There are often many causes to a non-communicable disease, particularly neurodegenerative diseases are more a consequence of a systemic failure than caused by a specific phenomenon. The multitude of papers assigning a specific mechanism, each time different, to neurodegenerative diseases is just noise that drags down knowledge acquisition in these domains. Some authors have hinted at a phase transition to explain the misfolding of some proteins, but what triggers this phase transition was elusive.

In this post, I discuss a very general paper. https://elifesciences.org/reviewed-preprints/107962v1

In simple terms, the authors have discovered how our innate immune system launches an extremely powerful and rapid response to a tiny signal from a pathogen. This has implications for age-related diseases such as cancer or neurodegenerative diseases.

The Core Problem:

Our immune system needs to react decisively to a single bacterium or virus. This involves a massive cellular response like inflammation or programmed cell death (pyroptosis, apoptosis). However, the initial detection of a pathogen (a single molecule binding to a receptor) provides almost no energy to power this massive response.

The Discovery - "Metastable Supersaturation": The authors found that key immune signaling proteins, specifically those containing Death Fold Domains (DFDs) (like ASC, FADD, BCL10, MAVS, TRADD), exist in a unique physical state inside our cells called metastable supersaturation. These full-length adaptors retain nucleation barriers and are able to exist supersaturated in cells. In contrast, many receptors and effectors do not. This localizes the “spring-loaded” behaviour to central adaptors that link receptor sensing to downstream cell-fate decisions.

A subset of death-fold domains (DFDs) are intrinsically “supersaturable.” Using a systematic screen of 109 human DFDs with a distributed amphifluoric FRET (DAmFRET) assay in yeast, the authors show that a minority of DFDs switch from soluble → assembled in a discontinuous (nucleation-limited) manner — the hallmark of a large intrinsic nucleation barrier. These discontinuous DFDs can therefore exist metastably above their saturation concentration (Csat) while remaining soluble (i.e. supersaturated).

Imagine a supersaturated solution of sugar water. It holds far more dissolved sugar than it should be able to. It remains liquid until you drop in a single sugar crystal, which instantly triggers the entire solution to crystallize.

Similarly, these DFD proteins are present in concentrations far higher than their natural solubility limit. They are kept in a soluble, "primed" state only by a high energy barrier that prevents them from spontaneously assembling (like the sugar needing a seed crystal).

This state acts as a long-term energy reservoir. The cell expends energy to produce and maintain these high levels of protein, storing potential energy for a future immune response. The authors show that tissues/cell types with shorter lifespans (e.g., monocytes) tend to express higher adaptor supersaturation than long-lived cells (neurons), suggesting a trade-off between rapid innate responsiveness and longevity. They also find conservation of nucleation barriers in distant taxa (fish, sponges, bacteria), indicating the mechanism is ancient.

How It Works for Immunity: When a pathogen is detected (the initial signal), the pathogen-bound receptor acts as the "seed crystal." This seed triggers the instantaneous, explosive polymerization of the supersaturated adaptor proteins (like ASC or FADD). This amplification process consumes the stored energy from supersaturation, converting it into a massive biochemical signal that leads to inflammation or cell death.

This allows for a response that is immediate, decisive, and independent of the cell's current metabolic energy (which is often hijacked by pathogens).

The Trade-Off is Immunity vs. Longevity:

This mechanism comes with a cost. Maintaining a supersaturated, "primed" state means there's always a risk of a spontaneous, accidental activation (a stochastic nucleation event). This would lead to unwanted inflammation or cell death without any infection. The authors found evidence that this trade-off is real: short-lived immune cells (like monocytes) have much higher levels of supersaturation than long-lived cells (like neurons). This suggests a fundamental thermodynamic drive where the need for strong immunity may inherently limit a cell's lifespan.

The authors also showed this system is highly specific (DFDs from one pathway don't accidentally trigger others) and that the mechanism is evolutionarily ancient, found in everything from humans to sponges to bacteria, indicating its fundamental importance.

This groundbreaking discovery opens up entirely new avenues for treating a wide range of diseases by targeting this "supersaturation engine."

