Scientist Stories: Hans Waterham, The value of phenotypic assays in peroxisomal disorders

Okay, thank you very much for the introduction and welcome to this session. So the session was Proximal Disorders Still Need for a Metabolic Assay, and my part was to give an insight into the value of functional assays in proximal disorders and whether they have any value at all. But I think already after the first presentation it's clear they have. But still I'd like to give my presentation. So there will be some overlap with the presentation that Nancy gave, but it's— actually it's not so much. You better hear things twice than only once and forget about it, especially in educational sections, I think. So proxisomes can be found in all eukaryotic cells ranging from yeast— OK, this is the point— from yeast, worms, plants, to mammalians including humans. And they're common in all eukaryotic cells except for erythrocytes. And actually most of the information about the functioning and protein import in proxisomes have been coming from studies in yeast. Okay. Okay, and proxysomes were first described in 1954 by a Swedish PhD student, Roden, who described looking through the microscope so-called microbodies, which are protein-dense particles surrounded by a single membrane. And at that point, he didn't know what the function of these organelles were. And actually, the first insight in that came in 1960 by Christian de Duve, got a Nobel Prize for that, who did some biochemical characterization of subcellular organelles, including peroxisomes. And he found that these peroxisomes contained one or more FAD-containing hydrogen peroxide-producing oxidases and the enzyme catalase, which decomposes these hydrogen peroxide into water and oxygen. And actually, that's the reason why these organelles were called peroxisomes thereafter. And in 1976, there was the first indication of a complete metabolic pathway, which is the fatty acid beta-oxidation pathway, which was localized in red liver peroxisomes and later also identified in humans. But at that time, the organelles were still not considered as very important for health and for cell survival or organism survival. And it changed after the appreciation of the underlying defect in Zellweger syndrome. And the Zellweger syndrome was first reported in 1964, and at this moment we know of at least or more than 500 patients reported with this defect. And as Nancy already mentioned, the patients with this severe presentation of the Zellweger spectrum show a great number of abnormalities and multisystem defects. And usually these patients die in the first year of life. And actually in 1973, Sidney Goldfisher published in Science some interesting observations. He noted that in cells from patients from the Zellweger syndrome, There were some clear mitochondrial abnormalities, and we already pointed to that just a minute ago. And the mitochondria looked swollen, distorted, and there were some defective complexes, electron transport complexes in these mitochondria. What they also noted, but they didn't think that was too important at that time, is there were some proximal abnormalities of— actually, they didn't— they could not find any proxysomes in hepatocytes and renal tubular cells of these patients. But then they still thought there must be something wrong with the mitochondria. But then only in 1982, people in Baltimore where also Nancy worked previously identified a very interesting defect, and that's namely that the fibroblasts of patients with the cell vega syndrome were not able to degrade very long-chain fatty acids. And actually this was already pointed at by studies in plasma shown here at the top, because if you study the plasma of these patients, you see accumulation levels of very long-chain fatty acids. And normal levels in control cells. But if you then look in the cell Vega cells, you see that C26 is not degraded by these cells here, where it is degraded by controls. And then the C16, which is actually a substrate for mitochondrial beta-oxidation, is degraded. So this pointed to different beta-oxidation pathways in the mitochondria and in the peroxisomes. Actually, the mitochondria can handle long-chain fatty acids, medium-chain, and short-chain fatty acids, whereas the peroxisomal beta-oxidation handles other fatty acids, including the very long-chain fatty acids, but also branched-chain fatty acids like pristanic acid, bile acid intermediates, and the dicarboxylic acids. And after several cycles of beta-oxidation, the substrates from the peroxisomes will be transported to the mitochondria for further completion of the beta-oxidation of these substrates. If you don't have peroxisomes, this chain shortening of these these very long-chain fatty acids will no longer occur. And then they become substrate to the ER fatty acid elongation pathway. And then the C22 fatty acids will become very long-chain fatty acids up to C40 very long-chain fatty acids and so on. And that's toxic and found in these patients. So at this moment we know more than 50 different enzymes that are located in the peroxisomes of humans. And most of these enzymes are involved in different metabolic pathways, and the most common pathways are the beta and alpha oxidation of fatty acids, the etherphospholipid or the plasmalogen biosynthesis pathway, and the bile acid synthesis, which were already mentioned by Nancy. So defects in any of the enzymes or proteins involved in these different pathways can lead to different groups of proximal disorders. Generally, they are subdivided into two main classes. You have the single enzyme deficiencies, in which the activity of the function of only one proxosomal protein is defective. And it can be a proxosomal enzyme, but it can also be a proximal membrane protein, for instance, involved in the import of substrates. And usually, typically in these enzymes, only one proxosomal pathway is affected. The second group of disorders, which has gained more interest, at least in research, is the proxosomal biogenesis disorders, in which the assembly, the maintenance, and the function of the entire proxosome is affected. And in this case, obviously, most multiple proximal pathways are affected. About the genetics of proximal disorders, it's important to realize that nearly all proximal disorders that we know now today are autosomal recessively inherited, and they're— in most cases, the mutations are found back in the parents of these patients. There's one X-linked genetically derived disorder, which is the X-linked adrenoleukodystrophy, and there the inheritance is from one of the parents, and not always from the mother. Sometimes also the father has the mutation, but does not have