TMAO and heavy hydrogen

A 2026 review subject was possible involvement of deuterium in TMAO levels, which contrasted with the usual TMAO meme. I’ll curate this paper through an outline of its sections:

“The human gut microbiome plays many essential roles, but an often-overlooked role is to maintain an abundant supply of deuterium depleted (deupleted) nutrients to fuel the host mitochondria. Excess deuterium (heavy hydrogen) damages mitochondrial ATP synthase nanomotors, leading to a decrease in matrix water production with increased reactive oxygen species (ROS) and inefficient ATP production. A microbial metabolite, trimethylamine N-oxide (TMAO) is a powerful signaling molecule whose plasma levels are high in association with many chronic diseases.

In this paper, we present a hypothesis that TMAO is a marker for deuterium overload in the methylation pathway, in addition to its role as an indicator of a disrupted gut microbiome. The original study that brought attention to TMAO involved feeding mice synthetic choline with fully deuterated methyl groups. Fully deuterated TMAO was subsequently detected in the plasma. By contrast, a diet rich in eggs, a natural source of choline (a precursor to TMAO), does not raise TMAO levels. Many of the pathologies that are linked to elevated TMAO can also be viewed as strategies to promote the supply of deupleted water to the mitochondria, systemically.

1. Introduction – DNA and histone methylation regulate epigenetic modifications and imprinting. Phosphatidylcholine is a precursor to acetylcholine, an excitatory neurotransmitter in the brain. Trimethylated lysine molecules, recovered during protein metabolism, are precursors to L-carnitine, which facilitates the transport of fatty acids into the mitochondria to be oxidized for fuel production. Dietary phosphatidylcholine and dietary L-carnitine, in addition to endogenous sources, are important nutrients that provide methyl groups to the methylation cycle. Choline and L-carnitine, as well as a close relative, betaine (trimethylglycine), are all precursors to trimethylamine (TMA), a small methylated amine produced through microbial enzymatic action. Obligate anaerobic hydrogen-dependent archaea called methylotrophs in the gut can reduce the three methyl groups in TMA to methane gas.

2. Evidence that deuterium disrupts mitochondrial function – Deuterium (2H) is a heavy isotope of protium (1H; hydrogen), and it is pervasive in nature, found in seawater at a concentration of 155 parts per million (ppm) relative to protium. Deuterium is highly damaging to the F1F0-ATP synthase (ATPase) nanomotors in the mitochondria that produce ATP, the primary fuel source of the cell. Deuterium loading suppresses the activity of many fundamental biologically important hydrolytic enzymes that depend on proton tunneling. It is likely that deuterium increases the frequency of unrepaired nuclear DNA mutations, by suppressing the activity of deuterium-sensitive repair enzymes.

The inherent collective proton tunneling (ICPT) process, which uses membrane-bound ATPase nanomotors in living organisms, is nature’s ultimate tool for discriminating hydrogen isotopes. This is because a deuteron (2H) cannot replace a proton (1H) in its tunnel protein during enzymatic transmembrane transport due to its doubled mass and twice larger atomic nuclear size. The result is large compartmental, inter- and intramolecular deuterium disequilibrium in 2H/1H ratios in all biomolecules, which readily distinguishes respiration from aerobic fermentation with adaptive significance. Deuterons irreversibly clog single proton tunneling ATP synthase nanomotors in mitochondria, resulting in the complete breakdown of ICPT.

This initiates many disease-causing molecular crowding mechanisms, which we review herein from the perspective of prokaryotic proton pumping and H2 gas formation in the organic molecular realm of mitochondrial proton-donating substrates. Understanding at the systems level how humans protect mitochondrial ICPT processes and ATP synthase is a fascinating journey reviewed herein. We hypothesize that the process uses TMAO as the microbial stepping stone, employing deuterium discrimination to become an active player in forming the biological reaction.

2.1 Quantum tunneling and proton-coupled electron transport – ICPT is a theoretical quantum mechanical phenomenon proposing that a single proton spontaneously passes through a potential energy barrier, typically within a hydrogen bond, in a manner that can be functionally irreversible. Unlike classical particles that must surmount energy barriers, protons can ‘tunnel’ through them due to their wave-like nature. Often, this single proton tunneling is part of a larger process where a proton and an electron are transferred simultaneously (or sequentially) as a single kinetic step, often in the presence of strong electric fields that stabilize the transferred state. This process is referred to as ‘proton-coupled electron transport’ (PCET).

Mitochondria exploit PCET to build the proton gradient that powers the nanomotors to produce ATP in the electron transport chain (ETC). Many enzymes, such as dehydrogenases and lipoxygenases, exploit proton tunneling to carry out their reactions. Deuterons are much less capable of tunneling, so this becomes a way to select for substrates containing protons rather than deuterons. The very large kinetic isotope effect (KIE) for soybean lipoxygenase is an example of this phenomenon.