1. Autoinflammatory and Autoimmune Diseases Examples: Crohn's disease, rheumatoid arthritis, lupus, CAPS (Cryopyrin-Associated Periodic Syndromes), type 1 diabetes.

2. Infectious Diseases Examples: Sepsis, severe viral infections (e.g., COVID-19, flu).

3. Cancer Application: Some cancers evade the immune system by preventing immune cells from initiating cell death (apoptosis) in cancerous cells. They might do this by interfering with the supersaturation or nucleation of proteins like FADD.

4. Neurodegenerative Diseases Examples: Alzheimer's, Parkinson's, ALS.

Therapeutic Strategy: This research provides a deeper biophysical understanding of how proteins form aggregates. Insights into controlling nucleation barriers could lead to strategies for preventing the initial "seed" event that sparks the catastrophic aggregation of proteins like amyloid-beta or alpha-synuclein.

Risks, trade-offs, and practical challenges

Immunity vs longevity trade-off. The authors argue a thermodynamic tradeoff: lowering supersaturation protects cells from spontaneous death but reduces rapid responsiveness to pathogens. Therapies that blunt supersaturation may increase infection susceptibility.

Off-target/cross-seeding risk. Although the interactome is relatively specific, some cross-nucleation exists (e.g., PYD↔DED). Inhibiting one adaptor could have unintended effects on other pathways, or conversely, seeding one adaptor therapeutically could accidentally trigger another.

Drugging interfaces is hard. Filamentizing interfaces and nucleation kinetics are complex to target with small molecules; biologics or degradation approaches may be more tractable but have delivery challenges.

Temporal and quantitative control required. Because the system is switch-like, small quantitative changes in concentration or barrier height can produce large outcome differences; therapies need tight control to avoid tipping the balance toward immunodeficiency or hyperinflammation.

In conclusion This study moves beyond simply listing the components of immune pathways to explaining the fundamental physics and energy dynamics that make them work. By understanding that immunity is powered by a "loaded spring" mechanism of metastable supersaturation, we can now think about designing much smarter, more precise drugs that either stabilize this spring (for autoimmune diseases) or trigger it on command (for cancer).

Can Exercise Mimic Dopamine Effects on Brain Signals in Parkinson’s Disease?

Parkinson’s disease (PD) is a progressive neurological disorder that affects movement and is partly caused by the loss of dopamine-producing neurons deep in the brain. While medications that replenish dopamine can help control symptoms, researchers are increasingly asking: Can other interventions, like physical exercise, offer similar benefits?

A recent study investigated this by examining how regular, vigorous cycling influences brain activity in people with Parkinson’s. The findings don’t suggest that exercise replaces medication, but they do reveal something intriguing: exercise may partially imitate how dopamine alters the brain’s internal rhythms.

What kind of exercise was used in the study? The study focused on a specific type called dynamic cycling. It’s not just casual biking; it involves riding a motor-assisted stationary bicycle at a steady, fast pace, usually between 75 and 90 pedal rotations per minute. This style of cycling has been linked to improvements in motor symptoms like stiffness and slowness in people with Parkinson’s. But this study wasn’t about clinical symptoms; it was about electrical signals inside the brain and what they reveal. What did the researchers do? Nine individuals with Parkinson’s, all of whom had deep brain stimulation (DBS) implants, participated in up to 12 cycling sessions over a month. These implants allowed researchers to directly record brain activity from a small structure deep in the brain called the subthalamic nucleus (STN)—a region strongly involved in movement control and a common target in Parkinson’s treatments.

Before and after each exercise session, the researchers measured electrical signals from the brain, called local field potentials (LFPs). They analyzed two types of signals: - Periodic activity: rhythmic brain waves like beta waves. - Aperiodic activity: the background “noise” of brain signals, analyzed through a measure called the 1/f exponent.

What is the relationship between the hindbrain and subthalamic nucleus? Parkinson’s primarily involves the degeneration of dopamine-producing neurons in a midbrain area called the substantia nigra pars compacta (SNc). These neurons produce dopamine, a neurotransmitter vital for controlling movement. They project to other brain regions, notably the striatum. Loss of dopamine in the striatum causes the main motor symptoms of PD: tremors, slowness, stiffness, and balance issues.