clear clinical symptoms. There are some rare exceptions that we encountered in the laboratory. There are some de novo occurrences of mutations. That's, of course, more common for XALD, but we also have observed it in the autosomal recessive disorders. We have a few cases of uniparental disomy. We have a few cases that we're working on at this moment of dominant-recessive mutations causing a Zellweger spectrum disorder. And there are some cases of skewed X chromosome inactivation in females presenting with severe XLDL-A presentations. So far we know 13 different single enzyme deficiencies, and most of these deficiencies have distinct symptoms, but some of them have overlapping symptoms, and that was already mentioned also by Nancy. Some of the— especially in the fatty acid oxidation disorders, some of the defects have Sjögren's spectrum-like presentations, whereas other defects in the etherphospholipid biosynthesis but also have a more RCDP-like presentation. It is expected that we can, in the next coming years, can find additional single enzyme deficiencies, especially with the next-generation sequencing technology. The proxysome biogenesis disorders have a defect in the formation of proxysomal membranes or in the import of proxysomal membrane and matrix proteins in the growth in the division or the proliferation, and that's a large— that's several processes involved to make a functional proxisome. And generally, disorders of proxisome biogenesis are now subdivided into the cell bigger spectrum disorders, RCDP, and there's a new group added to this biogenesis disorders is proxisomal deficient defects. So proxisomal disorders can be readily diagnosed in plasma as well as in cells.— because of the metabolic aberrations that can be measured in these different compartments. If you find a lot of different metabolic aberrations, that usually points to a biogenesis defect. If you only find one specific aberration, that can point to a single enzyme deficiency. Then we have several other functional tests that usually are performed in fibroblasts from these patients, and these include enzyme assays, immunofluorescence microscopy, for matrix as well as membrane proteins, metabolite analysis that can be also performed in the cells, immunoblot analysis to look for expression of proteins and for the processing of proteins, because some proxisomes— proxisomes of proteins are processed after importing proxisomes. We can also do flux measurements, whereas we can measure the activity of the entire pathway, for instance, the alpha and the beta oxidation pathway. And, of course, you can do complementation studies, including genetic complementation studies, and I will mention a few of them. So the cell-baker spectrum of Proxysomal Biogenesis Disorders is really a large spectrum of phenotypic variants with overlapping clinical signs, and typically all the patients have— at least most of the patients have sensorineural hearing deficits and ocular abnormalities, and then they have additional clinical symptoms that make the disease more severe and also often is a reflection of the severity of the peroxisome biogenesis defect in these patients. And in the Selleweger spectrum, most of the genetic defect of the patients with a PBD are caused by mutations in any of 13 different PEX genes, and these PEX genes encode proteins involved in import of matrix proteins as well as in the import of membrane proteins into the peroxisomal membrane, and Nancy alluded to that already. So how do we identify the mutation— how did we identify so far the mutations in these patients and the genetic defects? For this, we set up a genetic complementation assay, and this involves co-transfection of sequential co-transfection of one of the PEX gene cDNAs together with fluorescent peroxisomal reporter into patient cells. And then 2-4 days after transection, we look under the microscope for the localization of the peroxisomal reporter protein. If it's in a punctated pattern, then we know that the peroxisomes have reappeared so that we have genetic complementation. And thus we know which gene is defective in these patients, which can be sequenced then. If we don't see functional genetic complementation, the fluorescent protein will end up in the cytosol, no peroxisomes are made. And so we have to continue with the next excedinase. This has been very successful for identifying the genetic defect in cell Vega patients. However, it's very important that these cells don't have proxisomes, because if they don't have proxisomes, then this assay is a little bit difficult to perform. And it also means that it is working very well for the more severe patients and intermediate patients, but not so much for patients with a milder presentation. And we have encountered several patients now in the last 5, 10 years that have a more milder presentation. And here are some cells, and I don't know if you can dim the light more, but it's very difficult to see, I think. But what you see here is in the middle is the normal pattern in control cells where you have a lot of peroxisomes in the middle. But then we encountered patients that have a very strange peroxisomal phenotype with either enlarged peroxisomes, sometimes localized at the border of the cells. Sometimes we have this peroxisomal mosaicism, where some cells do import proximal matrix proteins, other cells do not, and some cells do both. And we sometimes have a lot of cells that have normal proxysomes. And so how did we find the defects in these cells? We first decided to exclude mutations in the PEX genes that we knew at that time. And we did that at the time in the sequence order based on the frequency that we have found patients with these defects. But nowadays we have a PEX gene panel available for that so that we can immediately sequence all the PEX genes. And if we don't find mutations in one of these PEX genes, then of course we come up with a candidate gene approach for what other gene may be affected in these patients. And if that doesn't work out, then we continue with exome sequencing. And then we had a surprising observation that in all these different cell types that we thought may have a a different gene affected, we found the common genes that already had been found previously in patients with severe presentations. And actually, especially in the PEX3 and PEX16 gene defects previously were very severe, and now we found all these milder phenotypes with hypomorphic mutations in these specific— in these particular genes. There were several cells, however, that had not mutations in one of the known PEX genes, and the elucidation of the genetic defect underlying these patients led to this new subgroup of proxosomal