Tune in weekly to our virtual series "Seventeen Minutes of Science" every Tuesday at 11am PST / 2pm ET where we go live on Facebook with a new guest each week to talk about how science and biotechnology is woven into their lives for (you guessed it) 17 minutes!
For episode 33 of 17 Minutes of Science we are pleased to be joined by Dr. Jeffrey Noebels from the Baylor College of Medicine. Dr. Noebels uses the mouse model but believes strongly in the benefits of the multiple model approach. His research focuses include gene control of neuronal excitability within the developing mammalian central nervous system, inherited neurological diseases, and epilepsy.
Dr. Janis Weeks (Host): [00:00:02] Looks like we're now live, so I'd like to welcome everyone to the first episode of 17 Minutes The Science of twenty twenty one. And it's a pleasure today to introduce my friend and colleague Jeff Nobles from the Baylor College of Medicine in Houston, Texas. His topic is gene mutations involved in epilepsy and cortical hyper excitability. Just a quick bio. He's currently the Cullen Trust for Health Care Endowed Chair in Neurogenetics. He's a professor of neurology, neuroscience and molecular and human genetics. He got his B.A. from Reed College here in Oregon, M.D. from Yale, Ph.D. at Stanford, and then did a residency in postdoc at Mass General and Harvard University. So as we'll hear about, Jeff is the leader in the field of neurogenetics, that is how gene mutations contribute to neurological diseases and how taking a genetic approach can help lead to new therapeutics.So welcome to the show, Jeff. I'm going to start my timer for seventeen minutes right now.
Dr. Jeffrey Noebels (Guest): [00:01:16] I just want to thank you for the invitation. I'm really pleased to be with you and especially with no PowerPoint slides. It's a treat.
Dr. Janis Weeks (Host): [00:01:23] Exactly. Yeah, it's like a real conversation. So let me start with this question. So you primarily use mouse models in your lab, so can you tell us how you got started using animal models for research on human diseases.
Dr. Jeffrey Noebels (Guest): [00:01:40] Sure so models – You know, I thought about it and realized that models have really defined my entire career. I began interested in getting interested in the brain by wanting to record from single neurons using glass microelectrodes. And I went to a laboratory at Stanford that was an expert in doing that technique. And we were studying epilepsy and I didn't really care about that, but I knew that it would make neurons fire a lot so I would get a lot of results. So we made epilepsy occur by putting a convulsive drug on the cortex of a cat, which is what the model was in those days. And I quickly realized that that's not what's wrong with people and, you know, why am I doing this? And so what happened next was really a defining moment, which was I went into the library and read two things. One was a paper by Lilly Jan describing some abnormal activity in Drosophila mutants, which she supposed was due to a mutation of the potassium channel, which hadn't even been cloned yet. In fact, we weren't even sure there were genes for channels. But I said, well, that that behavior that you described in the shaker mutant fly looks like what I'm studying in the cat brain. And I wonder if there are any other modern genetic models for epilepsy that I could study. And so I ran into another small little book called The Neurological Mutants of the Mouse, it's a catalogue. And they described a lot of spontaneous mutations that were being held at the Jackson Laboratory, which was then the Mecca of mouse genetics. Probably still is. And I went there. I said that that's what I'm looking for. And I went to Boston to the laboratory of Richard Sideman who was working with Pascha Orquiz defining mutant cerebellar mutations. And I said, do any of these mice have epilepsy? They said, we don't know. Why don't you find out? And I said, well, the only way to do that would be to do EEG recordings in mice, which had never been done in a wake behaving mouse. And I figured out a way to do that. And the funny part of the story is that I needed an EEG machine. And so right next to Boston was the Grass Instrument Company who sold all the EEG machines in the in the country at the time. And I wrote a letter to Ellen Grass, who was the president. Her husband actually helped develop the first EEG machines in the United States. And the next day a truck arrived at the laboratory with an EEG machine. And I used it to great advantage and found actually many mouse models of epilepsy.
Dr. Janis Weeks (Host): [00:04:34] Oh, that's a great story, especially with – that's so typical of the Grass family to do something like that.
Dr. Jeffrey Noebels (Guest): [00:04:41] Very, very generous. And then I began to experience some interesting questions that people who use models are probably familiar with, which is that people say, well, OK, that's just a mouse, it's not human or my model is better than your model. And then I realized that, well, you know, everyone thinks someone else's model is uninteresting and their own model is spellbinding. So I decided what I need to do is not make what might be another, but start looking at all the models together to develop systems of, you know, gradually of genes and that that saturate a certain pathway. So the more models the better actually as we try to explore the phenotypes in these mice. And even at one point I started a mouse sharing website in the early days, it was called mymouse.org, and we even had t shirts that said, "ask me about my mouse". So we tried to get the community to start sharing their models at an early point. But actually now the idea is very strong. And just recently at Baylor, someone has invented the something called match maker or Model Matcher, which is sort of like the gene variant servers that matched patient genes to basic scientists. So models are everywhere and they're great.