The striatum uses dopamine to regulate two pathways: - The direct pathway (which promotes movement) - The indirect pathway (which inhibits movement)

The subthalamic nucleus (STN) is part of the indirect pathway and sends excitatory signals to the globus pallidus internus, which inhibits the thalamus, which in turn controls activity in the motor cortex.

In Parkinson’s loss of dopamine leads to overactivity of the indirect pathway, increasing activity in the STN. This overstimulates the GPi, which then excessively inhibits the thalamus. As a result, the thalamus cannot properly activate the motor cortex, causing slowed and reduced movement.

What did the researchers find? After a month of exercising, consistent changes appeared in brain signals from the dorsolateral part of the STN: - Power of rhythmic activity increased, indicating stronger brain waves. - The 1/f exponent also increased, suggesting the background neural activity became more organized and less noisy.

Interestingly, similar increases in the 1/f exponent occur when people take dopamine-based medication. This suggests that repeated exercise might induce brain states similar to those achieved with drug therapy, at least in certain regions.

However, not all areas of the STN showed changes. The ventral part of the STN remained unaffected, highlighting that the effects were both region-specific and gradual.

What does this mean? This study doesn’t claim that exercise cures Parkinson’s, nor does it prove symptom improvement directly linked to the brain changes observed. Instead, it shows that repeated, structured exercise can produce measurable alterations in brain electrical activity—some resembling the effects of dopamine therapy.

In other words, exercise might help “tune” the brain’s motor circuits to support better movement control, even if the mechanisms differ from medication. By directly measuring brain activity, this study provides early evidence that exercise can influence key brain areas involved in Parkinson’s.

It’s a reminder that improvements in Parkinson’s may come not only from drugs and surgery but also from consistent, targeted physical activity that stimulates the brain in new ways.

There are two ways to counteract the effects of dying neurons, and in both cases, physical activity plays a role. The first involves neighboring neurons forming new synapses to replace those that are lost. The second involves stem cells creating new neurons, but this process is limited by several factors: some neurons have complex structures connecting multiple brain areas, and in older individuals, the supply of stem cells is significantly decreased.

Although effective symptomatic therapies have been developed for Parkinson's disease (PD), treatments that modify the disease itself still do not exist.

Drug repurposing studies help identify potential disease-modifying treatments. One advantage of drug repurposing is that, since these drugs are already approved, their safety profiles are known. However, further industry investigation is unlikely because companies hesitate to invest in drugs whose intellectual property is owned by others. Generally, they might consider repurposing their own drugs, which extends patent life by claiming new uses. This can also lead to questionable combinations of approved drugs just to file a patent.

In an issue of Neurology, Tuominen et al. publish a nationwide observational cohort study in Norway aiming to identify new candidates for disease modification in Parkinson's disease. They analyzed Norwegian health registries from 2004 to 2020 to identify people with Parkinson's. The study included 14,289 individuals and found 23 drugs among a total of 219 drugs associated with a reduced eight-year mortality risk.

Nonetheless, this reduction remains minor, and there is no proof that these drugs caused improved survival. The authors note that, although their findings are exploratory and cannot be directly applied clinically yet, the identified drugs could be considered for future clinical trials. Indeed, funding should be sought for these future trials, and since it is almost always private investors who finance clinical trials, they would require more information before making any commitments.

However, this study has notable strengths compared to similar studies. Tracking specific disease milestones is crucial for evaluating the effects of potential disease-modifying treatments.

For nearly all of the 23 drugs discussed by Tuominen et al., the adjusted eight-year mortality curves for Parkinson's patients and healthy controls diverged and showed different slopes, suggesting these compounds may have a disease-modifying effect. This pattern has not been observed in other trials testing potential disease-modifying drugs.

When interpreting Tuominen et al.'s results, it is important to remember that correlation does not necessarily imply causation. A drug associated with lower mortality does not automatically prove that it caused the reduction.