NADH-ubiquinone oxidoreductase (Complex I) couples the transfer of two electrons between NADH and ubiquinone to the translocation of four protons across the membrane. This process provides the driving force for ATP synthase, which harnesses the gradient to produce ATP, but it also assures that few, or no, deuterons arrive on the other (intermembrane) side of the membrane, protecting the ATPase nanomotors.

SAMe, the universal methyl donor, plays a crucial role in regulating oxidative phosphorylation (OXPHOS). It is primarily synthesized in the cytoplasm and imported into the mitochondria via the import protein SAMC. SAMC is the only mitoSAM carrier and is required for OXPHOS and oxidative tricarboxylic acid (TCA) metabolism, showing a strong dependency of mitochondrial health on one-carbon (1C) metabolism. We hypothesize that the importance of SAMe to mitochondrial health is directly linked to the plausible theory that SAMe’s methyl groups are normally highly deupleted.

3. Are microbially synthesized methyl groups and butyrate deuterium depleted? – A careful tracing of multiple metabolic processes taking place in a human cell reveals that they are plausibly designed to greatly restrict the number of deuterons that are in the mitochondrial water. This strategy helps to minimize exposure of the ATPase nanomotors to deuterons. In part, this feat is accomplished through enzymes such as flavoproteins that greatly favor protium over deuterium in their reaction, i.e., that have a high deuterium KIE. The physics usually involves configuring the enzyme to support proton tunneling, since deuterons are much less capable of such tunneling.

Another way to support a reduced deuterium supply to the ATPase nanomotors is to select nutrients that are naturally low in deuterium to feed into the tricarboxylic acid (TCA) cycle. This is what makes the metabolites produced by the gut microbes via hydrogen recycling very significant. The enzyme expressed by anaerobic archaea that metabolize TMA, TMADH, is a flavoprotein with a high deuterium KIE (~ 8.6) due to vibrationally assisted hydrogen tunneling.

This means that the hydrogen recycling that takes place during its metabolism further scrubs deuterium from methylation pathways, while TMA that is left behind becomes enriched in deuterium. This unmetabolized TMA is converted to deuterium-enriched trimethylamine N-oxide (TMAO) in the liver and released into the circulation. Elevated TMAO levels in plasma are associated with increased risk to cardiovascular disease and a long list of other inflammatory diseases, as we will detail in the coming sections.

4. Natural and synthetic choline have different effects on TMAO levels: does deuterium play a role? – The original 2011 paper that first identified TMAO as a risk factor for heart disease, involved feeding mice phosphatidylcholine where all the protons in the methyl groups attached to the nitrogen atom were replaced with deuterium, so that the researchers could trace the products of the nutrient in the body. It turns out that they accidentally conducted an experiment testing what happens when phosphatidylcholine is extremely enriched in deuterium. 🙂 They also determined that supplementation of mice with deuterated choline, TMAO, or betaine resulted in upregulation of multiple macrophage scavenger receptors linked to atherosclerosis. TMAO was not produced if the mice were pretreated with antibiotics or in experiments with germ-free mice, confirming that microbial enzymatic action was a necessary precondition.

A paper in 2021 on human subjects compared choline intake from natural dietary sources with supplemental choline bitartrate and found that the latter but not the former raised blood TMAO levels. Notably, these authors wrote in the conclusion of the abstract: ‘Despite high choline content in egg yolks, healthy participants consuming four eggs daily showed no significant increase in TMAO or platelet reactivity.’ However, TMAO levels rose significantly following synthetic choline bitartrate supplementation. This occurred even though the subjects had normal kidney function, showing that elevated TMAO is not just a consequence of kidney disease.

The authors of the original 2011 study had published a follow-on study on human subjects in 2013, in which they supplemented the subjects with D9-PC, essentially repeating the mouse study but with humans as the subjects. They confirmed that D9-TMAO levels were sharply elevated in the plasma and urine following supplementation. Furthermore, an elevated TMAO level predicted an increased risk of major cardiovascular events, after adjustment for traditional risk factors. This study shows that it may not be phosphatidylcholine vs. choline bitartrate that matters, but rather whether the choline is deuterium depleted or deuterium enriched. By contrast, a survey involving over 14,000 participants found that dietary choline protects from both heart disease and stroke. L-carnitine is also a precursor to TMAO, and a mouse study in which the mice were fed deuterated L-carnitine also showed a sharp increase in plasma TMAO following supplementation, further supporting the idea that deuteration is the primary factor promoting TMAO accumulation.

