May's ALS Research Roundup

Bacteria-Derived Protein Aggregates Contribute to the Disruption of Host Proteostasis

Reference: (Walker et al., 2022)

While ‘traditional’ pathogens such as viruses and pathogens are almost certainly not the primary drivers of ALS pathology, there is a body of evidence that shows that they may play a role in some ALS cases. In particular, they have been proposed to contribute to the bundling of disease-associated misfolded proteins into toxic aggregates which are commonly held to be disease drivers. However, the evidence for the role of bacteria is complicated, with studies showing that antibiotics increase risk of but slow progression of ALS and similar neurodegenerative diseases. This was proposed to be due to interactions with the gut microbiota, a mass of beneficial microorganisms including bacteria which live in the digestive tracts.

In this study, genetic and chemical methods were used to demonstrate the effects of bacteria on the folding of proteins, and how this led to the development of toxic protein aggregates. This was done in C. Elegans (nematodes), which were cultured in bacteria-seeded growth media.

The level of bacteria-derived protein aggregates was shown to correlate positively with bacteria-induced polyglutamine (polyQ), a previously disease-associated protein segment. Several specific bacteria were also assessed in relation to this protein aggregation, with P. aeruginosa and P. corporis inducing and suppressing aggregation, respectively. This suggests that specific bacteria have different and even opposite effects on the aggregation of proteins, rather than aggregation being affected by the presence of bacteria in general. This is notable, as while various microbial products have previously been shown to induce aggregation of host proteins this study indicates that there are exceptions. Some which have been shown to contribute to protein aggregation, particularly disease-associated amyloid proteins, were found to secrete surfactants which enhance protein aggregation. It is plausible that this and similar mechanisms are the drivers of the association between bacteria and disease, rather than aggregates developing around the bacteria themselves.

This study also highlighted protective effects exerted by bacteria which produce butyrate, a short-chain fatty acid which benefits host health. Neurodegenerative diseases have been linked to lower levels of butyrate and butyrogenic bacteria, and this study found that low butyrate levels led to increased production of bacteria-derived protein aggregates. The authors proposed a balance between these protective and other toxicity-driving bacteria, and suggested that the disruption of this balance (tipping towards toxicity) may contribute to disease-associated disruption of proteostasis.

While it is easy to ‘paint all bacteria with the same brush’, it is becoming increasingly clear that many play beneficial or even necessary roles in human physiology. When they began to emerge in ALS research as a possible disease-linked factor, several theories arose as to how they drove the disease. ‘Overloading’ the immune system was one, and serving as nuclei around which disease-associated proteins could aggregate was another. However, as with many aspects of ALS, the reality seems less simple. While this study addressed protein aggregation in general rather than specifically addressing ALS-linked proteins, it did show that bacteria can both induce and counteract protein aggregation. This limits the likely value of broad-spectrum antibiotics as an ALS therapeutic but may help us to develop more targeted measures. It may be that more specific therapeutics, whether drugs or gene therapies, can be developed to address aggregation-enhancing bacteria or increase levels of the protective. It may even be that these changes can be made with simple lifestyle changes, providing clinical benefits without chemical intervention at all. However, as shown here the processes involved are complex and multi-faceted, and further research is needed to determine exactly which bacteria are on which side of the balance.

Tags: mechanistic, microbiota, bacteria, non-ALS, nematode

Patients’ stem cells differentiation in a 3D environment as a promising experimental tool for the study of amyotrophic lateral sclerosis

Reference: (Scarian et al., 2022)

Cell culture models are extremely valuable for many types of research, particularly regarding diseases such as ALS. Ethics considerations prevent most testing on humans, and animal models have the potential to differ significantly from human physiology and so give flawed conclusions. However, cell models are limited in scope, in that they can only represent a small area of tissue with limited avenues for complex interactions. The advent of 3D cell cultures has challenged this limitation, providing the ability for significantly more comprehensive models of cell dynamics and interactions. They often involve cells which are incorporated into a ‘bioink’ and suspended in a hydrogel-based 3D construct using a bioplotter. This method has been used to create models of various cell types including blood vessels, bones, cartilage and heart cells.