fission defects. So the first case was a severely affected dysmorphic baby girl who died 1 month after birth without any improvement. And what we found in the laboratory was persistently raised lactate concentrations and occasionally mild raised very long-chain fatty acid levels. So that indicated that these patients had a problem with the mitochondria as well as with the peroxisomes. However, if we then perform all these functional assays in fibroblasts, we did not reveal any abnormal mitochondrial or peroxisomal biochemical parameters. But then when we looked through the microscope, we observed that the peroxisomal phenotype in these patients was very distinct compared to the control cells. We had this very elongated structures, which is magnified here in the corner. And then we also looked at the mitochondria under the microscope, and there again, we also saw that the mitochondria had a very tubular structure instead of this sausage-shaped mitochondria usually observed in control cells. So this indicated to us that there was a general defect, at least a defect shared by peroxisomes and mitochondria, in the fission of these organelles. And there were a couple of candidate genes at that time known, including the DLP1 gene. So this was the gene that we sequenced first. And indeed, what we found, we found one heterozygous dominant mutation in the DLP1 gene. And the interesting thing is that if we overexpressed the mutant DLP1 in the control cell line, we could convert control cell line into this typical mutant cell line with these elongated mitochondrial structures. Whereas if we overexpressed control DLP1 in the patient cell line, we could functionally complement the patient cells into a normal phenotype with respect to the mitochondria. So that indicated that this mutation, which was de novo, and the parents are biologically the parents of this patient, that this mutation had a dominant negative consequence. The second group of patients that we identified soon after that, and we already know at this moment 10 patients, have a have a defect in the PEX11 beta. And although it's called PEX11, it's not a protein involved in the protein import or the assembly of the peroxisomes, but it has been shown in yeast and in plants that it's probably involved in the fission of peroxisomes. And what we saw here in all these patients is that all the peroxisomal parameters, as well in the plasma as in the fibroblasts, were completely normal. Only on a very few occasions there was some elevated or increased very long-chain fatty acids in the plasma of patients. And what's also striking is all the patients had biallelic loss-of-function mutations in the PEX11B gene. So the PEX11B protein is not synthesized. We have not found patients yet with hypomorphic mutations. So we complemented cells with these— of patients with this defect.. And what we saw is that at the— and this is the control at the wild type, we see the normal peroxisomal staining. If you look at the patient at 37 degrees, we see this elongated peroxisomal structure indicating a fission defect. And then we found out that at 40 degrees, if we put the cells at 40 degrees, they completely lost the protein import of the peroxisomes, so there are no peroxisomes there. So what we did, we used this condition here to try to complement the cells, and then we see that we got the same phenotype again as in the wild type, confirming that PEX11B was indeed the problem in these cells. So at this moment, we have this PEX11B deficiency, which only affects peroxisomal fission, by the way. And then we have the DLP1 deficiency, which affects both peroxisomal and mitochondrial fission. And we know of 4 patients now, there's only one reported. Then we have 2 additional genes that are in the identified at the meantime is MFF and GDAP1, which also both affect peroxisomal and mitochondrial fission. So where do these genes or these proteins play a role? Well, we think that PEX11B and the MFF, GDAP, and DLP1 play a role in the maintenance of peroxisomes. And the maintenance is defined as peroxisomes that divide and grow from each other, and then there's this de novo pathway where proxisomes are derived from the ER. And we have another component here, PEX11 gamma, which we think plays a role in this part of the pathway, but I'm not going into detail in this part. But we think that the proteins that we have identified so far in the patients play a role in this part of the pathway. So combined, so we have now seen that we can have different mutations in different genes that can give rise to different subtypes of the biogenesis disorders. We can have a problem in the membrane formation, but that only occurs if we have loss of function, two loss of— a biallelic loss of function mutations in PEX3, 16, and 19. Because if we have hypomorphic mutations in these genes and in any of the other PEX genes, then we can end up with a lot of clinical phenotypes, and we can have several different peroxisomal phenotypes in cells. And then we have these deficient subtypes where we have identified several gene defects now that give rise to aggregated peroxisomes. We have not found any genes yet involved in the degradation of peroxisomes, but I expect that this will be found soon. Okay, most of the examples I've given so far, I don't know how much I Okay, thanks. Most of the examples that I present so far were obtained, or the results were obtained in the traditional approach, so in the pre-next-generation sequence area. So the traditional approach of diagnosing inborn errors of metabolism, including proximal disorders, is first you start with clinical symptoms, then you do metabolite studies, usually in plasma, and then you follow it up with functional studies in cells, and then you end up with the genetic studies. That is finding the genetic defect. So of course, with the implementation of the NGS and with the strong push towards genetic testing, this has changed a bit. So what we often observe now in our laboratory is that you have the clinical symptoms and then people send in DNA, so not plasma, not fibroblasts, immediately for targeted genetic-first approach. For the gene affected they think may be interesting. Obviously for Stellweger syndrome patients that's not so easy, but we have the gene panel now, the peroxisomal gene panel that we can use for that. And then afterwards you still have to imply metabolite studies or functional studies in cells if you find mutations that have not been found before to settle the diagnosis. And of course the future, and actually the future is already here, is more generic genetic-first approach. And that's that you have clinical symptoms, you provide whole exome sequencing