise our key findings. This is called a Manhattan plus, and this is how juicy results are presented. 175 00:21:45,680 --> 00:21:51,050 It's called that because it resembles the Manhattan skyline on the y axis, you have the p value. 176 00:21:51,050 --> 00:21:57,620 In the association study that Don talked about. And on the x axis, you have the chromosome and the genomic position. 177 00:21:57,620 --> 00:22:03,470 The dotted red line represents the threshold for genome wide significance of five times 10 to the minus eight. 178 00:22:03,470 --> 00:22:11,300 So we discovered 16 regions or loci across the genome that are significantly associated with carpal tunnel syndrome. 179 00:22:11,300 --> 00:22:17,810 This is what the results looked like in tabular formats and the first column you have the chromosome where your top snip is located. 180 00:22:17,810 --> 00:22:21,320 You've got the genomic position. You've got the name of the snip. 181 00:22:21,320 --> 00:22:28,610 And as we were saying earlier, there are several steps to the locus, but we're focussing on the top snip as each locus. 182 00:22:28,610 --> 00:22:36,290 Next, we have the odds ratio for developing carpal tunnel syndrome if you have the risk variant and then you have the P value dissociation. 183 00:22:36,290 --> 00:22:41,270 So we run these summary statistics through various points somatic tools and they 184 00:22:41,270 --> 00:22:46,250 tell us which genes are nearby and are likely to be related to these variants. 185 00:22:46,250 --> 00:22:51,530 We discovered there were three genes in particular that we were drawn to initially. 186 00:22:51,530 --> 00:22:53,600 I'm going to talk about this briefly. 187 00:22:53,600 --> 00:23:00,830 We found variants within genes called Adam C17 and Adam Kasten, and both of these happened to be missense variants. 188 00:23:00,830 --> 00:23:07,010 So these are these are genetic variants to lead to an amino acid substitution in your protein products. 189 00:23:07,010 --> 00:23:12,260 So very rare to pick up in a once, and it's another gene called SCMP, one that we were interested in as well. 190 00:23:12,260 --> 00:23:19,610 But we'll start with Adam, 'cause these are a family of proteolytic enzymes related to matrix metallic proteases. 191 00:23:19,610 --> 00:23:27,830 They have a wide range of functions, but importantly, extracellular matrix maintenance, sanitation and migration, and that we picked up in Adam. 192 00:23:27,830 --> 00:23:38,120 C17 causes an amino acid substitution in the protease domain and is predicted to be deleterious mutation since the RNC has 10. 193 00:23:38,120 --> 00:23:44,930 This leads to an arginine to glutamine substitution directly adjacent to an important sites within the enzyme. 194 00:23:44,930 --> 00:23:54,800 And this reduces the enzymes activation efficiency. So these are just zoomed in Manhattan plots, essentially, some of which is two earlier, 195 00:23:54,800 --> 00:23:59,360 and you can see that there aren't many other snips in that region apart from those two steps we picked up. 196 00:23:59,360 --> 00:24:06,500 So we're quite confident that we have actually found the causal variance in these two cases, which is very rare. 197 00:24:06,500 --> 00:24:10,450 Yes, if you want. Codes for a protein called Sibilant three. 198 00:24:10,450 --> 00:24:16,360 And this is a glycoprotein associated with basement membranes, elastic fibres and matrix components. 199 00:24:16,360 --> 00:24:22,410 And it's been associated with various cancers, hernias and also varicose veins. 200 00:24:22,410 --> 00:24:30,630 And the snip that we found is an intranet SCMP ones within the gene, but it's in a non-coding gene, but it's actually an enhancer region, 201 00:24:30,630 --> 00:24:38,430 which means that this variance increases the likelihood of transcription factors binding to it and therefore initiating transcription. 202 00:24:38,430 --> 00:24:43,770 And we certainly know that in skin, which isn't necessarily the tissue, were interested in print skin. 203 00:24:43,770 --> 00:24:51,210 It's been shown that's having the risk. Variants of the people actually produce more of this protein products than the heterozygous. 204 00:24:51,210 --> 00:25:01,340 The people who in turn produce more than the homozygous for the non-racialism, gallium carpenter carpal tunnel syndrome. 205 00:25:01,340 --> 00:25:08,850 We form something called a gene based enrichment analysis, which is where you take all the genes that were implicated in you guys. 206 00:25:08,850 --> 00:25:14,420 You see if they're overrepresented in particular biological pathways or gene ontologies. 