This study addressed the use of 3D cell cultures incorporating induced pluripotent stem cells rather than plotting mature cells, and their potential for use in the study of ALS. Peripheral blood monocytes (PBMCs; immune cells derived from blood) were reprogrammed into neural stem cells in a 2D cell culture before being transferred into a 3D hydrogel construct. They were then differentiated into motor neuron progenitor cells and finally mature motor neurons.

Cell differentiation was demonstrated to have occurred correctly, with the mature cells from both ALS patients and controls living and proliferating in the hydrogel. The properties of these differentiated cells were then assessed, including their electrophysiological and mechanical properties and ability to survive. The stem cell-derived motor neurons were not harmed by inclusion in the hydrogel or the printing process, with cell viability actually increasing after the printing process. It may be that the cell-cell interactions enabled by the 3D structure provide benefits for cell survival compared to a traditional 2D culture. The hydrogel also did not interfere with the electrophysiology of the cells, which is key to communication and interactions between neurons. Differences in cell viability between ALS and control groups were also maintained when the cells were printed in 3D. While this does mean that ALS samples did not have as many cells reach the mature motor neuron phase, this is important to have an accurate disease model. Similarly, several protein biomarkers of ALS compared to controls were shown in both motor neuron progenitors and mature motor neurons.

While studies such as this do not provide any new understandings of the mechanisms of ALS, they are essential foundations for future research. The key outcomes of this study were that ALS patient-derived stem cells could be differentiated into stable 3D cell cultures, both gaining viability and complexity compared to 2D models and retaining disease-associated differences. Considering how much research (particularly treatments) shows promising success in laboratories and yet fail in treating living people, more accurate models of living systems are needed. This is particularly key in multifactorial diseases such as ALS, where interactions between neurons and other types of cells have been demonstrated in animal models and are difficult to represent in a traditional 2D cell culture.

Tags: model, iPSC, human_cell

TDP?43 regulates cholesterol biosynthesis by inhibiting sterol regulatory element?binding protein 2

Reference: (Egawa et al., 2022)

One of the many non-disease-driving factors which contribute to ALS is the levels of lipids, nonpolar biomolecules such as fats and oils. One study found that patients with abnormally high lipid levels had delayed onset of ALS symptoms, suggesting a protective role. Accordingly, low levels of lipids were found to occur in mouse models of ALS early in disease progression, before the onset of symptoms, suggesting a role in pathogenesis. A genome-wide study showed that cholesterol pathways were disrupted in ALS (dyslipidemia), with the disease-driving protein TDP-43 likely to be involved. This study further investigated TDP-43-driven gene expression changes related to dyslipidemia. Expression microarray and RNA sequencing was performed using cell lines expressing high levels of TDP-43 compared to non-TDP-43 control cells, quantifying the expression of genes into RNA which would go on to generate proteins. This was paired with an assessment of the amount of cholesterol in the cerebrospinal fluid (CSF) of human ALS patients and spinal fluid of ALS mice.

This study identified 434 significantly altered genes between ALS and control cells. Notable among these was sterol regulatory element-binding protein 2 (SREBP2), a master regulator of cholesterol homeostasis, and several genes downstream of it. Elevated TDP-43 impaired transcription of the SREBP2 gene into RNA, lowering its functional activity and inhibiting the generation of cholesterols. Accordingly, the amount of cholesterol was significantly decreased in the spinal cords of ALS patients and ALS mice compared to controls. Interestingly, the same effect was not found in the blood serum, suggesting that this effect is limited to the nervous system. SREBP2 and several other prominent proteins were assessed for their ability to bind to TDP-43 mRNA, and SREBP2 was able to do so. This suggest that it may be inhibited through direct interaction with or incorporation into aggregates with TDP-43. 