or whole genome sequencing, and then you contact a laboratory like our laboratory that you have a finding in one of these genes that have something to do with peroxisomes, and then we can— and they ask us to find out if this has something to do with the clinical presentation of these patients. So then we ask for material for metabolite studies and functional studies in cells. And this was already indicated several years ago by Aroel Hennecombe and Leslie Biesecker also, that the next-generation sequencing is really demanding a next-generation phenotyping. So with respect to the main question of this session, yes, we really do need functional assays to confirm the findings of next-generation sequencing. Actually, next-generation sequencing is only a start, and after that, the real work really starts. And that's not so much only for proximal disorder. That is really true for all proximal disorders and maybe for all genetic disorders in general. Okay, a few examples. Last year, we identified a novel, very mild proximal biogenesis defect, the Heimler syndrome, with colleagues from Belgium, Morocco, and England. And actually, the first patient with Heimler syndrome was— patients were published in 1991, and these were CIPs with with hearing loss and a mild hypoplasia, so teeth problems, and they had some nail abnormalities. Otherwise, they were completely normal. They had no development delay, no intellectual disability. And since then, 5 additional patients were identified, or families were reported with similar symptoms, and occasionally these patients may have retinal pigmentation. So actually, the Belgian group performed WES, whole exome sequencing, in one of the patients, and then called me with a finding that they found a missense mutation at the end of the PEX1 gene, and they wondered whether that would be significant at all because of these very mild clinical symptoms. And then we said, well, send over cells because this could very well be important. And then we found out in the cells that indeed if you grow the cells and you look through the microscope, that you again see this mosaic pattern where some cells do import catalase,— and made exposed when all the cells do not import catalase. However, if you look in the plasma or in the fibroblasts and you look at the biochemical parameters, all these biochemical parameters are completely normal. So again, we needed a microscope to find out that this is indeed something important. But then if you culture the cells at 40 degrees, they completely lose their peroxisomal protein import. And what's more important than at 40 degrees, we could genetically complement these cells with PEX1 or with PEX6, because in the meantime, based on this data, we contacted a group in England and asked— because we knew that they had also whole exome sequenced their patients— and we asked them to look specifically for PEX1 or the related PEX6 protein genes to see if they had mutations in there. Then it turned out that the coverage of these genes was not so well in this whole exome sequencing, so they repeated it. Manually repeated the sequencing, and they indeed also found mutations, biallelic mutations in these genes. So we ended up with 6 families with this Heimlich syndrome. So we tested the different genetic variants by expressing them in cells that completely lack either the PEX1 or the PEX6 protein, and then you see that at least one of the mutations was hypomorphic and gave rise to some peroxisome-positive cells. So apparently Heimer syndrome is due to biallelic mutation in PEX1 and PEX6, and all these patients have at least one hypomorphic PEX1 or PEX6 allele. And this really puts Heimer syndrome and also Usher syndrome, which is a similar presentation, at the very mildest end of this cell degeneration spectrum disorder. Another case which is interesting given the discussion we had after the former talk is the talk, is a boy with a completely normal psychomotor development, but then at 5 years he presented with ataxia, dysarthria, and learning disability, and some MRI problems. The sister died at age 9 with similar symptoms but had not been diagnosed. And then the people performed whole exome sequencing, and they found a homozygous missense mutation in PEX13, and so they contacted us whether this would be relevant and maybe be a proxosomal biogenesis disorder. So we did all the biochemical testing, we looked in plasma, we looked in fibroblasts, all the biochemical test results were normal again. But then again, if you look through a microscope and we look for the protein import, proximal protein import, you see this mosaic pattern. And if you then put the cells at 40 degrees, they lose their proximal protein import, confirming that indeed this is the mutation. And if If at 40 degrees you bring back the PEX13 cDNA, you will refer the cells to the normal wild-type phenotype, confirming that these mutations are relevant. So all these examples show that the genetic-first approach may provide a genetic diagnosis or a tentative diagnosis, but often do not or does not precisely predict the biochemical or the clinical consequences of a genetic defect, and that is why you need a additional studies to determine what the genetic variant, genetic defect may be. And this is— these studies are needed to determine the pathophysiological consequences. What does the mutation do in the cell? What does it do for the organism? But it's also important to come up with the short and the long-term clinical consequences for patients. And of course, in relation with that, it's also, of course, very important to follow the natural history of these patients so that later we can tell something about the specific effect of these mutations. So I already mentioned this group of biogenesis disorders where specific mutations in all the different specific genes can give rise to different clinical presentation or spectrum presentation even within the peroxisome biogenesis disorders. But another interesting thing is that we are very good in diagnosing metabolic disorders, and we've been so for 30, 40 years. And that means that we pick up metabolic patients very early in life, and because we can treat them well, or at least take care of them well, they grow older. And what we see now is also in our hospital, we have a group of about 20 Zellweger spectrum patients that have reached ages between 10 and 50 years old. And what we have observed is that in some of these patients, they all have the same most of them have the same hypomorphic mutation in the PEX1 gene. What you observe is a clinical differentiation between these patients. And we also see many patients that have progressive clinical manifestations, but then