June's ALS Research Roundup

Identifying amyloid-related diseases by mapping mutations in low-complexity protein domains to pathologies

Murray et al., 2022

A key element in the development of ALS is alterations in the shapes of certain disease-driving proteins known as amyloids. These include TDP-43, SOD1 and FUS, with most patients having at least one such amyloid protein take up disease-associated aberrant conformations. These types of aberrant proteins are prone to clumping together into aggregates, particularly into insoluble fibre-like aggregates known as fibrils. One particular structural element which has been linked to fibrillar aggregation is the presence of low complexity domains, segments of protein without the usual range of hydrophobic amino acid residues. However, these kinds of aggregation are not uniquely disease-associated, and there is evidence that certain types of aggregation drive healthy cellular processes. The key difference is that unlike healthy fibrillar aggregates, disease-associated ones are ‘irreversible’, and require extreme intervention to break up once they have formed.

This study investigated the processes underlying this irreversibility, particularly in regards to a type of structural segment in low complexity domains referred to as LARKS (low-complexity amyloid-like kinked segments). Several amyloid proteins enriched in LARKS have been established to reversibly aggregate in healthy conditions but irreversibly aggregate when modified by ALS-associated processes, and are the focus of the paper. These include TDP-43 and FUS, two of the most common ALS-associated amyloids. Previous research has shown that LARKS form ‘steric zippers’ by mating the β-sheet structures in amyloid proteins, resulting in fibrillar structures. A key element of irreversible fibrillation is believed to be the interlocking of side chains protruding from these sheets. It has been theorised that ‘kinks’ in the LARKS can prevent this, possibly explaining why some do not form these irreversible aggregates.?

The study involved computational screening to identify mutations in LARKS that may drive the transition from reversible to irreversible aggregation, which were further investigated through structural and biochemical methods. This screening was performed on the intermediate filament protein keratin-8 (KRT8), which plays a role in the organisation and assembly of and signalling by simple epithelial tissues. These are typically sheet-like surface tissues including the skin, surfaces of the eyes and the surfaces of tracts including the digestive and respiratory.

KRT8 was found to form amyloid aggregates, and this aggregation was promoted by a group of mutations identified in this study. Each of the mutations which were predicted to promote steric-zipper formation were confirmed to accelerate amyloid-like protein aggregation. Notably, a hydrophobic segment in the low-complexity domain of KRT8 was found which had similar structure to the ‘fibril core’ of other amyloids. This may represent a site from which further fibrillar aggregates can grow from free-floating monomers of the constituent protein. There was also some indication that ethanols can contribute to amyloid growth, with pro-aggregation effects in KRT8 as well as other established amyloids α-synuclein, which play roles in Parkinson’s disease. This was theorised to be due to a high concentration of polar residues in the ‘fibril core’-like region, as another amyloid, Tau, which lacks such a region did not experience the same ethanol-enhanced aggregation.

While the individual structures of amyloid proteins differ, the presence of some consistent properties has emerged which may serve as future targets for research and therapeutics. In an ideal world these consistent structures would be key to their aggregation, and so provide a universal target for therapeutics which could allow us to address many neurodegenerative disorders at once. However, this is unlikely. Instead, it is important to develop an understanding of how different types of structures interact with amyloid aggregation, with the hopeful goal being a way to prevent this aggregation which is likely to be a key driver of pathology. This paper found evidence that the steric zipper-like structures were linked to the formation of irreversible pathological aggregates as opposed to functional, reversible aggregates. It may be that we can find a was to artificially modify these proteins, or create binding agents that block these steric zippers to prevent pathological aggregation. The ‘fibril core’-type structures may likewise serve as therapeutic targets, possibly playing a role in identifying amyloid-seeding proteins in the future so that they can be culled in favour of their non-amyloidogenic counterparts. The finding that ethanol can play a role may prompt beneficial lifestyle changes around alcohol, or even help with research into intentionally aggregating proteins (the explanation is beyond the scope of this, but it may help slow disease progression in some cases). Ultimately, understanding how aggregation occurs is the first step to stopping it, and ideally, the diseases driven by it.