On the one hand, the decreased mortality might simply reflect that healthier Parkinson's patients are more likely to be prescribed these drugs. For example, patients using tadalafil (for erectile dysfunction) may have better overall health and longer survival.

On the other hand, for the more difficult cases, some physicians may have focused solely on treating essential symptoms rather than prescribing a large number of medications.

Additionally, the number of patients in the "treatment" group is often small; for instance, only 170 patients used Levothyroxine sodium.

These days, major news stories about neurodegenerative diseases are rare. One headline in the trade press claims: "Harmless Virus Could Trigger Parkinson's Disease."

Most cases of Parkinson's disease are idiopathic, meaning the cause is unknown. However, several genetic mutations can also lead to this neurodegenerative disease, with approximately 20 to 25 percent of cases having a genetic cause. One of these mutations is in the gene encoding LRRK2, which can result in enzyme levels two to three times higher than normal. These mutations are more common in North African, Arab, Berber, Chinese, and Japanese populations.

There is some good news coming out on this topic, but I'll talk about another publication.

Every viral infection alters the genetic material of some (but not all) cells in our body. This is how a virus forces the host cell to rapidly produce thousands of copies of itself. In most cases, viruses are first abruptly eliminated by the innate immune system, killing the host cell. The adaptive immune system uses a more intelligent mechanism, producing specific antibodies that bind to the virus and often render it non-infectious. This process is called humoral immunity. The LRRK2 protein is highly expressed in immune system cells, particularly in response to bacterial pathogens like Salmonella.

The hypothesis that some neurodegenerative diseases are caused by viral infections is attractive because it could explain why these diseases appear with age and why they affect the nervous system, as many viruses tend to accumulate there with age.

In a new study, researchers analyzed brains provided by the Rush Alzheimer's Research Center (RADC) in Chicago, from 10 autopsied Parkinson's patients and 14 non-PD patients. They found traces of human pegylated virus (HPgV) in five Parkinson's brains, but none in healthy brains. Human pegylated virus is a virus related to hepatitis C. The virus has also been detected in the cerebrospinal fluid of Parkinson's disease patients, but not in the control group. However, the small sample size makes these results inconclusive.

Next, perhaps seeking stronger evidence, the researchers analyzed blood samples from 1,393 participants in the Parkinson's Progression Markers Initiative, a biological sample bank for Parkinson's disease research. Only about 1% of Parkinson's patients had HPgV in their blood, which is consistent with the infection rate in the general population.

Nevertheless, the scientists say that people infected with the virus exhibited different immune signals, particularly those with a mutation in the LRRK2 gene. They explained that since mutations in the LRRK2 gene are known to influence immune signaling, autophagy, and viral processing, these genotype-specific responses suggest that host genetics and viral interactions could influence immune responses to human pegivirus, promoting neuroinflammation and the development of Parkinson's disease.

At this point, I'm confused; they performed two experiments, one not significant and the other negative, but in the article, the scientists still suggest a possible link between Parkinson's disease and human pegivirus. However, the headline and abstract (which will likely be the only parts colleagues read) are much less definitive: they only suggest that human pegivirus alters the transcriptomic profiles of patients with Parkinson's disease.

This is a far cry from the headlines in the trade press: "Harmless virus might trigger Parkinson's disease."

A growing number of studies are exploring the role of perineuronal nets (PNNs) in Parkinson’s disease. PNNs are specialized structures in the brain’s extracellular matrix that limit synaptic plasticity, like a stabilizing "glue" between neurons. While this stabilization helps maintain brain function in adulthood, it also reduces the brain’s ability to rewire or adapt, which is important for learning and recovery after injury. PNNs are especially important during brain development. They help close the "critical period" of heightened plasticity in childhood. Interestingly, while PNNs are degraded in adults, this plasticity can be partially restored. For example, PNN removal can promote recovery in stroke models

PNNs serve multiple functions: protecting neurons from oxidative stress and harmful molecules, regulating plasticity, and helping stabilize long-term memories. Abnormal changes in PNNs have been observed in aging and various neurological conditions, making them a promising target for therapeutic intervention.