if you look at the biochemical parameters later in life, we see that they completely normalize. And here's an indication of pipicolic acid and the bile acid intermediates you see at higher age of this patient here. We see that they basically normalize, and also if we group the patients and we look for these parameters, Paxol metabolite parameters in patients throughout the years, we see that there's a normalization, but there's not a normalization in the clinical presentation. So it's very difficult to predict in the first year of life on the basis of the mutation what's going to happen with these patients. And also there's another— the main disorder, Paxol disorder, is X-linked adrenal leukodystrophy. And it's known there are several families that have exactly the same mutation, and they have a complete— may have a complete different clinical presentation ranging from asymptomatic to a very severe lethal CCALD cerebral involvement presentation. And the basis for this phenotypic variation is, at this moment, still unclear, and we are also doing studies on that. There is no— so apparently there's no genotype and phenotype correlation. Correlation, but you can also wonder whether this was the case in the Kellweger spectrum patients. So you may wonder whether there are environmental factors involved and whether there are genetic modifiers. And to identify these genetic modifiers, NGS may be of great help also. So in general, I think we can say for monogenetic diseases or simple Mendelian diseases, it's clear that they may often present as complex or polygenic diseases, and that can be due to environmental factors or the effect of additional DNA variations somewhere else in the genome. So genetic or environmental modifiers, but it could also be caused by the fact that the primary gene defects may cause or promote complex secondary defects, and that's actually also a bit what we discussed after the last talk. So we looked into that a bit because we have been working on cell-like spectrum disorder patients So one of the striking abnormalities is there, the very long-chain fatty acids that are accumulating in these patients and are not observed in control cells. Well, and very long-chain fatty acids are part of phospholipids. So phospholipids have a polar head group. It can be choline or serine. Then they have this glycerol backbone, and then they have two fatty acids connected to the glycerol backbone, and then they form this phospholipid bilayer with the fatty acids here in the middle. And you can imagine that if you have an accumulation of very long-chain fatty acids in Zellweger syndrome, that the tails, the very long fatty acid tails of these phospholipids have been changed. So we looked into that using lipidomics, which is a sort of metabolomics that we have set up in our laboratory, and which actually is a large-scale study of cellular lipids in biological systems. So the setup of the experiment is as follows. So we use cultured skin fibroblasts from healthy controls, from PEX1 patients with a severe mutation, a lethal mutation, and from PEX1 patients with these mild hypomorphic mutations that we also have a lot of patients from. So we analyzed these within the lipidomics platform, and then we looked for, in this case, for the phosphoglycerine species. And what you now observe is that— so from the left to the right, you see an increase in the length of these very long chain— of the fatty acids that are built into these phospholipids. And then you see that in control, mostly these are long-chain fatty acids, whereas in the PEX1 patients, you see an increase to the larger fatty acid chains., which makes sense because these are accumulating in these patients. So the phospholipid composition of these cells is completely changing due to the accumulation of the metabolites. And you may wonder what it does to the membranes and what it does to the proteins that are localized in these membranes and functioning in there. So you can also see that here that the shorter forms of phosphogolin are higher in the control cells and the longer forms of the phosphogolines are higher in the PECS1, the severe PECS1 patients. So this is for the PECS1, but it's also, of course, important to look at the milder patients. And also in the milder patients, you see a similar phenomenon. It's less clear, but you can clearly see from the coloring that also in the milder patients, this is happening. What you can now use is— what you can do now is you can also use different species of these phosphogolines. Colines and use it and make ratios of it and see whether they can perform as good biomarkers to determine the severity of the disease in these patients. And you can actually see if you pick out a nice phosphogoline species that you can really make a distinguished— distinguish between mild praxisomal disorders and severe praxisomal disorders. So this can also be a way a good biomarker for the future. Okay, to conclude my talk, so praxisomal disorders, is there still a need for metabolic assays? Well, I think it is. I think, I'm sure it is. I think it's very important, also because the next-generation sequencing for praxisomal disorders, but actually that's in general for all the inborn errors of metabolism, may provide a genetic diagnosis but does not often precisely predict biochemical and/or clinical consequences. And therefore, I think it requires the input of multidisciplinary teams for interpretation and validation and functional assessment of the next-generation sequencing. I think it's very important to involve also metabolic physicians, clinical geneticists, clinical biochemists, molecular biologists, and protein biochemists in the annotation of these data. And of course, it requires the development and implementation of new and improved techniques for the functional analysis of NGS findings. And especially with regard to— for metabolic diseases, the metabolomics platforms are— will be very important. And what I really find very interesting is the CRISPR-Cas9 technology, which allows us now to introduce mutations in specific cell lines without acquiring cells from these patients itself. And especially in patients that have died and there is no availability to get cells from these patients anymore, this really gives a nice possibility to study the effect of these mutations on a cell basis. So at the end, the group that— my group are working on the Proxthon biogenesis disorders, and the people that are doing the diagnostics also in the laboratory, Sasha Fred and Ronald Wonders, and then I'd like to I mentioned people with whom we collaborated on the Ehlers-Danlos syndrome, and I did not mention the RCTP in this case because of time.