207 00:25:14,420 --> 00:25:21,860 And you can see here that there is a lot of enrichment for extracellular matrix related ontologies. 208 00:25:21,860 --> 00:25:28,760 So we went back and looked at our genes and we found that, yes, there was an overabundance of extracellular matrix protein genes. 209 00:25:28,760 --> 00:25:39,210 But where are these genes acting in the context of carpal tunnel syndrome? It's long been suspected that the substance of your connective tissues, 210 00:25:39,210 --> 00:25:43,110 by which I mean the connective tissues around the median and around the flexor tendons 211 00:25:43,110 --> 00:25:47,560 in the carpal tunnel are somehow involved in carpal tunnel syndrome pathogenesis. 212 00:25:47,560 --> 00:25:51,390 And this shows an extended carpal tunnel decompression. 213 00:25:51,390 --> 00:26:00,150 The blue arrow shows the median nerve, and you can nicely see how it becomes quite injects it and quite hyperemesis as it enters the carpal tunnel. 214 00:26:00,150 --> 00:26:07,200 And below that you see one of the flexor tendons. Now flexor tendons are supposed to look like that's that's supposed to be white and shiny. 215 00:26:07,200 --> 00:26:12,600 But in this particular carpal tunnel, patients is covered in an inflamed sign over him, 216 00:26:12,600 --> 00:26:20,100 and this is something you often find when you open up carpal tunnel. So we wondered, are genes expressed in this tissue? 217 00:26:20,100 --> 00:26:26,790 So I extracted RNA from 41 patients who were undergoing carpal tunnel decompression, 218 00:26:26,790 --> 00:26:35,370 from whom tennis on medium was resected and run an RNA seq experiments and found that our three candidate genes are highly expressed in this tissue, 219 00:26:35,370 --> 00:26:41,560 suggesting that they have a plausible biological role and seen as synovial. 220 00:26:41,560 --> 00:26:46,800 Well, there are any practical applications of this in the near future. Well, this sort of thing. 221 00:26:46,800 --> 00:26:55,350 And so here are some key statistics again. These are 16 snips, and this is the risk that increases your risk of developing carpal tunnel syndrome. 222 00:26:55,350 --> 00:26:59,880 Now we all have either zero, one or two copies of each of these aliyu's, 223 00:26:59,880 --> 00:27:09,150 so we can construct a genetic risk score by multiplying the number of risk deals an individual has by the natural logarithm of the odds ratio. 224 00:27:09,150 --> 00:27:16,860 We do this across all 60 snips and some them to create a weighted genetic risk score within the UK Biobank cohorts. 225 00:27:16,860 --> 00:27:21,420 Individuals with carpal tunnel syndrome had a mean genetic score of one point sixty two, 226 00:27:21,420 --> 00:27:29,940 whereas controls had a genetic risk score of one point forty seven. We hypothesised that within those patients who had carpal tunnel syndrome, 227 00:27:29,940 --> 00:27:36,030 those who've had surgery are genetically more predisposed to a more severe phenotype. 228 00:27:36,030 --> 00:27:46,480 And consistent with this, we found that they operated carpal tunnel group had a significantly higher genetic risk score than the unaffiliated group. 229 00:27:46,480 --> 00:27:55,860 Therefore, genetic risk or correlates with disease severity. Our next hypothesis was whether patients with carpal tunnel syndrome. 230 00:27:55,860 --> 00:28:02,680 They have a higher genetic risk than those who have not needed a revision operation and have only had a single operation. 231 00:28:02,680 --> 00:28:09,040 And although we found this was the case, we were underpowered to find statistical significance. 232 00:28:09,040 --> 00:28:15,130 So we can't use this in clinic just yet because our genetic risk score is based on 16 snips. 233 00:28:15,130 --> 00:28:20,260 But as we discover more and more sniffs, our genetic instruments will become stronger for detecting this. 234 00:28:20,260 --> 00:28:25,770 And this is the sort of thing that can be used on an individual level for prognostication. 235 00:28:25,770 --> 00:28:28,890 This was particularly interesting to us when we looked at our results, 236 00:28:28,890 --> 00:28:35,520 we found this four of us snips were previously reported to be genetic determinants of human height. 237 00:28:35,520 --> 00:28:40,560 So we wondered would we see a height difference between carpal tunnel cases and controls in UK Biobank? 238 00:28:40,560 --> 00:28:44,760 And to our amazement, we found that there was a two centimetre difference in height. 