TAGS: mechanistic, amyloid, TDP-43, FUS, in_silico

In silico design of a TLR4-mediating multiepitope chimeric vaccine against amyotrophic lateral sclerosis via advanced immunoinformatic

Saleki et al., 2022

While the lack of a single driving pathogen makes the idea of a ‘vaccine’ against ALS and similar neurodegenerative diseases initially seem untenable, some researchers have taken that as a challenge. In particular, the field of ‘reverse vaccinology’, also known as ‘in silico vaccinology’, uses computational screening to select targets which are likely to produce robust immune responses, even in the absence of more traditional pathogens. To do this they have developed suites of tools such as C-ImmSim and GROMACS which are capable of digitally simulating the adaptive and innate immune systems. This paper aimed to develop a means of targeting misfolded proteins which are prone to forming aggregates which are likely disease drivers in a subset of ALS case. A further challenge was in doing so without targeting the healthy form of the protein, which plays a key biological role. Several such immunotherapies have been tested in Alzheimer’s disease (targeting the amyloid β protein), although none have yet been able to accomplish the stated goal. One was unable to distinguish healthy from pathological proteins, while the other had significant safety issues.

Several key ALS-associated proteins were selected for use in this process: SOD1, TDP-43 and TRAF6, a protein which specifically interacts with dysfunctional SOD1. Antigens from the surfaces of each of these proteins were used to determine epitopes which bind to these surface antigens using machine learning. These were then incorporated into a digital construct which also stimulated the CD4 T lymphocyte, a key element of the adaptive immune system which coordinates immune responses. A number of candidate ‘vaccine’ structures were created, with the most favourable selected for further testing. Programs were used to predict various properties of this vaccine including any allergenic properties, solubility and its molecular structure. Another was used to simulate the immune response in a biological system, predicting favourable interactions between the ‘vaccine’ and target proteins.

While admittedly this field of research is beyond my area of expertise, the concepts underlying it are fascinating. Being able to induce an immune response to specifically misfolded disease-driving proteins would be a phenomenal leap in treating and preventing ALS. As a likely step in the ‘life cycle’ of these proteins is transmission between cells to ‘seed’ aggregation in other cells, training the immune system to remove these ‘infectious’ proteins as they travel between cells could be a key step in stymieing the spread of pathology. However, this relies heavily on the accuracy of the modelling software used, which appears to be fairly recent. Furthermore, I am uncertain how this digitally modelled vaccine would be translated into the ‘real world’ as a therapeutic tool. The aforementioned attempts in AD suggest that it is possible, but their apparent failure also suggests that success is far from assured. This study’s vaccine also seems particularly ambitious, targeting three distinct disease-associated proteins. It seems almost too good to be true, but while I remain sceptical, I would be overjoyed to see it achieve success if the physical world. ?

TAGS: therapeutic, machine_learning, TDP-43, SOD1, in_silico

Multimodal?in?vivo?staging in amyotrophic lateral sclerosis using artificial intelligence

Behler et al., 2022

One of the most common elements in ALS is the spread of modified TDP-43 proteins, with a broad base of evidence suggesting that aggregation-prone variants are either linked to or drive pathology. TDP-43 aggregates are capable of converting healthy forms of the protein into their disease-associated form, spreading both within and between cells and propagating the spread of the disease. The spread of TDP-43 follows a sequential pattern, typically moving along networks of interfacing neurons. Four stages have been identified in this spread. It begins in the motor neocortex, then progresses to the spinal cord and brainstem, then to the frontal and parietal lobes and finally to the anteromedial temporal lobes of the brain. Notably, the cognitive symptoms which are present in a subset of ALS cases also closely mirrored these stages of progression. Altered executive function was paired with stage 2, behavioural disinhibition with stage 3 and impaired memory with stage 4.

The staging of ALS has previously been studied using diffusion tensor imaging (DTI), a variant on magnetic resonance imaging (MRI) which generates contrast using the diffusion of water molecules. In this study, DTI, cognitive and oculomotor parameters from 245 ALS patients were used in conjunction with machine learning/artificial intelligence to differentiate staging groups. Their aim was to show whether these measures were associated with each other, as well as if the combined body of data could be used to more accurately stratify cases of ALS based on their disease stage.