And I thank you for your attention. Thank you, Hans. This is— if we had time, we'd do your talk again too. Oh, okay. And I know you have a reputation as being very helpful, in Europe at least. People contact you, and you're interested. Interested in these diagnostic nuts. So you won't get fewer emails after this.

We also like other metabolic disorders, by the way. Okay.

So think about your questions. I wanted to— I said, you know, it's okay to ask stupid questions, so here comes one. When you were talking about mosaicism and I'm sorry we couldn't bring the lights down, but we could see your green things pretty well. You showed some pictures of cells that had a mixed— where it looks like some things were working and others weren't. So this mosaicism, is it the product of revertent mutagenesis or postzygotic mutagenesis or— why are those— what is the basis of— or heterozygote dominant mutations, what is the basis of mosaicism as you have used that word in this talk?

Yeah, that's— that question has been asked all the time after our presentations. And actually we still don't know. What we see is that in many of the hypomorphic— of the Zellweger patients with hypomorphic mutations, you see this phenomenon. And for instance, the patients that are homozygous for this mild PEX1 patient, that it's very common that you see that there. Well, if you have the cells, you grow them at 37 degrees, they have a mixed population of cells with protein import competent and other. But if you then, if you put them at 40 degrees, they lose their import. But if you put them at 30 degrees, they all have normal proxysomes. So it has nothing to do with additional mutations or whatever. So it seems to be a kind of a stochastic chance that it's happening or not. So one of the things we're doing now actually is to introduce this mutation by CRISPR-Cas in HeLa cells, and then we will start with a clonal cell and see if a clonal cell will still not come up with this mosaicism. So that indeed it is not something else which is there, but that it really has to do with this mutation. And we believe it has to do with the folding of the protein. Protein or something like that, but we're still not sure what's happening there.