Dr. Janis Weeks (Host): [00:06:17] All right, well, you designed and led the Human Channelopathy Project at Baylor, along with the Human Genome Center. So, can you tell us a little bit more about that project, about human genes?
Dr. Jeffrey Noebels (Guest): [00:06:31] That was exciting. So in the early days of epilepsy (and it made a lot of sense), the mutations were in ion channels, voltage gated ion channels. And in fact, everyone thought, well, maybe that's the story. It's a membrane disease of excitability and it's channel related and so before a lot of other kinds of genes started to pop up, and we can talk about that, I thought, well, let's find out if in humans, people are walking around with ion channels and how many can we find. So I walked over to speak with Richard Gibbs, who was directing the Baylor Human Genome Sequencing Center, and said –and this was back in 2010 or so– and said, would it be too much to sequence some of these ion channels in maybe 100 patients or so and compare them to one hundred people without epilepsy? And he said, well, how many channels are there? And I said, well, you know, at the time there were about 300. And he said, well, let's do them all! So I said, Really? Wow, OK. So actually, we didn't get all the way there, but we developed a platform to sequence about 250 known ion channels at the time and looked at a large cohort of people with epilepsy and without. And actually what we found was really important, and that is that we all carried variants and ion channels, no question about it. Some of them are deleterious and others are just benign variants, but they're all there. And so our first hypothesis was that people with epilepsy would have more of them than people without. And when we looked at the data, we found that that wasn't true. In fact, the loads of ion channel variants were about similar in both populations. So that was an 'aha' moment because that meant that it's really which ion channel variants you have and what pattern are they in. Because what's cool about ion channels is they modify each other, they all are contributing to whether a cell fires or not. So if you have too much of one, a little less of another, you might be fine, which is probably true for most of us. But if they add up in the wrong direction, you can have a big problem. So that was a major contribution because at the time people thought the minute you detected the variant in the gene, you're doomed. We said, no, no, no. You know, really this will confound this simplistic idea that, you know, there's one gene, one disease. And the minute you find a variant in the gene that there was something wrong with it. So it was in the early days, but I think made an important point.
Dr. Janis Weeks (Host): [00:09:16] Yes. So more recently, your lab discovered that mouse models of Alzheimer's disease show a non convulsive cortical hyperexcitability. Could you explain a little about what that phenotype is and what the implications are for understanding Alzheimer's disease?
Dr. Jeffrey Noebels (Guest): [00:09:33] Sure, yeah. So that was another kind of paradigm shift and it was all based on models, again. So as a neurologist, I was interested in epilepsy, not Alzheimer's disease. I didn't know that much about it, believe it or not. But a friend of mine, Leonard Mooky, who trained with me, was an expert in Alzheimer's disease. And one year we met at a neuroscience meeting and he said, Jeff, we've been doing a project in our lab and we found some wiring changes that you could show with neuroanatomy and neuropathology stains that look like the same changes you see in the brain with epilepsy in our mouse model of Alzheimer's, the J20 model. And he said, do you suppose they could be have had seizures or what? And I said, I don't know. But if you send us some mice, we'll certainly have a look. So we did. And lo and behold, I remember the day we plugged in the mouse and we're watching it walk around and all of a sudden we saw seizures on the EEG machine, but the mouse seemed to be behaving normally. And that turned out to be what we call a non convulsive, or even electrographic seizure, that didn't seem to perturb behavior that much. And it turned out after a lot of study that it was really probably originating in the hippocampus and that the reason.... So, of course, we took this in a very excited way. There were a couple of paradigm shifts here that we had to convince people of. One is the first reaction is that, well, this is a neurodegenerative disorder, that's not what epilepsy is. And to that, I would explain to people, well, you know, it's a balance between inhibition and excitation, and as cells die, if you lose more inhibitory neurons first, you can have epilepsy, at least for a time period. So people accepted that as possible. But then they said, well, you know, I've taken care of 80 patients for 20 years and none of my patients have epilepsy. That turned out to be a mistake too, because it turns out that the younger you are when the dementia begins, the more likely you are to actually have seizures. Very late onset seizures, people thought, well, that's not related to the disease, they're just getting old. But that turns out to be three times more in people with Alzheimer's. So there really was a human link between seizures and Alzheimer's disease. And more recently, a colleague of mine at Mass General Alice Lam, used a very clever technique to record from the hippocampus of people with Alzheimer's disease, which, you know, for most purposes you would have to drill a hole in the skull, which nobody likes. But she found a way of putting electrodes through the foramen ovale, a small hole in your cheekbone and getting the electrodes close enough to the hippocampus. And lo and behold, she found hippocampal seizures in patients with no sign of EEG abnormality on the scalp and off. So that that turns out to be now a very robust finding, both in animal models. There are about 100 mouse models of Alzheimer's now, all different kinds. They all seem to show hyperexcitability of one form or another. And in patients, the more we look carefully at them, the more we can find epileptic form abnormalities. So what's important about this is one, follow the data no matter what people tell you. And the second, of course, is that maybe this, if there's a vicious cycle and we think that there is between hyperexcitability and amyloid deposition and just cellular cell death, that that's a totally different kind of drug that would calm down the brain might be another way of slowing down the progression of dementia. So we're very excited about that discovery.
