I spoke with Prof. Andrea Meredith about BK channelopathies, her new KCNMA1 foundation, treatment strategies for KCNMA1 patients, her ideal lab, the value of review articles, Twitter for scientists and even sewing. That was a great and motivating conversation and I’m sure that many people will like Andrea’s analogies between sewing and patch-clamping among other things.

Have a good read.

---

---

So, Andrea, could you tell me your story, your ion channel story? Where and how did you get into ion channels?

That’s actually a great question for me because I didn’t do my graduate training in ion channels. As a graduate student, I was actually well trained in molecular biology in a lab researching neural development. And I liked molecular biology, I really liked the precision of it. But I was not in a very quantitative field, and so I found myself really sort of yearning for a much more quantitative approach in neuroscience. My rotation in a C.elegans lab, where I initially learned how to do electrophysiology strengthened my inclination that I really wanted to be doing something more quantitative like electrical signaling. But what kind of sealed the deal for me is when right after the KcsA structure came out, I invited Rod McKinnon to give a talk at UT Southwestern, where I was a graduate student. Each year graduate students got one slot to invite an outside lecturer, and I happened to be head of the committee that year, so I invited Rod. And that really changed my direction entirely. First it was reading the paper and presenting it as a journal club. I really got in touch with the elegance of protein design and how perfectly different parts of this channel basically mediated its function. Nowadays the mechanism of ion selectivity is a classic mechanism that we teach in every ion channels course, but to actually be a graduate student when that mechanism was revealed for the first time, it just completely hooked me. Then Rod came and gave his lecture, and I was even more hooked. By the time I graduated, I knew I wanted to go further in this field. Then, I did a postdoc in Rick Aldrich’s Lab at Stanford University, where I started to work on BK channels. Rick convinced me that no one had made a knockout of the BK channel yet, and so that was really the first thing that I should work on. I ended up doing that – I made the KCNMA1 knockout in mouse and studied the function and physiological role of BK channels in his lab. Then, thinking about all of that, I didn’t realize it at that time, but I already sort of formed the idea of what I wanted to do for my entire career in ion channels and that’s basically what we do in my lab now.

Yes, I see that you really stick to that idea. Since then, your whole scientific career has been revolving around BK channels.

Yes, but I think at this point we’ll probably branch it out because one of the interesting things for us is that BK is a calcium-activated channel, and so one of the major drivers of how BK controls excitability in cells is its calcium source.  That’s something that we’re going to research in-depth in the circadian system. The circadian clock is really interesting because there are two different calcium sources BK channel relies on. During the day, its activation is more driven by L-type calcium channels, and at night it’s predominantly release through intracellular RYR receptors. This is the first example I can think of where there is a dynamic switching of BK with its calcium source. And we understand the signal that produces it, which is the circadian clock, which brings together several mechanisms that change the way the BK channels participate in setting action potential firing rate during the day versus the night. BK’s expression oscillates, and this allows the channels to have at time of day specific effect on firing rate in that system. There is also diurnal regulation of BK’s biophysical properties that involves splice variants and beta subunits. So this type of dynamic regulation becomes one of the key things that I want to understand across all systems, especially ones with rhythmicity over different time scales.

It’s super interesting what you tell. Such a complex regulation and you are talking about it with such a passion. I’m curious, what motivates you to do your research? Is it just your internal curiosity?

Well, my sort of core motivation for a very long time was really just understanding the biophysical mechanisms of information coding and understanding which ion channels would be dominant during action potential signaling in such a way that would allow you to vary that action potential signaling and vary information coding. I was always super motivated to just drill down into those mechanisms and start to understand the generalizations that different systems use for information encoding. I would’ve been perfectly happy to do that until the end of my career but then…