The study found a clear correlation between cognitive parameters and structural changes in ALS-associated tracts, as measured by DTI. Performance in executive oculomotor tasks also correlated strongly with cognitive test scores. This data was used to by the machine learning system to group patients into clusters A-D, with cluster A having the best performance in executive oculomotor tasks and cognitive tests, cluster D having the worst performance and B-C showing intermediate results. The authors believed that clusters A and F corresponded closely to stage 1 and 4 of ALS progression. Considering that the machine learning clustering was independently performed without prior biases, this serves to further validate the staging system. While B and C were clearly intermediary between A and D and had some features indicative of the stage 1-4 progression, they were more difficult to clearly distinguish from each other in a statistically significant manner.

While the main result of this study is validation of the existing stages of ALS progression, it is valuable that they were independently reproduced by computational analysis and not just artifacts of human judgement. Being able to clearly determine the progression of ALS based on quantifiable features may help us to both identify how far a case has progressed, but also to predict how it will progress in the future. This is likely to contribute to stratification of patients, separating them into more targeted treatment groups which can be treated for the stage of ALS which they have reached. Given the lack of success with most attempted ‘general therapies’, these sorts of ‘personalised therapies’ are likely the next step of ALS treatment. There is a significant difference in treating early-stage ALS, in which pathology is still spreading, compared to late-stage where most motor neurons are already infected, and effective treatment must be tailored to incorporate this. However, this study does not answer all of the relevant questions. They were only able to study propagation patterns within the central nervous system and not map these changes to peripheral symptoms, and the results were not able to be confirmed post-mortem. Confirming how the features analysed here correspond to the more functional symptoms could do a great deal to help us understand the spread of ALS pathology, and how best to respond to it.

TAGS: diagnosis, stratification, machine-learning, TDP-43, human

NUP62 localizes to ALS/FTLD pathological assemblies and contributes to TDP-43 insolubility

Gleixner et al., 2022

ALS cases have a wide variety of genetic factors involved, although most have mutations in at least one protein which misfolds into an aggregation-prone form. These aggregates are believed to be primary disease drivers, both damaging cells directly and sequestering materials needed for healthy function. However, an outlier among these proteins is TDP-43, which is found to aggregate in up to 97% of sporadic ALS patients even in the absence of mutations in the actual TDP-43-producing gene. The reason for this isn’t clear, although theories include that it is prone to being ‘seeded’ by other protein aggregates, or that it is always close to aggregating and the changed cellular environment in ALS tips it over the edge. This study investigated the connection between TDP-43 and the most common known variant of ALS, known as C9-ALS. In this, a repeat expansion in the C9orf72 gene results in the production of a set of aggregating proteins known as dipeptide repeat (DPR) proteins. In particular they studied the mislocalisation of TDP-43, a pathological factor in which the protein moves from the nucleus of neurons into the cytoplasm, the fluid which fills most of the cell. As pathological aggregation only occurs in the cytoplasm and not the nucleus, this is likely to contribute to the development of TDP-43’s role in ALS. Their analysis was performed induced pluripotent stem cells (iPSC) grown from tissues taken from the spinal cords and several brain regions of human C9-ALS patients, as well as transgenic mice.

The major connection found between C9-ALS pathology and TDP-43 mislocalisation was in the nuclear pore glycoprotein p62 (NUP62). This is one of a group of proteins which comprises the nuclear pore complex, a permeability barrier which governs traffic between the nucleus and cytoplasm. NUP62 was found to be both depleted and mislocalised to the cytoplasm in iPSCs from C9-ALS patients. The cause of this appeared to be polyglycine arginine (poly-GR), one of the DPRs produced from pathological C9orf72 repeat expansions. Poly-GR initiated the formation of cytoplasmic RNA clusters which collected both NUP62 and TDP-43, as well as several other less immediately relevant RNAs and proteins. This also resulted in NUP62 and TDP-43 undergoing further interactions which further removed them from a functional soluble state. Notably, previous studies have identified aggregates containing both NUP62 and TDP-43 in ALS models without C9orf72 repeat expansion. One possible explanation is that these two proteins are both prone to binding to certain aberrant proteins, and that poly-GR is merely one example of a protein which can initiate the binding process. It is also worth noting that abnormal nuclear/cytoplasmic transport occurred in ALS models even without NUP62 dysfunction, suggesting that there are also other dysfunctional pathways involved in this process. However, it does appear that the specific interaction with TDP-43 is strongly linked to NUP62 and not other nuclear pore proteins.

