Alexander Disease Research Update
Discussion of recent publications from the research literature on Alexander Disease, hosted by Albee Messing from the University of Wisconsin-Madison.
Alexander Disease Research Update
Alexander Disease Research Update - Episode #1: genetic anticipation?
Use Left/Right to seek, Home/End to jump to start or end. Hold shift to jump forward or backward.
Albee Messing and Michael Brenner discuss:
Hunt, CK, et al. Does genetic anticipation occur in familial Alexander disease? (2021). Neurogenetics (in press)
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8241638/
https://medlineplus.gov/genetics/understanding/inheritance/anticipation/
Send your questions to: axdrupodcast@waisman.wisc.edu
Help support research on Alexander Disease at the University of Wisconsin-Madison's Waisman Center:
https://alexander-disease.waisman.wisc.edu/donate/
Hello and welcome to Alexander Disease Research Update or AXDRU. I'm Alby Messing, and this is episode number one, the first of which I hope will be a regular series of discussions about newly published results from the research literature. With me today is my long-term collaborator on GFAP and Alexander Disease, Michael Brenner. And our topic will be a paper by Hunt et al. entitled, Does Genetic Anticipation Occur in Familial Alexander Disease? Which was published, which is in press in the journal Neurogenetics. But since this is the first episode, for those of you who don't know me, I should introduce myself. I got my training in experimental neuropathology at the University of Pennsylvania, and then for 30 plus years taught, ran a lab, and for several several years ran a large research center on intellectual and developmental disabilities at the University of Wisconsin in Madison. Along with Mike Brenner, I began working on Alexander disease around 1995, and it's been the sole focus of my lab since 2000. Mike, do you want to say a few words about yourself?
SPEAKER_00Sure. So I trained in biochemistry, and that was at the University of California, Berkeley. I then transitioned to molecular genetics and studying the controls of gene activity. For the past 30 years, our gene of interest has been the one encoding GFAP glial fibrillary acidic protein. And except for the initial three years, this has been in a collaboration with Albi. It was in the course of this study of GFAP that we discovered that mutations in the GFAB gene are responsible for most cases of Alexander disease.
SPEAKER_01Thank you. So, and thank you for joining me today on this first podcast. So I want to make another general point about the goals of these podcasts before we dive into the paper itself. Looking forward, I view these this as a good opportunity to try and answer questions about Alexander disease. So if you wish, please email your questions to me and we'll try to address them in future episodes. I'm not a physician, however, so getting answers may have to wait until we have the appropriate person on as a guest. The email address for questions is A M E S S I N G. That's just my first initial and last name at Wisc.edu. Now let's turn our attention to the main topic for today, the paper by Hunted Al, which asks whether genetic anticipation occurs in Alexander disease. And I want to start by just offering a brief definition of anticipation, which is a worsening of the phenotype in progressive generations. So in this paper, um the authors describe a mother-daughter pair. The mother is 71 years old when she present presents to the clinic with a four-year history of weakness and recent dysphagia. At presentation, she was wheelchair dependent. On examination, she had extreme weight loss, prominent palm mental reflexes. I honestly don't know what those are. Spastic dysarthric speech. Tone was increased in all limbs, and there was a pyramidal pattern of weakness. Brain MRI demonstrated cervical medullary atrophy, suggesting a diagnosis of adult onset Alexander disease. Her daughter presented to the neurology department at age 45 with an eight-month history of stiffness in all four limbs and poor balance. She had poorly responsive pupils, prominent jaw jerk and slight dysarthria, abnormal speech, slow tongue movements, marked spasticity in all limbs with brief, brisk reflexes, and normal strength except for one particular muscle, dorsal interosseous muscle. There were no lower motor neuron features, no cerebellar signs, and sensory exam was normal. MRI scans of her brain were done and were reported as normal. So this is at the age of 45. Her condition progressively deteriorated so that over the course of six years, she um eventually had to retire from work, um, and then she was lost uh to follow up. So together this mother-daughter pair suggested to the authors that uh that there could be uh there was there was a worsening in the phenotype in the daughter, and that this could one explanation for this could be genetic ancestry anticipation. I should say that um DNA analysis of the mother revealed a mutation or a variant in GFAP, uh which was uh asparagine 386 to serine.