…the newspaper article came along…

Yes, in August 2018 I just stumbled upon the New York Times Magazine article about a little girl, Kamiyah, with a rare undiagnosed genetic condition. I thought I recognized some of the Kamiyah’s symptoms as being described in the literature associated with the mutation in BK channels.  She has a particular type of dyskinesia that is sometimes called a ‘drop attack,’ and it has a very distinctive presentation. The follow-up article confirmed that Kamiyah had a mutation in the KCNMA1 gene that encodes the BK channel. At first, I was really uncomfortable with getting involved. I thought that as a basic scientist, I had nothing to offer. But then I realized that they were basically talking about precision medicine and a gene mutation of unknown causality, and they were talking about it as if the discovery of a gene mutation would explain everything. I knew that it wouldn’t. I felt compelled to actually weigh in with what I knew, which is that just because you’ve discovered a mutation in a genome, it doesn’t mean that you understand the mechanism of a disease. When the patient is getting a whole-genome or whole-exome sequencing and comes up with a single gene mutation, the immediate assumption is: “It’s causative”. Well, is it? There are labs that have basically been studying this for years. One is Jeff Noebels lab. He had a Cell paper several years ago asking: “Is there a constellation of single nucleotide polymorphisms in ion channels that actually can be predictive for seizure?”. The conclusion was basically “no”, we’re still not there yet in terms of understanding the landscape. I felt like I had to explain this to the people that wrote that newspaper article, and then later to Netflix.

Being involved with patients has really been a game-changer for me. For many years, most of my time has been spending in the lab at the rig very focused on the narrow mechanisms that we’ve been interested in. Then suddenly Kamiyah’s story came along, and this just opened up a whole different world for me. Now we have a cohort of patients, and part of what I’ve been trying to do recently is pivot towards that in ways that I would’ve never considered at earlier points in my career. We’ve basically changed the direction and invested almost everything in studying patients’ mutations. It was scary to me to spend money on this line of research when there was no guarantee that it was going to have a return. I decided this was absolutely what I wanted to do. I spent as much money as I had, and now I’m writing the grants to get it funded and continue doing it. Hopefully, the grants will get funded.

I hope it will. This story is so motivating. Could you tell me where you are now in your study of patients’ mutations?

The first most important thing for us was to basically understand whether Kamiyah’s mutation is causative. At this point, we actually are convinced that her mutation is causative, and there are multiple reasons for that. One is that her mutation ends up being the most common de novo mutation in the entire patient population and across this set of patients they have a very similar phenotype. The patients are sort of telling us that there is evidence for causality. The second is it is a really strong gain-of-function mutation. I think that I’m going to end up being surprised and sort of blown away when we look into the animal model about what the different neuronal and muscle populations are doing when they have this mutant BK channel there. Because this gain-of-function mutation is so strong that in some cells it may hyperpolarize the membrane so that they can never even get to the threshold. But in other cells, having more rapid activation, it could actually enhance firing because it could enhance the repolarization phase of the action potential. I’m so excited to get in there and figure it out because it could end up informing us about why is that some people that have the one type of mutation have more of the seizure phenotype and less of the dyskinesia phenotype whereas most of the people with this other mutation have very strong dyskinesia phenotypes.

Along with this, another part of what we are super interested in figuring out is why different patients carrying the BK mutation can have different types of seizures? Why does one patient have a grand mal seizure and another patient could have an absence seizure? Is it that the mutation is expressed differently between those two patients? Is it the cell type that is more susceptible to that mutation? Is it the age of the patient, or maybe some other mutations or genetic variation that they may carry in their genome?

And what about the circadian system? As BK expression is modulated by the circadian system, there could potentially be a relation between KCNMA1 patient symptoms and the circadian rhythm. These seizures, are they different during the day and at night?

It’s certainly conceivable that the circadian clock is gating some sort of propensity toward seizures at certain times of day. But I will say that the jury is still out on that, and hopefully animal models will help us to better understand the intersection between sleep and seizures. We now have a cohort of patients that we’re doing genotype-phenotype analysis on and many of the patients are reporting some sort of sleep problems. However, very few of the patients have had chronic EEG monitoring, so we have very limited information on their nocturnal behavior.

OK. And what are the existing therapeutic approaches for KCNMA1 patients? If we take, for example, gain-of-function KCNMA1 mutation, the first thing that comes to my mind is to try BK inhibitor.

No, it’s not that simple. Maybe for some channelopathies this could be relatively straightforward, but for BK I don’t think it’s straightforward in the slightest.

BK channels are regulated by so many different factors that you can think that BK channels represent a different entity in every cell type they are acting in.