Understanding the role which cholesterols play in ALS is key to developing targeted therapies. Previous studies have shown that the ‘brute force’ method of introducing cholesterol-like drugs was insufficient, failing in a phase 3 clinical trial. It is unclear whether this is due to the mechanism of disease association being more complex than simply removing cholesterol function, or if the drug used was insufficient in some manner. Identifying SREBP2 as a key element may provide a valuable avenue of approach, such as using gene therapies to upregulate it into normal levels in potential disease cases. Such approaches have been shown to have some effect in Huntington’s disease but have yet to be attempted with ALS.

Tags: mechanistic, lipids, cholesterol, TDP-43, RNA-seq, human, mouse

BDNF and pro-BDNF in amyotrophic lateral sclerosis: a new perspective for biomarkers of neurodegeneration

Reference: (Riolo et al., 2022)

Brain derived neurotrophic factor (BDNF) is an abundant and tightly regulated protein in the central nervous system and is involved in several key processes including both creation and death of neurons. It goes through several stages before entering its functional form, with the most relevant here being the second-last stage as the precursor protein pro-BDNF. Pro-BDNF is bioactive in its own right, with biological functions independent of mature BDNF. It triggers several signalling pathways which some have theorised have opposite effects to mature BDNF. The maturation of pro-BDNF to BDNF is a delicate balance which can influence cell survival or death, and which has a notable role in neurodegenerative diseases. As BDNF can cross the blood-brain barrier (BBB) which isolates the CNS from the peripheral body, it has the potential to be a valuable biomarker, allowing more accurate diagnosis and prediction of ALS.

In this study, patients with ALS, patients with other neurodegenerative diseases and healthy controls had genetic analysis performed and both CSF and blood serum extracted. These were used for several analytical processes intended to detect levels of BDNF and pro-BDNF in the CSF and serum. They observed higher BDNF and lower pro-BDNF levels in the serum of patients with both ALS and other neurodegenerative disorders compared to healthy controls, and the ratio of BDNF:pro-BDNF proved accurate in distinguishing ALS patients from controls. BDNF levels in the CSF were not notably different between ALS patients, healthy controls, and patients with other neurodegenerative diseases. Pro-BDNF levels in the CSF were unable to be assessed due to the very low concentrations of pro-BDNF, below the detection limits of current devices. Interestingly, higher BDNF levels in the CSF and lower pro-BDNF levels in the serum were linked to faster-progressing ALS with shorter survival times. This was observed despite the lack of association between CSF levels of BDNF and disease conditions, possibly indicating that this is part of a disease-modifying process independent of serum levels. ALS variants containing the C9orf72 repeat expansion mutation were found to have different properties to other forms of ALS, with BDNF levels comparable to controls and significantly lower than other ALS patients. It is unclear why this might be.

The authors proposed that rather than BDNF, pro-BDNF is a valuable candidate for an ALS biomarker. However, further research is required, particularly assessment of pro-BDNF levels in the CSF. The ratio between the two seems to have immediate potential as a biomarker, with the divergence between the two becoming very pronounced in disease conditions. While some studies have investigated the functional role of BDNF in ALS, and even considered it as a therapeutic which counteracts glutamate-induced neurotoxicity, it can also reduce long-term survival of motor neurons. As such, its diagnostic role seems likely to be its greatest contributing to ALS research. Accurate neurodegeneration biomarkers can help to determine the ‘true’ onset of ALS pathology, rather than just the onset of clinical symptoms, and in doing so enable early treatment. Most effective therapies have their greatest effectiveness in early stages of the disease in mouse models, although the difficulty in predicting onset in humans means that clinical assessment of this is very rare.