The research detailed here helps to explain one mechanism by which TDP-43 may become involved in diverse ALS pathologies, even without direct TDP-43 mutations. Like many pathological processes in ALS it appears to include both gain of function, inducing the creation of aggregates, and loss of function, with dysfunction in the barrier between the nucleus and cytoplasm. While the aggregation-inducing interactions detailed here were the result of C9-ALS factors, it is plausible that similar processes and interactions may exist within other ALS variants, resulting in the ubiquity of TDP-43 aggregation. Understanding their mechanisms is, as always, the first step to understanding how to counter it. If we can prevent the activity of poly-GR or otherwise prevent NUP62 mislocalisation, it may be possible to prevent the development of TDP-43 aggregates. While it is unclear exactly how much of the neurodegeneration in ALS is the direct result of TDP-43 aggregation, there is almost certainly value in mitigating at least a portion of ALS pathology.

TAGS: mechanistic, TDP-43, C9orf72, iPSC, human

Cryo-EM structure of an amyloid fibril formed by full-length human SOD1 reveals its conformational conversion

Wang et al., 2022

A key element of ALS pathology is the misfolding of certain types of proteins into an aggregation-prone form. But more than that, this misfolded form is also ‘infectious’, and can be transmitted from pathological proteins to healthy ones, reshaping them through an unknown mechanism. This had led to these proteins being referred to as ‘prion-like’ or ‘prionoid’, as this ‘protein infection’ mechanism is shared with prion diseases such as bovine spongiform encephalopathy (‘mad cow disease’). In this study, researchers sought to understand how this conversion takes place. Misfolded SOD1 proteins were generated from recombinant human SOD1, and cryogenic electron microscopy (cryo-EM) was used to analyse them. Cryo-EM is a form of 3D molecular imaging which takes place under low temperature conditions, and is typically used to determine the structures of biological molecules.

The SOD1 generates was in two distinct forms, as confirmed by the cryo-EM structural analysis. The first was ?unstable dimers, small proteins consisting two SOD1 molecules. The second form was fibrils, fibre-like pathological aggregates which are the mature, stable end-state of most prionoid proteins. Fibrils likely contribute to the propagation and spread of ALS between cells rather than directly exerting toxic effects, but have been linked to certain pathological activities such as disruption of mitochondria and inducing inflammation.

Previous studies of SOD1 fibril structure using several methods, including protease digestion experiments and analysis using mass spectrometry. These resulted in several structural models, including the ‘three key region’ model and the ‘N-terminal core’ model, each of which were partially compatible with the structure found in this study. However, the prior models and the one presented here all agreed that SOD1 fibrils were produced from the immature dimeric form under reducing conditions, in which a lack of oxygen prevents oxidation reactions. While both dimeric and fibrillar SOD1 were rich in ‘β-sheet’ structures, the immature form had a barrel-like structure compared to the more compact fibrils with more internal interactions.

By design, this study investigated a homogenous group of SOD1 fibrils for consistent analysis. However, when known ALS-associated SOD1 mutations were mapped to locations on the SOD1 protein, it was found that almost 85% were in a fibril core region identified in this study, while others were in segments which connected the terminal ends of the fibril together. The authors predicted that these mutations are likely to disrupt important interactions within the fibril structure, which is likely to induce the formation of fibrils with structures and neurotoxicity distinct from what they found. This may help to explain the high degree of structural diversity found in ALS-linked SOD1 proteins, with different ‘strains’ displaying different properties.

The information in these sort of studies is by its nature fairly dense and difficult to both interpret and explain without specialist knowledge. However, a key takeaways from this research was that we can identify and distinguish the structures of different forms of aggregating proteins, and identify the sites in the protein which are most likely to contribute to this aggregation. Understanding these structures allows us to better predict how the spread of ALS pathology occurs, as well as being able to identify disease-related versus healthy forms of the protein. If we are able to more fully explore how different mutations interact with this structure we may even be able to identify subtypes of ALS or predict its progression based on the structural properties of the misfolded proteins present. While the development of treatments which directly target the structural variations are difficult due to the broad range of mutations which can occur, it may be that one day we can create a large enough body of information to fully characterise and address each. Alternatively, it may be possible to?create a treatment which ‘neutralises’ the fibril core somehow, halting the prion-like spread and thus the spread of ALS pathology.