ANDREA MEREDITH

You know, BK channels are regulated by so many different factors (we can talk about beta subunits, alternative splicing, lipids and other intracellular factors) that you can think that BK channels represent a different entity in every cell type they are acting in. One of the most interesting features of BK channels is that in some cells they can actually increase firing rather than decrease firing. The classical role for BK channels is when they act during the afterhyperpolarization and mediate a suppressive effect on firing frequency. But when a BK channel participates in repolarization of the action potential, that channel can actually enhance repolarization so that the cell can fire the next action potential more quickly. This could be the type of channel that is potentially deranged in patients with seizure-linked mutations.

Several big pharmaceutical companies actually had drug development programs targeting BK for urinary incontinence, blood pressure, erectile dysfunction. You basically want an agonist in those cases. You want to be able to lower blood pressure, you want to dampen the excitability of smooth muscles for all of these phenotypes and that would be done with an agonist that opens the channel. The problem is that the opening of BK channels may be associated with seizure activity in certain parts of the brain. Obviously you don’t want to deliver a drug that relaxes peripheral smooth muscles, but then if it crosses the blood-brain barrier, you’re basically seizing. So what would you do about that? You either generate a drug that doesn’t cross the blood-brain barrier, so it’s just peripherally acting, or you try to understand more about the types of channels that are in the target tissue.

I’ve talked to neurologists and patients, and they tend to think that because there is a basic research lab working on this problem, it’s just a matter of time before somebody comes up with the drug that corrects the channel function. I would say that this is not the case. For seizures you want a centrally acting drug, but without knowing the locus of seizures, which may differ according to the mutation that the different patients carry, you have no idea what to target with this “precision” drug. Sure I like the idea that there would be one day when we’ll be able to come up with super-specific drugs that basically only bind a particular splice variant with a particular beta subunit with a particular level of activation on it from a particular calcium source. ‘Precision medicine’ is a noble goal, and there are companies working on this.  But this is not where we are at now. People have been having seizures for millennia, and we have highly effective drugs for seizure. In my opinion, all of these patients should basically be trialed on standard anti-seizure medications because those medications are targeting broad swaths of the brain that are known to be involved in seizures, and they have well known safety profiles.

Another point is: what about the gene correction? The parents now are relatively savvy. They know that there is gene therapy now available for some limited subsets of disorders and that this gene therapy can be used for single-gene mutations. So, obviously, they are interested. But I can’t recommend this approach as a priority at this point. I certainly wouldn’t do that with my own child.

Why not?

I don’t think the technology is there for gene therapy for these types of brain disorders yet. I can see delivering something into peripheral muscles like for muscular dystrophy, for example. I can see delivering it into the lung for CFTR but I cannot see it for BK – an ion channel that is widely expressed in the brain. There are too many unknowns. That’s not to say that we won’t get there, and obviously, the goal of all the research that we do is to find a cure. Not just a therapy, but a cure. I feel some responsibility to raise the real stumbling blocks and challenges along the way and be the expert on this issue.

Yes, that’s a complicated story. And in order to sort it out you certainly need a strong team. Could you tell me about your lab?

Yes, I have a tiny lab of super-hardworkers. Two postdocs, two graduate students, and a medical student right now. As my lab is so small, I’m still intimately involved in experimental design and analysis. Also, some of the human genetics that we’ve been analyzing and doing are actually on my plate. That’s fun for me because I get to go back to that genetics background and keep my skills up in that area. With my current academic responsibilities, it’s hardly possible to carry out electrophysiological experiments myself, but I guess the true joy of science to me is really in experimental design and interpretation. I’m side by side at the rig with my people every single week looking at the data and turning it around in different ways to analyze it and designing the next experiment with them.

So, as your lab is tiny, and you still have to do genetic experiments from time to time, probably it’s not how you see your ideal lab. I’m curious, what’s your view of an ideal lab?