SPEAKER_00So can I break in?
SPEAKER_01Please do, sure.
SPEAKER_00So I think the DNA analysis was done. So the daughter they said was lost to um examination, but then then reappeared um four or five years later.
SPEAKER_01Right.
SPEAKER_00Um, and I think it was the daughter's DNA that was analyzed, not the mom's. Um, and an important point for this paper is the daughter was seen and diagnosed before the mother. And that's one of the things that led them to think about this um genetic anticipation.
SPEAKER_01Right. So then what they do is conduct a literature search and try to capture all examples of parent child uh pairs with a diagnosis of Alexander disease and ask the question of whether the phenotype is worse in the child than in the parent. Um their sole uh criterion is age of onset. Um so they do a fairly extensive um literature review, uh, which is presented in table one. I honestly could not figure out why they ordered um all of the patients in table one the way they did, um, because it was somewhat laborious going through to figure out um which one was which. It's not done by the variant in the sequence from amino terminal to C terminal, it's not done by chronology of publication. Um, it doesn't seem to have any real rhyme or reason to it, but it's fairly comprehensive. Um nevertheless, it's not it's not complete and it's not entirely accurate, which becomes a problem later. Um you know, I know for a fact that one of they cite as one pair um a uh a pedigree that we published together in 2012 with the D417A variant. Um but they omitted another family reported in the same paper, which was with which had an S247P variant, um, both of which are relevant for their argument, if you you know, if you follow their logic. But what do you think about the data in this paper and how reliable it is for judging differences between severity?
SPEAKER_00Are you you're asking me?
SPEAKER_01Yeah.
SPEAKER_00Um the question I have is there are two things. One is once a parent has been diagnosed, um you'll be more likely to pick up an earlier age for offspring because you're looking for it. And often in Alexander disease, um, the initial symptoms are very general. They could be almost any kind of disease, they're not identified as Alexander disease, usually for often for many years later. And so there's a question is for the parent, then what is the age of onset? And they often, I think, take it as once is when the definitive, more or less definitive diagnosis is made. But then you can look back in their clinical history, and there were other things that maybe you know could have been interpreted as that five, 10, 15, 20 years earlier. But once the parent's been diagnosed and you look at the child, once they show any of these, you'll say that's a symptom. So I think there's a bias there and just how you interpret the data. Now, in in their particular case, that's not an issue because the the offspring, the daughter was diagnosed before the parent. What we don't know in these others is they do not give the information as to who was diagnosed first, but probably because they don't comment on it, maybe it was the parent. So you have that then this bias that um it's going to be earlier for the offspring simply because people are more are going to categorize what they see. First of all, they'll be looking more carefully. And first, when that when they do find something, they'll say that's the age of onset for Alexander disease. So I think there's that issue which which isn't addressed. And then I think there's another problem in that just the nature of doing research and reporting your data is you you tend to report it when you find it rather than waiting five or 10 or 15 or 20 years to see what's going to happen. So you report it when you get the initial results. So the initial results are you've got a parent and you've got younger kids, and you're not waiting to see whether other kids might develop symptoms in five or 10 or 15 years. So the data you'd want there is um, are there other kids who are known to have the mutation who are asymptomatic? And so that could really change um the nature of your conclusions. And um, for ethical reasons, I guess often kids are not not genotyped, they're not they're not determined whether they have the mutation or not if they're not sick.
SPEAKER_01Usually.
SPEAKER_00Yeah, so that data may simply be unavailable. Um, but I think it's probably some cases it is available, and it's in here it's it's um, you know, we don't know how much to what extent that is.
SPEAKER_01Yeah.
SPEAKER_00So those are two reservations about whether this is a real phenomenon or not.