In the cerebral cortex, PNNs are primarily found around inhibitory interneurons, particularly those that produce the protein parvalbumin. These interneurons help maintain a balance between excitatory and inhibitory signals in the brain. In Parkinson’s disease, where dopamine-producing neurons in the substantia nigra are lost, this balance is disrupted—suggesting that changes in PNNs may contribute to the disorder.

A recent study investigated the effects of temporarily reducing PNNs in the primary motor cortex (M1) of healthy adult mice using chondroitinase ABC (ChABC). This intervention caused temporary impairments in motor function, suggesting that PNNs contribute to motor stability. Using ChABC makes sense as perineuronal nets are composed of chondroitin sulfate proteoglycans, and ChABC is an enzyme that digests them. Chondroitinase treatment has been shown to allow adults' vision to be restored; moreover, there is some evidence that Chondroitinase could be used for the treatment of spinal injuries.

The researchers created a Parkinson’s disease mouse model by damaging one side of the midbrain of mice with 6-hydroxydopamine (6-OHDA). Two weeks after the lesion, PNN levels dropped in both hemispheres of the motor cortex but returned to normal within five weeks.

The researchers then applied ChABC to reduce PNNs again in the motor cortex and paired this with motor training. This combination improved motor recovery slightly in the Parkinsonian mice. The improvement was linked to an increase in parvalbumin interneurons surrounded by PNNs and a normalization of excitatory signals at their cell bodies. These findings suggest that PNNs respond dynamically—first to the injury and later to therapeutic intervention—and that manipulating them could help restore motor function.

In summary, perineuronal nets in the motor cortex appear to play a subtle but significant role in regulating movement. Modifying their structure could open new avenues for motor rehabilitation in Parkinson’s disease.

I am somewhat skeptical about the benefits of acupuncture beyond the placebo effect. However, this article suggests it could be beneficial for individuals with Parkinson's disease.
This study investigates whether acupuncture has any impact on long-term health outcomes, such as mortality, disease progression, or complications, in individuals newly diagnosed with Parkinson’s disease (PD) in South Korea. Indeed, it is unclear what constitutes an effective acupuncture session for Parkinson's disease, and individuals interested should receive at least one session every two months. It is commonly believed that in PD, reduced mobility due to tremors, postural imbalance, and rigidity likely contributes to poor circulation and decreased gastrointestinal motility, leading to bowel obstructions and impaired swallowing, which can, in turn, result in recurrent aspiration pneumonia. The benefits may arise from the fact that people with PD might find it easier to move.

As is often the case, the Korean authors used a database to select patients and gather information about their health over the following years. They did not see any patients in person; it was purely a matter of processing numbers in the database. The NHIS database contains extensive patient information, including diagnostic codes, healthcare utilization, prescriptions, vital signs, disability grades, sex, age, and socioeconomic factors such as health insurance. However, this database does not provide precise medical details.

Who Was Included in the Study?

The study focused on adults (age 19 or older) newly diagnosed with idiopathic Parkinson’s disease (IPD) between 2012 and 2016.

To ensure a cleaner analysis, the authors excluded individuals diagnosed with Parkinson’s before 2012 or who were disabled at the time of diagnosis. They also excluded patients diagnosed with dementia or who died within a year of their diagnosis. To strengthen the results, they excluded patients who had received acupuncture in the six months preceding their diagnosis.

After applying these exclusions, about 41,000 patients remained. From this group, the researchers employed "propensity score matching" to pair patients from both groups who had similar health and demographic profiles (like age, sex, income, location, and other health conditions). After further exclusions and matching, 6,394 patients remained in each group.

The researchers measured all-cause mortality (death from any cause), tracked for up to six years following diagnosis. They also recorded fractures (such as hip or spine fractures), emergency room visits, and deep brain stimulation (DBS) surgery. These outcomes serve as proxies to assess whether acupuncture is associated with slower disease progression or fewer complications.

How Were Acupuncture and Non-Acupuncture Groups Defined?