239 00:28:44,760 --> 00:28:49,860 The carpal tunnel patients have shorter than controls in both males and females. 240 00:28:49,860 --> 00:28:58,650 And is this simply association or is this causation? And to untangle that relationship, we performed a Mendelian randomisation analysis, 241 00:28:58,650 --> 00:29:03,570 which is a technique that is used to basically establish causation between a putative 242 00:29:03,570 --> 00:29:09,150 risk factor and an outcome using genetic variants as a natural experiments. 243 00:29:09,150 --> 00:29:15,090 And the important thing here is that the negative gradients of the course and the fact that this was highly statistically significant. 244 00:29:15,090 --> 00:29:20,730 And we've demonstrated the highs is inversely causal in the aetiology of carpal tunnel syndrome, 245 00:29:20,730 --> 00:29:24,840 and we calculated that each standard deviation increase in height is associated 246 00:29:24,840 --> 00:29:30,300 with a point seventy six odds ratio for developing carpal tunnel syndrome. 247 00:29:30,300 --> 00:29:34,900 We weren't the first to discover this association between heights and carpal tunnel syndrome, 248 00:29:34,900 --> 00:29:41,610 although we've explained it for the first time, hopefully. But some other people have discovered that carpal tunnel patients are not only shorter, 249 00:29:41,610 --> 00:29:46,260 but they also have shorter hands, wider palms and a greater risk ratio. 250 00:29:46,260 --> 00:29:50,250 So they essentially have a slightly square or around a wrist. 251 00:29:50,250 --> 00:29:55,950 And this actually fits very nicely with our findings because Heights is essentially a proxy for all skeletal growth. 252 00:29:55,950 --> 00:29:56,970 So if you're shorter, 253 00:29:56,970 --> 00:30:04,500 you're probably more likely to have this sort of risk configuration if you go back to the two genes that we talked about earlier. 254 00:30:04,500 --> 00:30:10,740 There is an ardent 10 mutation. In fact, there are several that can cause autosomal recessive vile Nakasone syndrome. 255 00:30:10,740 --> 00:30:17,550 This is characterised by short stature bracket athletes to short fingers and joint stiffness and a similar biomarker. 256 00:30:17,550 --> 00:30:25,920 Sony like syndrome in advancing 70 mutations, which manifests with short stature and iron on these and carpal tunnel syndrome has 257 00:30:25,920 --> 00:30:29,940 been reported in families with biomarker sahni syndrome and even the C children, 258 00:30:29,940 --> 00:30:33,750 which is vanishingly rare in the general population. 259 00:30:33,750 --> 00:30:41,190 So this nice little close closes the loop on the saga between heights, carpal tunnel syndrome and these two genes. 260 00:30:41,190 --> 00:30:47,460 So going back to this slide with our two competing hypotheses how can you genes confer risk carpal tunnel syndrome? 261 00:30:47,460 --> 00:30:52,020 The results this study certainly seem to favour this first hypothesis. 262 00:30:52,020 --> 00:30:58,770 Thus aberrations in your extracellular matrix architecture, either within the carpal tunnel or within the nerve itself, 263 00:30:58,770 --> 00:31:03,960 perhaps or simply being shorter and having altered wrists dimensions. 264 00:31:03,960 --> 00:31:08,910 These seem to be the important genetic determinants of carpal tunnel syndrome. 265 00:31:08,910 --> 00:31:13,500 So in summary, we have performed the status of a glass of carpal tunnel syndrome. 266 00:31:13,500 --> 00:31:16,410 We discovered 16 susceptibility loci. 267 00:31:16,410 --> 00:31:25,080 We identified some biologically plausible candidate genes and found that they are expressed in our tissue of interest, namely the Tino Sino-African. 268 00:31:25,080 --> 00:31:29,970 We developed a genetic risk score that seems to correlate with disease severity, 269 00:31:29,970 --> 00:31:34,830 and we have found this causal role of heights in carpal tunnel syndrome, 270 00:31:34,830 --> 00:31:39,240 and we provided some novel insights into the biology of carpal tunnel syndrome. 271 00:31:39,240 --> 00:31:43,170 And I'm going to finish by quickly talking about something almost completely 272 00:31:43,170 --> 00:31:50,310 different in the career of a plastic surgeon and indeed a plastic surgery trainee. 273 00:31:50,310 --> 00:31:55,890 The question we most commonly ask patients is probably this all you left to right handed. 274 00:31:55,890 --> 00:32:00,570 We can one further and we ask, why are you left right handed? 