SPEAKER_01Yeah, I think there are several types of potential ascertainment bias. One, as you say, um the parent is diagnosed first, and then so you're looking more carefully at the children and you pick something up at an earlier age. I think it also goes in the other direction that the child is diagnosed first. And then you want to know, just just in terms of the standard diagnostics, you want to know whether they it's a de novo mutation or whether they inherited from one of the parents. So then you start looking more closely at the parents, and then you pick up parent, parental, uh one of the parents being positive for the variant, and then you look at their phenotype more closely. And so uh so it's that there is a built-in bias there where the parent is obviously older than the child. And what we don't have, with one exception being um our 2012 paper, we don't have any really multi-generational family. So I guess one thing I'd really like to wait and see is when whether any of these children who now have a worse phenotype when they have children, if they're able to, um, what happens in the next generation down the line? So the common way of thinking about anticipation is that it gets progressively worse with generations. But there are situations um where, such as one that I know of, uh specifically is in the fragile X field, where you get where the anticipation is due to expansion of a tri-nucleotide repeat, where you not only get expansions, but you get contractions. And these these repeats are unstable, basically. Um, so so there's a there's a big issue about whether this particular individual is so severely affected that they cannot have children in themselves, or whether they're not so severely affected. And there are many on this list who um clearly have uh are listed as having an adult onset at an age where they could have had could have had children.
SPEAKER_00Now, do we need to explain what a trinuclear repeat is?
SPEAKER_01Um can you do that briefly?
SPEAKER_00Well, I think so so people know that DNA has four different nucleotides, uh A C T G, and a trinucleotide repeat is when three of these are just repeated one after the other. So, like AUG, A T G, A T G A T G A T G A T G many, many times. It might be 20 times, 30 times. And when this happens, um when the DNA is being replicated, there's chances for it to sort of slip the replication machinery and fall back three or six or nine of these, and thereby increase the number, total number in the DNA. Or it could jump forward and skip over some. So these trinucleotide repeats are unstable in terms of the number that they have in a row. So that's basically what goes on.
SPEAKER_01So what they end up focusing on, though, is they have several possible explanations for anticipation, if it exists, which we haven't agreed that it exists. But if it exists, they have several um hypotheses, and they focus on one particular one, mosaicism, that the parent is mosaic, um, but the child is not. And so therefore, the child has a higher burden of affected cells.
SPEAKER_00Uh and say what mosaic means?
SPEAKER_01Well, mosaic means that you are you have a mixture of cells in your body of different genotypes.
SPEAKER_00So some of the cells in the embryo and then in the person are going to be normal, and some of them are going to have this mutation. So the mutation occurred after conception. Um, and this is considered unusual. In most cases, it's believed that the mutation occurs in one of the germ lines of the parents, Yeager, the sperm, so that every cell is affected. Um but in the mosaic system, mosaic case, it means the mutation occurred after conception.
SPEAKER_01Right. But I think the um their idea is that it's the parent who's mosaic. And there are clearly examples where you are mosaic for both somatic cells and germ cells. So that's why the parents would be mildly affected, but the child then having it received the variant through the sperm or ova, uh, then is much more uniformly affected and severely.
SPEAKER_00Right. Yeah. So the parent is the mosaic and the mutation. So the the idea is that the mutation occurred in the parent after the parent was conceived. Yes.
SPEAKER_01Right. So they cite the only example in the literature that exists for mosaicism, which is um in Alexander disease. In Alexander disease, which again is one of our papers. Was that also 2012? Yes. I forget. We were we were busy that year. Um so so what do you think about the is there any evidence for mosaicism? Beyond it being a hypothesis, is there anything here that really supports mosaicism?