Tags: biomarker, BDNF, human

Alterations in metabolic biomarkers and their potential role in amyotrophic lateral sclerosis

 Reference: (Li et al., 2022)

Metabolism refers to a broad range of processes by which the body converts fuel, particularly glucose or lipids, into energy. Metabolic disruption has been frequently reported in ALS. This takes the form of altered glucose uptake and reduced glucose tolerance, increased lipid breakdown and peroxidation of lipids in mouse models of ALS. Lipid metabolism disorders have also been widely reported in ALS patients, presenting with elevated cholesterol, triglycerides, and other mixed dyslipidemias. Cholesterols in particular have possible links to protein aggregation, and so may contribute to more core pathogenic processes. Fatty adipose tissues have also been shown to secrete adipokines, a group of chemicals with roles in metabolic processes including fat distribution, energy expenditure, appetite and satiety regulation, insulin secretion and sensitivity, and inflammation. As some of these adipokines can cross the blood-brain barrier and function in the brain, they may provide a means of impacting brain function as part of ALS-driven processes. Dietary therapies have shown some therapeutic potential in animal models and ALS patients, including high-energy/high-fat diets and intake of medium-chain fatty acids or beta-hydroxybutyrate.

In this study, ALS patients and healthy controls were assessed for various metabolic biomarkers. Several non-metabolic measures were also taken, including body composition and clinical features of ALS. In general, patients with lower body fat, visceral fat and fat in limbs showed faster disease progression. ALS patients were also found to have higher levels of several adipokines (adiponectin, adipsin, resistin, and visfatin) and other metabolic biomarkers including glucagon, gastric inhibitory peptide, and C-peptide. Adiponectin and leptin were notable for being positively correlated with visceral fat, as well as faster ALS progression. Differences in leptin may also contribute to sex-linked differences in rates of ALS, as sex hormones are known to influence leptin levels. Previous data on leptin levels in ALS is highly conflicting, with different studies showing both lower and higher leptin levels in ALS patients compared to controls. This could be due to measures being taken at different times during ALS, as muscle and fat wasting can dramatically alter metabolic data. Adiponectin have been consistently reported to be elevated in ALS patients, although they did not connect it to ALS progression in the way this study did.

While not as focused as many, exploratory studies such as the one performed here can be valuable launching points for future research. The ‘shotgun approach’ is also useful to gather broad information on many molecules, genes, and other biomarkers. This and similar data contribute to a unified body of data which would not be feasible to gather in a single study. While metabolic biomarkers are difficult to conclusively assess given their dramatic variation across time, it is one of the easiest to address for patients. Dietary changes can have significant effect on the longevity of patients, although we have yet to find the ‘sweet spot’ of lipid and particularly cholesterol levels which most extends the lifespans of patients. While gene therapies such as those attempted with leptin in mice have had some effect in increasing fat mass and improving neurological function, such treatments are problematic in humans without a clear need due to the risk of health problems rising from the weight gain itself.

Tags: biomarkers, lipids, metabolic, human

Heat shock protein Grp78/BiP/ HspA5 binds directly toTDP?43 and mitigates toxicity associated with disease pathology

Reference: (Fran?ois-Moutal et al., 2022)

Proteostasis refers to the process by which proteins are regulated and maintained within living cells and is necessary for the survival of both cells and the organism at large. Proteostasis is significantly disrupted in ALS, with many of the main disease-driving proteins disrupting the processing of RNA (which produce proteins). Several downstream components of the disease interfere with different stages in the proteostatic network, particularly the removal of defective proteins, resulting in a cascade of defective proteins and dysfunctional regulatory systems. One of the core regulatory systems is the family of chaperone proteins which assist in protein misfolding, degradation, and the clearance of aggregated protein. A major subgroup of this is the heat shock proteins (HSP), which are produced in response to stressors and some of which are often protective against neurodegenerative diseases such as ALS. This study investigated 13 isoforms of HSP70 with regards to TDP-43, one of the main disease-driving proteins in ALS, with samples derived from human cells. Previous studies have shown that the HSP70-mediated stress response appears to be disrupted in ALS motor neurons, and that overexpression of molecular chaperones such as HSP70 can prevent the formation of TDP-43 aggregates. However, the specific interactions between HSP70 proteins and TDP-43 have not been explored, and so this study sought to do so.