TAGS: structural, mechanistic, SOD1, cryo-EM, protein

Plasma taurine is an axonal excitability?translatable biomarker for amyotrophic lateral sclerosis

Nakazato et al., 2022

One of the factors which makes ALS difficult to effectively research is that the core pathology takes place in the central nervous system (CNS), with the brain and spinal cord separated from the peripheral body by strong physiological boundaries. Accessing it requires invasive and often painful procedures such as spinal taps, and many tests can only be performed post-mortem due to the sensitivity of the systems there. As such, some researchers seek to use increasingly more sensitive measures to find usable markers in the periphery, which can be collected in more minimally invasive manners. These often investigate blood or DNA, hoping that the flow of proteins into the periphery or systemic genetic changes in ALS can give us valuable hints to identify, predict or understand the disease. This study investigated the blood plasma of ALS patients, a biofluid which contains proteins but not DNA-laden blood cells. Specifically they used two forms of mass spectrometry, a process in which the mass-to-charge ratio of particles are used to determine their chemical composition and structure. These were assessed against clinical phenotypes and electrophysiological measures of motor nerve excitability (which is lost in ALS) in ALS patients and healthy controls.

Changes were detected in several pathways connected to the generation of energy in skeletal muscle, cells which control voluntary movement and which are denervated and degraded in ALS. Key among those were the fatty acid β-oxidation and creatine pathways, and these changes were able to reliably distinguish ALS patients and healthy controls. Creatinine, a waste product of muscle activity which can bypass cell walls, was decreased in ALS patients. Given the loss of muscle function as ALS progresses this is unsurprising, and could be assume to be merely a side effect of muscle degeneration. Fatty acid β-oxidation is the pathway by which fatty acids are broken down to produce energy, mainly in skeletal muscles. Several fatty acids and other members of the pathway were decreased in ALS patients. However, other molecules with similar roles (eg carnitine) were not likewise changed, suggesting a broader role in the pathological pathway.

Of the specific molecules involved, taurine was identified as an independent biomarker of ALS. It was strongly correlated with SDTC, a measure of axonal excitability which predicts neuron survival in ALS. Taurine has shown various benefits in ALS clinical models including protecting against glutamate toxicity, oxidative stress and sodium channel dysfunction. It has been noted previously to be elevated in ALS patients, possibly as an active protective measure by the body. A drug containing taurine conjugated to ursodeoxycholic acid has shown some effectiveness in ALS, decreasing neuronal death and slowing functional decline in patients.

While peripheral measures of ALS are often distanced from the ‘true’ pathology and so only fairly broadly functional as biomarkers, the progression of information technology and bioinformatics has allowed the integration of numerous measures to create a more accurate picture. While the sample size for this study was relatively small and the statistical evidence for the involvement of each metabolite is individually low, it may be possible to integrate such data into an accurate model to identify and predict the disease course of ALS without invading the CNS. While peripheral treatments are extremely unlikely to be able to address the core pathology of ALS, they may be able to slow the death of neurons and so the development of symptoms such as loss of muscle function. Evidence of successful treatments using broadly available supplements such as taurine is obviously valuable, as allowing patients to even slightly mitigate ALS symptoms can have significant benefits to quality of life.

TAGS: biomarkers, therapeutic, human

Metabolic Profile and Pathological Alterations in the Muscle of Patients with Early-Stage Amyotrophic Lateral Sclerosis

Lanznaster et al., 2022

It is almost impossible to identify ALS before the onset of or even at the time of clinical symptoms, as cases will often start with minor symptoms which can be ignored as regular body changes or other diseases. As a result, the delay between onset and diagnosis averages between 9 and 13 months. But even that is likely misleading, as pathological processes begin long before symptom onset and gradually build up to reach the point where neurons are killed and muscle function is lost. There is consequently a great deal of importance placed on research which can pre-emptively predict ALS or diagnose it in its early stages. This study sought to use metabolic features of the muscle and blood serum of ALS patients as a diagnostic tool capable of distinguishing ALS patients from healthy controls without ALS pathology. Notably, the researchers sought out samples from patients who were in relatively early stages of ALS, hoping to identify early metabolic changes to more effectively predict the disease.