Hmm. You know, it’s not that different from what I have. The number of people is almost irrelevant. It’s more about the depth that each person takes in their project. The ideal lab to me is a set of very focused trainees or experts, each having a different piece of the puzzle. Right now, I have a postdoc from Chile – Hans Moldenhauer – and he’s an expert biophysicist and biochemist. Then I have another postdoc, Su Mi Park, who actually came from the animal model and neurophysiology side. So I almost have the dream team right now, because pairing these two postdocs together in a time when we’re analyzing causality for the patient mutations actually gives me almost perfect suite of skills within the lab. I say almost because now we have a postdoc position available in the lab and I encourage anyone interested in joining this “Team Science” to apply. The ideal lab to me is where each person brings a background expertise that clicks together like a piece of the puzzle, so that together we’re an ensemble that is more than the sum of the separate pieces. That puzzle can either be tiny or it can be a very big intercut puzzle. The critical thing is that all of the pieces lock together properly.

Well said. I agree with you that people are the most important asset of any lab. But you didn’t say anything about technologies. These are essential as well. Are there any novel technologies (e.g. automated patch clamp) that you are using in your lab?

Well, no, we don’t have an automated patch clamp robot in the lab. You know, we’re more in the qualitative and the mechanistic side of things, not the screening side. Would I like to have one in my lab? Absolutely. I can think of a gazillion projects that I would do. But the problem is unless you are funded for it, it’s not going to happen. But there is a side of me that wants to work on that scale. Going back to my ideal lab, one of those pieces of the puzzle that I would click into place would be that I would love to have a rapid response component to my lab where we have some sort of high throughput assay, be it at the level of the channel or the circuit. In my lab, we work with multielectrode arrays. At this point, we have a fourplex system so that we can record from four different brain slices simultaneously. Each of those probes has 16 electrodes on it, so we have 64 electrode amplifier and we can split those electrodes out however we want. But I would love to be able to do a high throughput sort of interrogation of different circuits.

Oh, it’s interesting to know that you’re attracted by high throughput assays. Have you ever thought about working in industry?

Well, actually, yes. You know, there is some frustration sometimes with the deep mechanistic work that we do in academia. It’s slow and labor-intensive. There was a point in my career when I was really seriously thinking about moving to industry. I wanted to work at scale, and I wanted to work very rapidly. The speed at which you can move and practical decisions that you make in industry are attractive to me. Early in my career, I actually applied to a German biotech company that had a presence in the US, and I got interviews, and it was something like “write your own dream job” kind of thing. But then all this dragged on and sort of petered out, and eventually, they shut down the US side of the research. So, had I gone to work for them, I would’ve been fired almost immediately. That was a valuable experience for me of what the industrial reality could look like. Since then, I am at peace with my academic position.

OK, so you stayed in academia. But, do you have any interactions with companies through collaborations or consulting?

No, I don’t have any industrial collaboration at this point. I would say that I’m amenable to it, but I think that there are certainly some bridges to be built between academia and industry. I think that academic scientists are not very savvy about different aspects of academic-pharmaceutical partnerships. I really would like to get savvier, and maybe your ionchannellibary.com could be helpful here.

As for consulting, I actually have some interesting consulting experience. Some years ago, I was a consultant for a wearable technologies company in the Silicon Valley. They make a digital eyewear product, and I consulted for them with my circadian background. That was really interesting, and for about a year I was having regular meetings with their hardware and software developers discussing different capabilities they could integrate into their device. That was really fun and enriching for both sides. I’m very positive about consulting, and I’m always open to new opportunities in this area, even beyond ion channels.

And could you tell me about your foundation? What are your goals?

When I got involved with KCNMA1 patients, I met neurologist Sotirios Keros who became the co-founder of the KCNMA1 Channelopathy International Advocacy Foundation (KCIAF) with me. It was largely his idea to recognize we needed this because he had already a background in nonprofits and because he deals more with patients he understands what patients need in terms of information and access. So, we decided to create this foundation together to cover both the patient side and the basic research side. My goal in the foundation is basically to advance scientifically correct knowledge about this disorder. What we find is that on the Facebook site, or in interactions with patients, there are a lot of different things that get discussed. Patients and families don’t know what level of evidence there is behind a particular statement. When one person takes a particular drug and has a response to it, what does that mean? Should my kid take that too? That’s not easy to build evidence. I’m the scientific side of the foundation, and that’s really important to me. Now the patients have a valid “official” place where they can find the information they know is correct.