The HspA5 and HspA8 isoforms were enriched in TDP-43 in the nuclei, but not the cytoplasm where TDP-43 moves to in ALS. They were also found to bind to nuclear TDP-43 in the absence of stress, suggesting that they interact with healthy TDP-43 in some manner, and that this function may be lost in ALS. HspA5 was noted to specifically bind to the RNA-binding domain of TDP-43, which is a key element in its role in RNA splicing and translation. Accordingly, the binding of HspA5 to TDP-43 was also found to be inhibited by RNA bound to the TDP-43, suggesting that only ‘free’ TDP-43 is a valid partner. This may be the means by which HspA5 controls TDP-43 folding, attaching to free TDP-43 which is more prone to misfolding- and aggregation-inducing protein interactions. The movement of TDP-43 out of the nucleus in ALS would counteract this effect, resulting in the loss of protective activity by HspA5 and contributing to the protein misfolding cascade which characterises ALS. With this in mind, overexpression of HspA5 in the prefrontal cortex of ALS patients (also observed in this study) may be an attempt to counteract this loss of function which fails as no matter how much HspA5 there is, it cannot act on TDP-43 in its new cellular compartment.

With TDP-43 being one of the very few nearly universal elements of the pathological cascade in ALS (~95% of sporadic cases have TDP-43-containing aggregates), restoring healthy TDP-43 function is a key therapeutic target. While we know some of the influence it exerts and what the downstream effects of its dysfunction are, there are many gaps in our knowledge of exactly how it functions in a healthy system. Similarly, molecular chaperones are often addressed as a whole, or in larger subgroups such as HSPs, rather than investigating their individual-level interactions. Unfortunately, this case appears to describe a relatively incidental downstream effect which contributes to the cascade rather than a primary driver. The primary driver of this effect would be the mislocalisation of misfolded TDP-43 out of the nucleus, which is a difficult problem to effectively address. There has been some research into modifications such as reintroducing nuclear localisation sequences to TDP-43, but ALS-associated aggregation makes this particularly difficult to achieve once pathology has set in. If there were some ways to induce HspA5 function in the cytoplasm that may prove beneficial, but we have no means of doing so as of yet.

Tags: mechanistic, HSP, TDP-43, human_cell

Diagnostic circulating miRNAs in sporadic amyotrophic lateral sclerosis

Reference: (Panio et al., 2022)

MicroRNA (miRNA) are small single-stranded RNA molecules which do not code for proteins, but rather play regulatory roles in the expression of other proteins from messenger RNA (mRNA). They are emerging as significant factors in ALS pathogenesis, with many being dysregulated prior to and during the main pathological cascade. miRNA function in complex, interconnected networks, with most miRNA regulating numerous mRNA targets. Their presence as part of environment-dependent epigenetic pathways is also of note, as their function can vary beyond purely genetic changes. This study was a meta-analysis which incorporated data from several previous studies which have investigated the use of miRNA as biomarkers for ALS, both for diagnostic and predictive function. They were particularly interested in the role of extracellular miRNA, whose main role is theorised to be in cell-cell communication, and which can be more readily found in circulation compared to more centralised central nervous system biomarkers.

From the accumulated research 162 miRNA were identified with previous links to ALS. However, different methodologies, tissues and sample sizes made inclusion of all of these questionable. Bioinformatic analysis of the capacity of the miRNA to distinguish ALS from healthy controls identified a panel of 10 miRNAs (miR-193b, miR-3911, miR-139-5p, miR-193b-1, miR-338-5p, miR-3911-1, miR-455-3p, miR-4687-5p, miR-4745-5p, and miR-4763-3p) with potential as diagnostic tools. They believe that levels of each of these in the blood can function independently as a diagnostic biomarker for ALS, with accuracy ranging from 66 to 83%. Some have previously been investigated in depth, with miR-338 and miR-193 confirmed in rodent models of ALS and several others found increases in blood plasma, serum, or skeletal muscle of ALS patients. Further analysis into combinations of miRNA with better combined function as biomarkers identified miR-193B and miR-4745-5p. miR-193b is linked to autophagy control while miR-4745-5p’s function is unknown, and both are downregulated in ALS. While their individual accuracies were 66 and 83% respectively, their combined accuracy was over 90% with a 100% true negative and 83% true positive rate.