Okay. So then there was one more question. I'll let you ask questions too. You talked about most of these disorders are Zellweger spectrum disorders except for ALDR are recessive, you said. Yeah. But then you said your exceptions were UPD and de novo mutations,— and you said dominant recessive.

By that do you mean that there are some—

do you want to explain what you mean by dominant recessive?

Well, we have now encountered one patient with a defect— well, actually a couple of years ago— that had a— was heterozygous for one mutation, and the parents do not have the mutation. They are the biological parents, and then this mutation gives results into a truncated protein that apparently screws up the peroxisome biogenesis. So if you overexpress this one, like the DLP1, then we end up with peroxisome biogenesis defect, and we can also complement the cell with that. So that looks like there's a dominant one. And then we have another— we have several families, and actually that's a collaboration with a lot of groups in the world, with a specific PEX1 mutation that also seems to be— is also dominant negative. Recessive mutation. Also one heterozygous mutation in PEX1, and that gives a plexus hormone disorder.

So that's through overexpression or through interaction between the normal and abnormal protein?

Also the interaction probably, yeah. And also in those case, in those families, we have 4 or 5 families, the parents do not have the mutation. And they are the parents, okay. But the complicated is one or two families do have, but they don't have— but it's getting— it will be very interesting story because there is more to it.

A little bit, actually. OK. Now while you've got him, we've got time for several questions. This is Jill Clayton-Smith from Manchester making her way to the microphone.

So this is another stupid question, maybe. It's about the temperature sensitivity.