The study was able to identify several major alterations in the metabolic profiles of muscle tissue in ALS patients compared to healthy controls. These included alterations in the metabolism of unsaturated fatty acids in the serum, carbohydrates in the muscle and amino acids pathways in both the serum and muscle. Most of the metabolites noted were related to known pathological mechanisms in ALS, reinforcing the disease relevance of this data. The biomarkers found were combined into several more complex prognostic models. The model from the serum metabolome was able to predict disease duration, while the muscle model was able to predict variation in weight in ALS patients.

Several possible independent biomarkers were also identified. Elevated muscle levels of lauroylcarnitine, a pro-inflammatory saturated fatty acid, were associated with low forced vital capacity (measure of lung function), and were associated with an unfavourable prognosis in ALS patients. Citramalate, a metabolite often used to diagnose dysfunction of the gut, was elevated in both the serum and muscle of ALS patients compared to controls. This may be linked to changes in levels of beneficial bacteria, some of which produce chemicals such as nicotinamide which improve ALS motor function. Several protective pathways were also altered, although those appeared to be elevated as a protective response against ALS. These included two antioxidant proteins, SOD3 and GLRX2, which were proposed to be linked to mitochondrial dysfunction and mutant SOD1 aggregation, a well-explored possible disease driver. Similarly elevated was glycine, which activates the glutathione which protects against destructive oxidative stress which has been clearly demonstrated in ALS.

The use of metabolic biomarkers to diagnose and predict the progression of ALS is valuable in that it can be done without invasive surgery on the central nervous system to extract more directly affected biofluids. If muscle and blood can be used to accurately predict or diagnose the disorder, this would make large-scale testing much more plausible, allowing us to catch cases earlier. Treatment in early disease stages is dramatically more effective than further down the line, so early diagnosis is in the best interest of patients. On the research front, being able to identify changes without the full systemic breakdown which can occur in more developed cases allows a clearer investigation of disease-driving factors, so everyone benefits. While the predictive models shown here were relatively limited, this is likely a factor of the relatively small sample size. Even if metabolic biomarkers can only be used to predict disease duration and weight change, incorporating these with other models such as morphological features or lipidomic changes may help to develop a more powerful collective model greater than the sum of its parts.

TAGS: biomarkers, predictive, metabolites, metabolism, human

Lipid Metabolism Is Dysregulated in the Motor Cortex White Matter in Amyotrophic Lateral Sclerosis

Sadler et al., 2022

Lipids are a group of biomolecules including fats, oils and waxes which play a variety of roles including hormone regulation, carrying nerve impulses, cushioning organs and storing energy. A common feature of ALS is hypolipidemia, an abnormally low level of lipids in the blood. This is accompanied by a variety of associated pathological features including defects in energy metabolism, with high-fat diets extending survival in ALS patients. Lipids play a particularly important role in the brain, with approximately 50-60% of the brain’s dry weight being lipids. Previous studies of lipids in ALS were all performed in grey matter, which is predominantly the cell bodies of neurons. This study extended their research to white matter, the nerve fibres extending from these cell bodies which often make up the deeper tissues of the brain. As these fibres are surrounded by lipid-heavy (70-80%) myelin sheaths, lipids were a key element of this exploration. White matter from 14 ALS patients and matched controls were assessed for lipidomic features, gene expression of lipid-metabolism enzymes and levels of myelin-associated proteins.

This study found some interesting sources of variation and consistency between ALS patients and controls. Lipid metabolism was changed in the white matter of the motor cortex, and lipid-metabolising genes were expressed at different levels in ALS patients. There was also a significant increase in cholesterol levels in the ALS motor cortex compared to controls, which has been previously identified as an ALS risk factor. However, there were no changes in the levels of some key myelin proteins, suggesting that the composition of myelin changed while the amount did not. Given the elevated level of myelin and its lipid- and cholesterol-rich composition, myelin was identified as a likely source of these changes. It is notable however that previous research on myelin protein levels have been particularly inconsistent, with separate studies finding higher, lower and unchanged levels. Another possible factor is oligodendrocytes, supporting cells in the central nervous system which carry out the function of applying myelin to neurons. They also have several roles in maintaining neuron metabolism, and ALS patients lack the tightly controlled metabolic regulation which these would typically provide.