5. TMAO directs metabolism towards deupleting peroxisomal-mitochondrial crosstalk – The synthesis of TMAO from TMA requires hydrogen peroxide. The source of hydrogen peroxide is via peroxisomal fatty acid chain modifications that utilize molecular oxygen dissolved in plasma to produce SCFAs, ketones, NADH and hydrogen peroxide (H2O2). The resulting metabolic water of the reaction is deupleted, as fatty acids are inherently deupleted molecules in biology. Although peroxisomes do not produce ATP directly, they reduce NAD+ for proton delivery to mitochondria via membrane-based intracellular proton transporters. Although the partial contribution of TMAO synthesis to intermediary metabolism is yet to be determined to efficiently deplete deuterium, it is certain that none of the above works well in the deuterium preserving glucogenic metabolic state. There is a strict dependence of peroxisomes on long chain saturated fatty acid substrates with particularly lower deuterium-related chemical mass.

Peroxisomal metabolism triggered by TMAO turnover utilizes very long and branched chain fatty acids (A) as well as dissolved molecular oxygen (B) carried in plasma. Peroxisomes produce SCFAs via β-carbon oxidation, ketones, NADH (C) and hydrogen peroxide (B). H2O2 is rapidly converted to metabolic water by catalase (CAT) that also yields molecular oxygen for the mitochondrial matrix (D) as well as for other cellular compartments. Peroxisomes can also reduce NAD+ for proton delivery to mitochondria via membrane-based intracellular proton transporters. oxidation of very long chain saturated fatty acid β carbons, purportedly of animal source, with the help of molecular oxygen, yields the most deupleted H2O2 by weight. CAT, one of the fastest enzymes in biology with that of isomerases, rapidly and irreversibly produces water from H2O2, while recycling oxygen. Metabolic hydrogen peroxide of fatty acid breakdown with low deuterium consequently provides ATP synthase nanomotor-sparing protons for energy production. High TMAO with hydrogen peroxide turnover can easily depend on CAT-mediated oxygen recycling.

6. Is TMAO an indicator of deuterium overload in the mitochondria?
6.1 TMAO inhibits S-adenosylhomocysteine hydrolase
6.2 TMAO induces reactive oxygen species
6.3 TMAO suppresses autophagy via PI3K/Akt/mTOR activation
7. TMAO and human diseases
8. Does oxidative stress lead to mitochondrial deupletion? – While oxidative stress is a major contributor to cellular damage, the processes involved in resolving ROS are an essential part of the mechanism by which the cell reduces deuterium levels in the mitochondria. Intracellular ROS are derived mainly from NOX, xanthine oxidase, and the mitochondrial electron-transport chain (mETC). Excess mitochondrial deuterium promotes increased ROS generated by the mETC. Superoxide dismutase (SOD) converts ROS to H2O2 , which can release the highly destructive hydroxyl radical in the presence of reduced iron (Fe2+). However, H2O2 is an excellent source of deuterium depleted water (DDW) in the mitochondria, as long as there is sufficient mitochondrial glutathione and both glutathione peroxidase and glutathione reductase are adequately expressed. H2O2 freely crosses the mitochondrial membrane, and, with adequate antioxidant support, it is rapidly converted to two molecules of DDW, catalyzed by glutathione peroxidase.

9. Do lipid-laden foam cells support mitochondrial deupletion?
10. Archaeobiotics
11. A crucial role for A. muciniphila
12. Strategies to lower TMAO levels – It is clear that elevated plasma TMAO is a risk factor for a broad range of chronic diseases, and therefore it is compelling that a strategy that reduces plasma TMAO should show health benefits. However, simply avoiding foods that provide precursors to TMAO is not likely to be productive. Choline, L-carnitine, and betaine are the primary sources that fuel the methylation pathway. Eggs and seafood, rich sources of these nutrients, also contain many valuable micronutrients and healthy fats that are also essential.

Deuterium depleted water (DDW) is commercially available, at dilution levels as low as 5 ppm. It can be mixed with tap water to simulate natural glacier water, typically containing around 100 ppm deuterium. Although the number of studies on the effects of therapeutic deuterium depletion on various health conditions is small, a review paper found that deuterium depletion has shown promise in preventing and treating cancer, improving long-term memory, enhancing sports performance, and reducing symptoms of depression.

It is apparent that the best way to reduce TMAO levels, while simultaneously boosting methylation supplies, is to promote an abundant colonization of anaerobic archaea in the gut, so that they can clear (fully metabolize) the TMA before it has a chance to become TMAO.