The breadth of interactions in the miRNA ecosystem and the newness of study into them means that there is a great many studies researching them and coming up with different results. Meta-analyses such as this are an attempt to consolidate this data in a manner which is likely to highlight true disease-associated differences, rather than variation within individual datasets. While this is complicated by many factors, chief among them tissue-associated differences in miRNA and different studies drawing samples from different tissues, the quantity of data has value of its own. Diagnosis of ALS is also a complicated task, with wide variations in clinical symptoms, rate of progression and the delay between actual disease onset and the presentation of symptoms. There are also many other diseases that initially resemble ALS, and so treatments which may work for ALS are attempted and fail. As such, accurate diagnosis of ALS is valuable in both treatment and determining the effectiveness of treatments from a research standpoint. With miRNA able to be assessed from blood alone it presents a minimally invasive biomarker, which could be used to both confirm ALS cases and in an ideal world, identify cases before the neurodegeneration has progressed to a point that treatment is largely ineffective.

Tags: biomarker, meta-analysis, miRNA

References

Egawa, N., Izumi, Y., Suzuki, H., Tsuge, I., Fujita, K., Shimano, H., Izumikawa, K., Takahashi, N., Tsukita, K., Enami, T., Nakamura, M., Watanabe, A., Naitoh, M., Suzuki, S., Seki, T., Kobayashi, K., Toda, T., Kaji, R., Takahashi, R., and Inoue, H. (2022). TDP-43 regulates cholesterol biosynthesis by inhibiting sterol regulatory element-binding protein 2. Scientific Reports 12, 7988.

Fran?ois-Moutal, L., Scott, D.D., Ambrose, A.J., Zerio, C.J., Rodriguez-Sanchez, M., Dissanayake, K., May, D.G., Carlson, J.M., Barbieri, E., Moutal, A., Roux, K.J., Shorter, J., Khanna, R., Barmada, S.J., Mcgurk, L., and Khanna, M. (2022). Heat shock protein Grp78/BiP/HspA5 binds directly to TDP-43 and mitigates toxicity associated with disease pathology. Scientific Reports 12, 8140.

Li, J.-Y., Cui, L.-Y., Sun, X.-H., Shen, D.-C., Yang, X.-Z., Liu, Q., and Liu, M.-S. (2022). Alterations in metabolic biomarkers and their potential role in amyotrophic lateral sclerosis. Annals of Clinical and Translational Neurology n/a.

Panio, A., Cava, C., D’antona, S., Bertoli, G., and Porro, D. (2022). Diagnostic Circulating miRNAs in Sporadic Amyotrophic Lateral Sclerosis. Frontiers in Medicine 9.

Riolo, G., Ricci, C., De Angelis, N., Marzocchi, C., Guerrera, G., Borsellino, G., Giannini, F., and Battistini, S. (2022). BDNF and Pro-BDNF in Amyotrophic Lateral Sclerosis: A New Perspective for Biomarkers of Neurodegeneration. Brain Sciences 12.

Scarian, E., Bordoni, M., Fantini, V., Jacchetti, E., Raimondi, M.T., Diamanti, L., Carelli, S., Cereda, C., and Pansarasa, O. (2022). Patients’ Stem Cells Differentiation in a 3D Environment as a Promising Experimental Tool for the Study of Amyotrophic Lateral Sclerosis. International Journal of Molecular Sciences 23.

Walker, A.C., Bhargava, R., Dove, A.S., Brust, A.S., Owji, A.A., and Czy?, D.M. (2022). Bacteria-Derived Protein Aggregates Contribute to the Disruption of Host Proteostasis. International Journal of Molecular Sciences 23.