275 00:32:00,570 --> 00:32:06,780 Because handedness is interesting across all cultures, about 90 percent of humans are right-handed, 276 00:32:06,780 --> 00:32:09,750 and this has been the case since at least the Palaeolithic period. 277 00:32:09,750 --> 00:32:17,370 As we know from cave paintings and looking at tools, the scheme and the distribution of handedness is a uniquely human trait. 278 00:32:17,370 --> 00:32:20,830 And the question of handedness has been culturally loaded for a very long time. 279 00:32:20,830 --> 00:32:23,280 This historically had negative connotations. 280 00:32:23,280 --> 00:32:30,960 It's been associated with evil and malady, so words like sinister and gauche candidate all refer to left handedness. 281 00:32:30,960 --> 00:32:38,640 On the other hand, left handedness is also associated with creativity and genius in popular culture. 282 00:32:38,640 --> 00:32:45,300 So it's certainly something that we've been interested in for a long time. So what do we know about the genetics of handedness? 283 00:32:45,300 --> 00:32:53,220 Well, very little considering it's been debated considerably in many academic spheres for well over a century. 284 00:32:53,220 --> 00:32:59,170 We know from twin studies that there is a heritable component to this. The estimates of 25 percent. 285 00:32:59,170 --> 00:33:01,470 And that's quite a modest heritability us. Perhaps why? 286 00:33:01,470 --> 00:33:10,110 Despite several previous goals, no one has found any genome wide significant associations with handedness in the general population. 287 00:33:10,110 --> 00:33:17,550 So we wondered, can we harness the size and power of UK Biobank to find associations within UK Biobank? 288 00:33:17,550 --> 00:33:26,970 We had about 350000 right handers, about 40000 left handers and about 6000 ambidextrous people, and energy was of right handers versus left handers. 289 00:33:26,970 --> 00:33:32,520 We left out the ambidextrous people to try and capture the two extremes the phenotype, 290 00:33:32,520 --> 00:33:37,680 and we were delighted to discover three genome wide association signals. 291 00:33:37,680 --> 00:33:40,740 We then went on to do a right handers versus non-white Hamdi's G. 292 00:33:40,740 --> 00:33:49,170 So we lumped the ambidextrous participants with the the left handers, and we picked up this extra hit here on chromosome six. 293 00:33:49,170 --> 00:33:55,060 And when we scrutinised these genomic regions for these hits were something very interesting became apparent. 294 00:33:55,060 --> 00:34:01,410 That is this three out of four of our loci, we found genes related to microtubules. 295 00:34:01,410 --> 00:34:05,670 This included top, which codes for beta tubulin map two, 296 00:34:05,670 --> 00:34:11,160 which codes the microtubule associated protein to unmapped map to see which codes for the total protein, 297 00:34:11,160 --> 00:34:15,330 which is fairly well known to be important in the pathogenesis of Alzheimer's disease and 298 00:34:15,330 --> 00:34:21,300 Parkinson's disease and microtubule proteins are very important in neuronal physiology, 299 00:34:21,300 --> 00:34:25,560 their importance in morphology, migration, axonal transport and morphogenesis. 300 00:34:25,560 --> 00:34:28,900 So it's entirely plausible that they should have effective phenotypes such as. 301 00:34:28,900 --> 00:34:37,960 This so we wondered that the disproportionate numbers of these microtubule genes that came up, was that merely a coincidence? 302 00:34:37,960 --> 00:34:43,130 So to try and establish whether that was the case or not, we performed a permutation analysis. 303 00:34:43,130 --> 00:34:49,090 So we looked at all 348 genes in a gene ontology called microtubule and skeletal organisation. 304 00:34:49,090 --> 00:34:57,310 We selected 150000 snips evenly spread across the entire genome, and we randomly permitted them into groups of four. 305 00:34:57,310 --> 00:35:05,390 And we counted how many times a three out of four of these groups of four would be in relatively close proximity to microtubule gene. 306 00:35:05,390 --> 00:35:09,910 We found that this happened in two hundred and ninety nine out of 10000 permutations, 307 00:35:09,910 --> 00:35:16,480 which tells us that there was an empirical probability of three percent of this occurring by chance alone. 308 00:35:16,480 --> 00:35:19,840 We then performed the snip based enrichment study, which is when we took all the snips, 309 00:35:19,840 --> 00:35:26,560 and Augustus showed some degree of association with handedness and correlated them with other genes phenotypes. 