Acupuncture was counted only if a person had six or more sessions in the year following diagnosis. This cutoff is based on previous studies suggesting that at least six sessions may be necessary to observe an effect. The acupuncture types included conventional methods as well as specialized forms like ocular or electroacupuncture.

Results:

Overall, the mortality in the acupuncture group (960) was significantly lower than in the control group (1,118). Deaths due to neoplasms (174 vs 234), circulatory (172 vs 208), and digestive diseases were also lower in the acupuncture group. In all cases, the acupuncture group had equal or better results than the control group. However, no significant differences were observed between groups in fracture risk, emergency room visits, or DBS procedures.

Key Takeaways

The study aimed to explore whether acupuncture might influence survival or disease progression in Parkinson’s disease. To achieve this, it utilized a large, high-quality national health dataset and careful matching to compare similar patients. While the text does not cover the results themselves (e.g., whether acupuncture improved survival or reduced complications), the methods ensure that any differences found are likely due to the treatment rather than other health or demographic factors.

At Padiracinnovation, we face a paradox: despite the deluge of scientific publications on ALS, Parkinson's, and Alzheimer's disease, academic and industry scientists seem unable to develop effective drugs.

In fact, the incentive for academic scientists appears to be publishing a large volume of work, as this is nearly their only path to career advancement. Industry scientists publish infrequently, but their work primarily serves to promote their company to potential investors.

How can we differentiate the wheat from the chaff? There are several telltale signs:

• single author

• authors who publish more than two papers per year

• publishing outside of prominent journals, such as in conferences

• articles based on specific intuition without testing other hypotheses

• articles concerning a specific drug without explaining the rationale for its initial selection

• articles making bold claims based on queries to a public database without conducting further research to validate the results and without considering confounding factors

• articles making outrageous claims, such as "breakthrough in disease X."

In our field, additional indicators exist: a strong publication should report results from human trials, as animal models often prove to be completely ineffective.

Consequently, we find few publications to discuss, even as popular science news organisations report daily on significant advances made toward new drugs.

Abnormal protein aggregation within cells is a recurring phenomenon in Parkinson's disease (PD), Alzheimer's disease (AD), and amyotrophic lateral sclerosis (ALS). Current approaches use antibodies to target these aggregates, but this is a rudimentary approach, as little is known about the causes of their formation, or whether they are the cause or consequence of the disease.

Cells are an incredibly crowded environment, and their molecules undergo Brownian motion, which thwarts their biological function. Making the cell less dense and more soluble would certainly alleviate some molecular problems. There are various approaches, including those that use phase transitions.

Recent research sheds surprising light on the dynamic relationship between mitochondrial activity, ATP levels, and neuronal cytoplasmic fluidity, all of which play a critical role in controlling protein aggregation.

The researchers used mouse giant goblet cell cultures to analyze presynaptic viscosity using real-time confocal microscopy. These cells are characterized by large glutamatergic nerve terminals, ideally suited for real-time imaging. Rather than focusing on individual proteins, the team took a holistic approach, using a technique called fluorescence recovery after photobleaching (FRAP) of soluble green fluorescent protein (cGFP) to assess the overall viscosity of the axonal cytosol.

Cytosolic viscosity can reflect the extent of protein aggregation; Greater aggregation means less free diffusion of cGFP, indicating a more "solidified" cytosol.

Synapses are hotspots for mitochondria, which provide the ATP needed for neurotransmission. By labeling active mitochondria and comparing their location to cGFP mobility, the study revealed that regions with greater mitochondrial activity exhibited higher cytosolic fluidity. This suggests a direct link between ATP production and the maintenance of a more soluble and functional presynaptic environment.

To further investigate this, the team inhibited mitochondrial function using FCCP and other mitochondrial blockers. As ATP production decreased, cGFP diffusion decreased sharply, suggesting that the cytosol was becoming more viscous due to protein aggregation. It is important to note that this effect was specific to mitochondrial inhibition: blocking glycolysis had little effect.

Even components of the synaptic release mechanism, such as synaptic vesicles (SVs) and active zones (AZs), exhibited reduced mobility under mitochondrial stress, reinforcing the idea that energy depletion disrupts the fluid phase of the cytoplasm.