That’s a great initiative. And how do you fund activities of your foundation?

The short answer – we need funding. We’re not big enough as an organization yet to write grants. There are other epilepsy foundations that are big enough to apply for NIH grants or to apply for foundation grants. We are not there yet, but the hope is that we can grow into that one day. What we basically need is the involvement of the patients and their families and maybe a little bit of seed funding to get to that point. But it’s going to be a multiyear process. The foundation has existed for barely a year, but we’re really working to get it off the ground. I don’t know that our patient population is ever going to be big enough where we can fund our own grants. We’ll see. If you want to support our foundation you can make a donation through the foundation website.

And what’s the general situation with the funding of research in rare diseases?

Now it’s a difficult time for rare disease research – it’s kind of weird to work on rare diseases in the time of coronavirus. It’s never been an easy thing to find the money for rare disease research. So, although we have a lot of biotech or pharmaceutical companies that have specialized expertise and approaches to target rare diseases, the problem for them is that the patient population in most cases is going to be so small ultimately that the specific rare disease drugs aren’t going to end up being humongous revenue generators.

So, the question is: if you develop a drug that hits a particular channelopathy, will that drug then also be applicable to other channelopathies? So, you see, the way we are thinking about it is: we can approach rare diseases if there end up being commonalities.

That’s one of the things that I like to think the most about. As we go forward in our understanding of BK-linked channelopathies, does that then inform how we think about any other seizure-linked channelopathies? Dyskinesia actually has relatively few drugs that are used to treat particular types of movement disorders, so understanding how BK channel mutations produce dyskinesia could end up being something that opens up the dyskinesia field.

That’s an interesting perspective. I’ve never thought about it in this way. Well, my next question sort of falls out of the discussion, but I was looking at your articles in Pubmed and found out that your first review article was published in 2019, and now you came up with your second review article. Just two reviews – that’s not much. What are your thoughts on the importance of review articles?

One of the things that irritates me is the proliferation of dogma in fields. I think review articles inadvertently can contribute to that because they summarize mechanisms, ideally in helpful ways, but sometimes in unhelpful ways. For example, there was one time I was asked to review a paper in a circadian field, and I went back to the editor and I said: “This field even doesn’t exist yet. The field basically consists of three papers.” I basically told them, there is no value to publishing this review. You can’t review data that doesn’t exist. Otherwise, it’s just a hypothesis paper, right? Much to my annoyance, they published the review anyway. So, as part of my argument with the editor, I looked at the number of reviews that are published in the circadian field versus the total number of publications, and I compared it to the ion channel field. The number that I came up with in the circadian field was something like a third of all publications are review articles. That does not help the field. The number of reviews in the ion channel field is about right – it’s somewhere less than 10% or whatever. But I don’t think that it’s appropriate to publish a review until you have a scientifically valid set of data to write that review on. So, for the first time in my career, this year I felt that I’m at the apex of that. So we had the KCNMA1-linked Channelopathy review, and then I wrote a Physiological Reviews on circadian regulation of ion channels. But what’s the point to writing a review before that? Actually, the great model for this is Rick Aldrich. I think Rick has written a grand total of two reviews in his career. If you write a review, it has to be iconic. It has to be seminal.

And what about social media? I could say that you’re rather active on Twitter. What’s the benefit of social media for you?

Well, actually I was a latecomer to social media, I’m very private. Up until a couple of years ago, I never had a Twitter or a Facebook. I was on LinkedIn, but I consider LinkedIn sort of a professional thing. In other social media, the boundaries are a little bit blurred, and I’m uncomfortable with that. But as part of finding the KCNMA1 patients, I basically had to come into a Facebook group. I also created a lab Twitter account, and asked my postdoc Hans: “Does anyone ever use Twitter?” And he’s like: “Oh yeah, I use Twitter all the time to find out about the latest papers”. So, I started routinely checking the Twitter feed and connected to different ion channel labs. I realized that it’s actually a really good way to have it filtered into your consciousness. It’s not just a cold citation tracker that’s giving you all of these different ion channel records. The people that I’m linked to through Twitter are my ion channel friends, and so there ends up being a level of personal satisfaction seeing them get their papers out. Then, I noticed that a lot of my female colleagues are using it to promote each other by, for example, retweeting their papers on Twitter. And so I’ve become even more committed now to Twitter, making sure that my female friends are getting maximal visibility. I feel like that is making me a better scientist and a better advocate.