Do patients react badly to fever, or if you have patients living in hot countries, do they do well? Should they move to Norway? Yeah. I think a lot of metabolic patients have a problem if they get fever or things like that, so it's complicating metabolic— so I look again at Nancy. I got that question a lot also together with the mosaicism. We have another— we are working on other defects that has a similar phenomenon, but that's a fever syndrome, so there you have a problem with the get fever. So, you know, but I don't know, but I can imagine that there is a— so for one of the PEX11 beta patients, we saw there that also if you put them at 40 degrees, the cells, that they lose their proxysomal import, protein import. There is an anecdotal suggestion that indeed he is worse when he got a fever. So, but that's not really studied that well.

But you can imagine that's a problem, yeah. With the PEX11 beta patients, the few that are reported, there are a couple of patients, and we have another, who've had sort of some weird, almost like psychotic episode after an intercurrent illness.

And I wondered whether the behavioral things were a prominent feature in some of the older patients. Well, the oldest one, which was the first one we reported, Dutch patient, I think he's in his 30s now. Well, he's really a Zellweger patient now, or NALD, so he's really progressively deteriorating a lot. But what you see with most of the PEX11 patients early, they have cataracts and they are short in stature, except for the Dutch patient, but then we have not measured the parents because maybe he is short compared to the parents because my parents are like here years, which is typical in the Netherlands.

Nancy, can you go to the microphone, please?

I just wanted to mention that again, we just don't have a whole lot of older patients, so we don't know exactly how they're presenting and what they'll do. But the concern about behavioral deterioration of older patients has come up, but there equally have been older patients who've done well.

Okay. So the normalization of biochemistry in the older patients, what is that? Is that just a problem for clinicians who— Yeah, that's definitely a problem. Or does it mean something? Is it a sign of— that this is a mild— is it a sign that these people are able to compensate in time somehow, or what is that? Just a funny thing that's a problem for clinicians, or does it mean something?

Well, to start, we don't know yet. One of the things I want to do is— so we have skin fibroblasts from these patients when they were very young, and I would like to get skin fibroblasts again from from them now that they are in their 30s and then compare, so do the whole exome of whole genome sequencing and transcriptome analysis and epigenetics to see if there's something genetically changed in there. But one of the things that we know is that peroxisomes become less active when they're in older— Normally. Normally from itself, so maybe the peroxisome metabolic processes are less prominent or so. Less important as you get older. Yeah, maybe, but it's definitely a problem for a diagnosis because it basically means if you suspect a patient from a paroxysmal disorder, you don't find anything in plasma, that does not mean that you have excluded a paroxysmal disorder. And actually we know of patients that have an aberration in plasma but nothing in the fibroblasts. We know of patients that have normal plasma, they have aberration in the fibroblasts.

We know patients that have no aberrations in both. The lipidomic slide you showed where you had correlation between the metabolic abnormalities and the clinical phenotype, was that people who were the same age or what? These are fibroblasts. Yeah, and so the fibroblasts are from people that are the same age?

Well, that's difficult to tell, but in most cases it definitely did seem a few patients there, or they must have been in the first year of life.

Yeah. So that is also something that might change with age.

Um, that's a good point to look up. Yeah, also in combination with the study size.

Yeah, good suggestion.

Yeah.

Do you, do you think that— is there any connection between aging and having good quality peroxisomes, or is— do you think there's—

do they play a role in normal aging? Well, there have some studies. I'm looking to Nancy, maybe she— there have some studies, at least in older cells, that the peroxisomal functioning is—

Nobody can hear you, dear. So, I mean, there is thought that peroxisomes play a role in some complex diseases, and those include neurodegenerative diseases. And so there's articles on low plasminogen levels in Alzheimer's disease and Parkinson's disease, and some proxim deficiencies in brains from patients with Alzheimer's disease over several stages that can correlate to the stage of the disease. But the idea is this is a secondary effect, but may also contribute. So that's another, another unanswered question. It's often good on our grant reports by saying, oh yeah, this is part of rare disease, but it also can be involved in other complex disorders—

parentheses Alzheimer's. Yeah, so, so it's speculation, but theoretically maybe you're— if you're peroxisome— if you have good quality peroxisomes, you're going to get less worse Alzheimer's than you otherwise would have. Who knows?

But it's something for a grant application. Maybe it's just part of the mix of all of those complex factors.