To test whether ATP could restore the altered cytosol state, the researchers administered ATP directly to neurons. They found that ATP not only restored cGFP diffusion but also reduced the size and number of protein aggregates. To test whether enhancing endogenous ATP production could mitigate the protein aggregation linked to mitochondrial dysfunction, the researchers turned to NMN, a molecule known to boost NAD⁺ levels and support mitochondrial health.

They treated neurons with NMN and observed the following key outcomes:

Partial restoration of cytoplasmic fluidity: In neurons with compromised mitochondrial activity (such as those derived from PARK2 or TDP-43 mutant patients), NMN treatment significantly improved the diffusion of soluble proteins like cGFP. While not as dramatic as direct ATP infusion, NMN nonetheless reduced cytosolic viscosity.

Reduction in aggregate burden: In both mouse neurons under mitochondrial stress and hiPSC-derived human neurons from neurodegenerative disease patients, NMN treatment lowered the accumulation of insoluble protein aggregates.

Improved ATP levels: NMN supplementation helped increase intracellular ATP concentrations, presumably by enhancing mitochondrial NAD⁺-dependent enzymatic activity, which supports oxidative phosphorylation.

These results suggest that NMN supports the same protective pathway as ATP, but indirectly, by restoring mitochondrial capacity to generate ATP and maintain a more fluid intracellular environment.

The mechanism appears to be biophysical rather than biochemical: ATP acts as a hydrotrope, a molecule that keeps other proteins dissolved and prevents them from forming aggregates.

The researchers then examined whether this principle held true for specific proteins involved in neurodegenerative diseases, including:

• α-synuclein (mutant SNCA and SNCA-A53T), PARK2 – Parkinson's disease

• APP, Amyloid, Tau – Alzheimer's disease

• TDP-43 – ALS

These purified proteins were able to undergo liquid-liquid phase separation (LPS) and form condensates in vitro. ATP was able to dissolve many of these condensates in a concentration-dependent manner, although mutant or misfolded versions (e.g., SNCA-A53T) required higher ATP concentrations to dissolve.

When the aggregates were left to incubate for longer, some (notably SNCA-A53T) began to form protofibrils, elongated, fibril-like structures similar to those observed in real-life pathology. Here again, ATP could reverse this phenomenon, but with reduced efficiency.

Even under crowded conditions (mirrored by the addition of PEG), ATP retained some ability to prevent or dissolve aggregates, although the effect was less potent.

The team then studied neurons derived from Human induced pluripotent stem cells (hiPSCs) from patients with Parkinson's disease (PARK2 mutation) and ALS (TDP-43 mutation). These neurons exhibited reduced cytosolic fluidity, lower ATP levels, and greater protein aggregation than healthy controls.

This supports the idea that ATP deficiency and mitochondrial dysfunction contribute to the condensation of pathogenic proteins in human neurodegenerative diseases.

Implications for Drug Development

This research redefines our approach to therapeutic targets in neurodegenerative diseases. Instead of seeking to eliminate aggregates after their formation, we could:

• Target mitochondrial function to preserve ATP production at synapses.

• Use small molecules that mimic the hydrotropic effects of ATP to maintain cytoplasmic fluidity.

• Develop drugs that prevent the formation of LPS (lipoproteinases) of key disease proteins by improving their solubility.

ATP itself is not a drug molecule in the traditional sense, but these results open new avenues for small molecules capable of acting like ATP to maintain protein solubility or prevent aggregate formation at an early stage.

Conclusion

Neurodegenerative diseases are often viewed from a genetic or protein perspective, but this study provides a biophysical perspective: the physical state of the cytosol itself is crucial. If cells cannot maintain a fluid and soluble environment, primarily due to energy deficiency, aggregation may become inevitable.

This is not just about treating symptoms or even eliminating aggregates afterward. It is about preserving the cellular environment so that neurons can withstand stress and maintain their function. As the field continues to explore how biophysical properties such as viscosity, solubility, and phase separation interact with disease, the role of ATP may prove central, not only as a fuel, but also as a key regulator of neuronal health.