So, as you mentioned women in science, has being a woman caused any challenges in your career?

Ahhh, yes. Yes, I think it would be nuts to not think so. I think that women have much fuller plates. We take on so many different roles besides just our scientific role, and it makes it incredibly difficult to be focused solely on job success. It’s a lot of hard work, and so advocacy ends up being a big issue for women because anything that can make your success greater as a scientist, or life easier at home, needs advocacy. But one of the things that I think is good about the ion channels field is that it feels relatively egalitarian. I’ve had excellent interactions with both my male and female colleagues. Again this may be bolstered by the use of social media, but there is a critical network of women in the ion channel field that all support each other and that are all relatively well supported by our colleagues.

That’s not to say that there haven’t been problems, there are certainly incidents of gender discrimination in this field, but the field as a whole, I think, is welcoming and egalitarian for women. We need more women in electrophysiology because the type of science that we do, the problems that we approach, all of these things are made better by having diversity. Lately, I’m welcoming the fact that structural biology, for example, is becoming more integrated. Vera Moiseenkova-Bell, Crina Nimigean, Sudha Chakrapani, Vasanthi Jayaraman, I can read off a long list of names of excellent structural biologists that are women. So, that’s just what we need. We need women in every corner of the field. Just why not?

100% agree. So, we’re approaching the end, and so maybe you can tell me some interesting facts about you? For example, I know that you can use a sewing machine as an electrophysiology rig. I’d really like to know more about it.

I would actually love to talk about this. I think one of the reasons why I love electrophysiology is because I learned to sew as a child. You know, sewing has so many commonalities with what electrophysiologists do. First, you work with your fingers, that’s really what we do when we dissect the neurons or are patching cells. Just think about threading a needle. Isn’t that the same as filling the patch pipette?

When I was a young child, I learned how to use a sewing machine. My mother worked for a dress shop and she had two different types of sewing machines. She had a regular sewing machine that just had two threads and then she had an industrial type of sewing machine (with 6 different threads) capable of sewing a seam and overlocking it at the same time. So when you sit down at the sewing machine, you are effectively setting up a rig in the same way that you set up an electrophysiology rig. It’s funny, because I didn’t even think about it until I recently sewed face masks for my family. I was sitting at the sewing machine, and I just suddenly realized that it feels like sitting at the rig. Really, the sewing machine is effectively an electrophysiology rig.

The other thing is when you’re setting up the sewing machine, you’re threading the fabric through and making decisions in real-time. You are sewing the seam, and you’re tracking along, and then you get to a curve or you get to a fold and you have to pivot, you have to make decisions in real-time. Especially if you are sewing on one of these industrial sewing machines that’s actually cutting the seam, sewing it, and overlocking it at the same time, every decision has to be made quickly. I loved that about sewing, and it’s the same real-time process I use at the ephys rig. I did embroidery and some fine needlework, so there’s also a creative side, but then there is this industrial side to me that was higher throughput on the big machines. I told my mom that I think that some of my technical abilities directly came from learning how to sew and the precision of that sewing. She was blown away. She couldn’t have imagined that. You know, in the same way that kids would engage in chemistry sets or whatever, why wouldn’t a complicated sewing machine with 6 different thread paths be the same type of technical stepping stone?

The sewing machine is effectively an electrophysiology rig.

ANDREA MEREDITH

Sewing has been a lifelong hobby for me. When I was in high school, I actually sewed many of my own clothes, and even though I’m a professor now and I can afford to have alterations done at a dress shop, I still do it myself. It’s not about the money, it’s all about the “patching”.

---

---

I’m thankful to Prof. Andrea Meredith for taking the time to talk with me and sharing her motivating story.

---

Images by Andrea Meredith, Artem Kondratskyi as well as Every Angle on Unsplash.