310 00:35:26,560 --> 00:35:30,910 And we found that nine out of 10 enrichments were for neurodegenerative phenotypes. 311 00:35:30,910 --> 00:35:35,170 You can see Parkinson's disease right at the top there, 312 00:35:35,170 --> 00:35:40,030 but then wondered whether there was any shared genetic architecture between handedness and these diseases. 313 00:35:40,030 --> 00:35:42,550 So we performed something called Elzie School Regression, 314 00:35:42,550 --> 00:35:49,930 which is basically we were comparing GWAS results with previously published G one summary statistics and looking for areas of overlap in the genome. 315 00:35:49,930 --> 00:35:55,300 And we found significant correlations with Parkinson's disease and schizophrenia, 316 00:35:55,300 --> 00:36:02,830 and interesting sets of schizophrenics are well known to be much more likely to be left handed. 317 00:36:02,830 --> 00:36:05,110 So now that we correlated handedness, Gina Typekit, 318 00:36:05,110 --> 00:36:12,160 the next thing we wanted to know was whether genotype at the handedness snakes could in some way affect your brain structure or function. 319 00:36:12,160 --> 00:36:21,070 So we teamed up with a team of brain imaging experts theme repair in Oxford to try and establish this relationship. 320 00:36:21,070 --> 00:36:27,940 And again, we used UK Biobank because the UK Biobank has detailed brain imaging for about 10000 participants, 321 00:36:27,940 --> 00:36:36,760 and our colleagues have derived the very sort of clever computational pipeline to take the raw imaging data and produce imaging to phenotypes. 322 00:36:36,760 --> 00:36:40,570 And these are distinct individual measures of brain structure and function. 323 00:36:40,570 --> 00:36:48,100 So this might be that the volume of particular brain areas where it might be the connectivity between two areas in the brain. 324 00:36:48,100 --> 00:36:49,780 So when we correlate, it's the genotype. 325 00:36:49,780 --> 00:36:57,190 At our fourth snips with brain imaging, we found that for the snip that was near the total protein gene and mapped gene on chromosome 17, 326 00:36:57,190 --> 00:37:01,810 there were many significant associations and these were particularly in white 327 00:37:01,810 --> 00:37:06,730 matter tracks such as the acute physical and the superior longitudinal cyclase. 328 00:37:06,730 --> 00:37:14,290 And these are tracks that connect language related regions in the brain, such as Broca's area and the plain and temporally and vulnificus area. 329 00:37:14,290 --> 00:37:23,110 So on this on this picture here, the blue represents the these language related tracts and the orange and green are the other grey 330 00:37:23,110 --> 00:37:31,450 matter areas related to language function and can see that they're being connected by the blue tracts. 331 00:37:31,450 --> 00:37:39,850 The next logical thing to do was try and close this triangle by correlating handedness with brain imaging, regardless of genotype. 332 00:37:39,850 --> 00:37:45,580 And here we found that left handers have increased functional connectivity between the right and left. 333 00:37:45,580 --> 00:37:50,470 Functional language networks and functional connectivity in this context refers to the 334 00:37:50,470 --> 00:37:56,620 temporal relation between two corresponding areas lighting up in the two hemispheres. 335 00:37:56,620 --> 00:38:02,080 And interestingly, can these language related tracks, such as the Acute for Cyclase, 336 00:38:02,080 --> 00:38:07,170 have been associated with schizophrenia and auditory hallucinations? 337 00:38:07,170 --> 00:38:13,290 So in conclusion, this is the first ever us to discover genome wide associations with handedness. 338 00:38:13,290 --> 00:38:18,870 And I'm going to caveat that by saying we have not discovered genes that definitely makes you left handed. 339 00:38:18,870 --> 00:38:23,640 We discovered a modest number of associations full of relatively small effect size. 340 00:38:23,640 --> 00:38:30,330 So this whole gene environment paradigm, the environmental factors are still predominant, 341 00:38:30,330 --> 00:38:35,970 saying that we did discover some interesting microtubule related genes and these genes have 342 00:38:35,970 --> 00:38:41,700 biological plausibility in contributing to differences in connectivity between language areas, 343 00:38:41,700 --> 00:38:49,380 which in turn might predispose to left handedness and certain neurodegenerative and psychiatric disorders. 344 00:38:49,380 --> 00:38:58,040 Thank you very much for attention.