It is fairly conclusive at this point that significant changes in lipid metabolism occur in ALS, although the specifics are still a matter of debate. This study proposed that either one or both of changes in the composition of myelin and the function of the myelinating oligodendrocyte cells are to blame. The role of oligodendrocytes is fairly unexplored in ALS research, so it is difficult to make any clear conclusions. However, if this can be confirmed it may represent a new target for therapeutic interventions. The data collected here suggests that it is still carrying out its myelinating function even if the composition of the myelin itself is changed, and so addressing the change in myelin may help to restore function without needing to alter these cells. Another potential use of this paper is concreting the lipid-associated changes for use as biomarkers. These could incorporate the levels of lipids themselves or lipid- or myelin-associated proteins, helping to distinguish ALS patients from healthy individuals and so potentially playing a future role in diagnosis. All in all, while the associations between lipids and ALS are unclear, they may represent another means of understanding and quantifying this infuriatingly diverse disorder.

?TAGS: mechanistic, lipids, human

References

Behler, A., Müller, H.-P., Del Tredici, K., Braak, H., Ludolph, A.C., Lulé, D., and Kassubek, J. (2022). Multimodal in?vivo staging in amyotrophic lateral sclerosis using artificial intelligence. Annals of Clinical and Translational Neurology n/a.

Gleixner, A.M., Verdone, B.M., Otte, C.G., Anderson, E.N., Ramesh, N., Shapiro, O.R., Gale, J.R., Mauna, J.C., Mann, J.R., Copley, K.E., Daley, E.L., Ortega, J.A., Cicardi, M.E., Kiskinis, E., Kofler, J., Pandey, U.B., Trotti, D., and Donnelly, C.J. (2022). NUP62 localizes to ALS/FTLD pathological assemblies and contributes to TDP-43 insolubility. Nature Communications 13, 3380.

Lanznaster, D., Bruno, C., Bourgeais, J., Emond, P., Zemmoura, I., Lefèvre, A., Reynier, P., Eymieux, S., Blanchard, E., Vourc, H, P., Andres, C.R., Bakkouche, S.E., Herault, O., Favard, L., Corcia, P., and Blasco, H. (2022). Metabolic Profile and Pathological Alterations in the Muscle of Patients with Early-Stage Amyotrophic Lateral Sclerosis. Biomedicines 10.

Murray, K.A., Hughes, M.P., Hu, C.J., Sawaya, M.R., Salwinski, L., Pan, H., French, S.W., Seidler, P.M., and Eisenberg, D.S. (2022). Identifying amyloid-related diseases by mapping mutations in low-complexity protein domains to pathologies. Nature Structural & Molecular Biology.

Nakazato, T., Kanai, K., Kataura, T., Nojiri, S., Hattori, N., and Saiki, S. (2022). Plasma taurine is an axonal excitability-translatable biomarker for amyotrophic lateral sclerosis. Scientific Reports 12, 9155.

Sadler, G.L., Lewis, K.N., Narayana, V.K., De Souza, D.P., Mason, J., Mclean, C., Gonsalvez, D.G., Turner, B.J., and Barton, S.K. (2022). Lipid Metabolism Is Dysregulated in the Motor Cortex White Matter in Amyotrophic Lateral Sclerosis. Metabolites 12.

Saleki, K., Mohamadi, M.H., Banazadeh, M., Alijanizadeh, P., Javanmehr, N., Pourahmad, R., and Nouri, H.R. (2022). In silico design of a TLR4-mediating multiepitope chimeric vaccine against amyotrophic lateral sclerosis via advanced immunoinformatics. Journal of Leukocyte Biology n/a.

Wang, L.-Q., Ma, Y., Yuan, H.-Y., Zhao, K., Zhang, M.-Y., Wang, Q., Huang, X., Xu, W.-C., Dai, B., Chen, J., Li, D., Zhang, D., Wang, Z., Zou, L., Yin, P., Liu, C., and Liang, Y. (2022). Cryo-EM structure of an amyloid fibril formed by full-length human SOD1 reveals its conformational conversion. Nature Communications 13, 3491.