SPEAKER_00So in our paper, we we proved it because we showed that the mutation could be detected in buckle DNA, which is inside the mouth, the skin, which is um in terms of development germinal layers, it's it's ectoderm, and brain is made from the ectodermal layer. So we could detect it in buckle DNA. We could not detect it at all in blood. So it's clearly a mosaic case. In this paper, there's no evidence whatsoever. Um, and I would say, I would argue that there's maybe evidence against it, because for um it to the mutation has to be present in the brain, we think, to cause Alexander disease, which is ectodermal tissue. But then for the parent to pass it on to their kids, it has to be in their germ cells, which are not ectodermal, they're a different um source. And so if it is mosaic, it's extremely widespread. So that's one thing. The other, another reason to be skeptical is let's say, even if they are mosaic, for the methods that are used to detect that a person has a mutation, and this is use a PCR, the polymerase chain reaction, it would have to be present in probably at least half the cells in the tissue that they take to test. So it's already probably at least a majority of the cells, so I'm a little skeptical that going from that to 100% would make that much difference. You know, maybe it could theoretically, but it's I think a little improbable. Um, so that's another reason to be a little skeptical of this. Um another reason is in four of the cases that they have in their table, they note that the parent had siblings who also had Alexander disease. So that means the parents couldn't have been mosaic. Um, and yet in those cases, they're claiming that the offspring had an earlier age of onset. Okay, but it's the but it's impossible that that could be explained by mosaicism. And you know, unfortunately, they don't mention that in their paper, but I think that's um shows at least for those cases it it can't be the explanation. And then in terms of the data they present, it would have been very helpful to know what was the source of the tissue they used for was that was used, not that they used, that was used in these cases to diagnose Alexander disease. Um it's a buccal swab, which would be then allow you, I think, to hold the mosaicism possibility a little bit more strongly. But often it's blood, which comes from another germinal layer that comes from uh mesodermal origin. And again, we in our case we found no um mutation in blood. But if they're finding the mutation in blood, to me that would strongly indicate that this is not a mosaic person. And and That data just was not presented.
SPEAKER_01Yeah.
SPEAKER_00So the bottom line is for multiple reasons, I think this is um extraordinarily unlikely possibility.
SPEAKER_01Right. So ultimately, uh it seems that mosaicism is a hypothesis. Um, and it's an interesting hypothesis, but none of the data here really um supports it. You know, I would also say that for many of these pairs, the differences in age of onset, it isn't really that great. And uh and one idea I don't think they even offered uh as a hypothesis is the impact of genetic modifiers. So uh, you know, I think there's there's uh there's certainly it's nice to have this tabulation, even though there are some errors in it. Um it's it's uh I'm glad people are interested in this this problem and trying to pursue it. Um now I think we're really going to wait for more hard data uh to to be able to evaluate it.
SPEAKER_00Agreed.
SPEAKER_01All right. All right, so Mike, do you want to say anything more about this or should I move on?
SPEAKER_00I think you can move on. All right, so um well, I let me just one more thing, just you know, technically in the table. Um they include a mutation, uh well, a a change which it's not clear that it causes Alexander disease, um the E223Q. They have another one that I have no idea what they're talking about, whereas M45 Roman numeral II. I don't know if that's a misprint or what, but I have no idea what that is.
SPEAKER_01Oh, I think it's supposed to be M451i.
SPEAKER_00Okay, that that was a possible guess, but I think there, I don't think there's an MR51 in your table, and I think that actually may be longer than the um protein.
SPEAKER_01Yeah, I'd have to go back and check that that reference. Right.
SPEAKER_00You might check that. I I didn't check the reference. All right. So the problem, so it just indicates, I mean, there's some you might say a little sloppy with the table. Yeah, it's not not rigorously done. And and again, for the for the audience, whenever people whenever we see something like that, it makes us worry about the science. Um, when it when the attention is not paid to the these kinds of details. But that's just an aside. All right, that's what I want to cut out.
SPEAKER_01Now you're gonna challenge my editing capabilities. We'll see. So that's all for today's episode of Alexander Disease Research Update. Um, you can send questions to my email address, amessing at whisk.edu, and we'll try our best to address them in a future podcast. I'll announce posting of the next episode through my regular email distribution list. And write to me if you aren't on that list and wannabe, um, as well as on the Facebook page. Thank you for listening, and thank you to Michael Brenner for joining in today. Our theme song was written by Charlie Allenson, special technical assistance from my daughters, Zoe and Rebecca. And of course, thank you to our donors, the Baron Riddle family and the Wanma Fund. I'm Alby Messing. See you next time.