Dr. Janis Weeks (Host): [00:13:40] Yeah, that's a great story with real potential application to patients. So we have about four minutes left, so I'm going to end with a big question here. Can you give us your view on how human neurogenetics is going to be or is and will continue to be contributing to personalized medicine?
Dr. Jeffrey Noebels (Guest): [00:14:02] Well, that's a great question to ask and Epileptologist, because right now, human and neurogenetics and personalized medicine has just exploded in the field of epilepsy, in particular with the pediatric epilepsy. So when seizures happen in babies, sometimes even in the delivery room, but, you know, within the first two years of life, it's very abnormal. And now almost all these patients, if they're in the right place, will be sequenced with clinical exomes. And by virtue of that, and comparing their exome with their parents, there's been an enormous discovery rate of de novo variants that actually cause the disease. And each one of those is, now that other people are getting diagnosed, they immediately said, "well, OK, now what drug should I take? And how can we cure this disease?" Because we always thought, oh, if you got the gene, you can cure the disease. Well, so what that's sprung is a huge number of patient family advocacy groups that want models of their child with the exact same mutation and made and then used as a test system to find therapy or somehow reverse the disease if possible, so thats... it's more than a cottage industry. Now, that is the way things are moving forward, that people are getting personalized models made of their gene. They put [the genes] into mouse, into zebrafish, various animals that allow you to rapidly screen the effectiveness of certain drugs and find out, you know, importantly which drugs might be better, but also which ones to avoid because some medicines make you worse. So we're beginning to really put these pieces together and it's so gratifying to watch this happen. I don't know where the end [is]... we have not reached peak model, you know, "peak oil", they're going to be so many more genes and so many more models needed to really get a grip on these diseases.
Dr. Janis Weeks (Host): [00:16:05] So are you involved in using that kind of approach for patients you have there at Baylor?
Dr. Jeffrey Noebels (Guest): [00:16:11] Well, yes. I mean, we vary – if you take a deep dive into the biology of what those genes do, you don't actually need to study every single personalized variant. But you, you know, start asking questions like 'is it a loss of function or a gain of function?' On the other hand, channels to more than just pass more or less current. They also, like other proteins, interact with other proteins and so there are nonpoor functions of every variant that need to be explored to find out what they do and also which circuits these occur in in the brain. So we have a mixture of mice that have been personalized with a, quote, "humanised mutation" and also other ones that are just, you know, knockouts or things like that that we use with cre-lox systems to actually target the location of the mutation into the certain brain circuits. We're learning a lot that way.
Dr. Janis Weeks (Host): [00:17:14] Great. Well, we are just about out of time here. Anything else you want to add in the final minute that we didn't touch on?
Dr. Jeffrey Noebels (Guest): [00:17:22] There's too much!
Dr. Janis Weeks (Host): [00:17:23] I mean, I know we could go on for hours.
Dr. Jeffrey Noebels (Guest): [00:17:26] Yes, and I usually do! Well, I just want to say what a pleasure it is to to join you for something like this. I think we need more conversations between scientists and other interested people to learn more about what's going on. It's so exciting these days in science.
Dr. Janis Weeks (Host): [00:17:46] It is! Well, on that note, I want to thank you again. We've got 15 seconds here. Thanks again, Jeff, for taking time out of your busy day to spend some time with us and to all of you out there watching. We'll see you again in a week for the next episode of 17 Minutes of Science.