13. Discussion – In this paper, we develop the argument that TMAO serves as a marker for excess deuterium in the methylation pathway, and, by extension, in the mitochondria, systemically. While methyl groups have powerful epigenetic effects, the ultimate fate of methyl groups is their metabolism to CO2 and water that is most likely deuterium depleted in the mitochondria. A microbial imbalance leading to reduced colonization by beneficial bacteria and an overgrowth of pathogenic species is the primary cause of overproduction of TMAO.

14. Conclusions – We have shown that TMAO, a causal factor for many diseases, may act as a marker for gut dysbiosis and for excess deuterium load in mitochondria, systemically. We traced through many of the biological pathways involving 1C metabolism and showed the integral role that gut bacteria play in stripping deuterium from methyl groups.”

https://link.springer.com/article/10.1007/s11306-026-02443-3 “The essential role of hydrogen gas recycling by gut microbes in reducing deuterium load in host mitochondria: is trimethylamine oxide a deuterium sensor?”


I take Now brand taurine, acetyl-L-carnitine, flush-free niacin, and betaine. I asked them whether these products are evaluated for their deuterium content. Will update with their response.

Year Six of Changing to a youthful phenotype with sprouts

1. I’ve continued daily practices from Year Five to experience another year without being sick (if I don’t count getting MSG poisoning from Chinese food.) I consequently scheduled a doctor visit next week to get a sumatriptan prescription refilled.

2. Two modifications to what’s mentioned in 2026 diet and supplement changes:

– In that post’s comments, Ole Bisgaard Pedersen asked if I took NAD+ supplements such as NAM, NMN or NR – forms of vitamin B3 that are precursors to NAD+. I didn’t note that last year I started taking Now brand Flush-free niacin 500 mg mid-morning.

Nicotinamide riboside has the most human evidence, including a Tru Niagen clinical trial that showed improvements in peripheral artery disease. It’s too expensive for long-term use, though.

Even if I could afford it, there isn’t a magic bullet for fixing vascular system dysfunction. PAD is just one symptom of a cardiovascular system that needs to be overhauled then maintained at a healthy level. There is no clinical trial that has a logical therapeutic end point to stop treatments where a person could say, “I’ve done enough for my vascular system, my physical and cognitive functions won’t backslide.”

– 2-3 years ago, I changed from microwaving broccoli sprouts in a plastic bag to microwaving them in a small bowl with a small plate covering it to keep them from popcorning out of the bowl. I use a 1000W microwave oven on 80% power for ten seconds.

3. The two vitamin C macaque studies I’ve recently curated both ran for a human equivalent of ten years. I was encouraged that both found Nrf2 activation to be part of their causal beneficial evidence, since vitamin C wasn’t on my radar as a Nrf2 activator.

I expect that in four years I’ll write a Year Ten post on eating microwaved broccoli sprouts. I haven’t seen human evidence for broccoli extracts that bypass small intestine absorption and metabolism per Glucosinolate and isothiocyanate human interventions, or enteric capsules, or nanoformulations as suitable substitutes. Maybe studies on broccoli sprout powder or a Nrf2 activator that tops sulforaphane will be published before then, who knows.

Eat broccoli sprouts for ALS?

A 2026 rodent study investigated sulforaphane’s ability to affect ALS-like symptoms:

“The objective of this study was to evaluate neuroprotective efficacy and safety of sulforaphane (SUFP) in a methylmercury (MMHg⁺)-induced preclinical rat model of amyotrophic lateral sclerosis (ALS). ALS is characterized by progressive motor neuron degeneration and muscle wasting, leading to impairments in gait, swallowing, salivation, and routine motor activities.

64 animals were classified into eight groups: 1st: normal control, 2nd: vehicle control; 3rd: SUFP perse (4 mg/kg, i.p.), 4th: MMHg + (5 mg/kg, p.o.), 5th: MMHg + 5 + SUFP (2 mg/kg, i.p.), 6th: MMHg+ 5 + SUFP (4 mg/kg, i.p.), 7th: MMHg+ 5 + omaveloxolone (OVX) (30 mg/kg, i.p.), and 8th: MMHg + 5 + dimethyl fumarate (DIMT) (50 mg/kg, i.p.). Neurotoxin MMHg + was orally administered at 5 mg/kg for the first 21 days. For the next 22 days, SUFP, OVX, and DIMT were administered intraperitoneally (i.p.).

SUFP modulates neurotransmitter levels such as acetylcholine (A), dopamine (B), GABA (C), glutamate (D), and serotonin (E).

SUFP4 exerted broad neuroprotective effects in ALS pathology by restoring antioxidant proteins (Nrf2, HO-1, SIRT1), suppressing apoptotic (Bax, caspase-3, Bcl-2) and inflammatory markers (TNF-α, IL-1β), and enhancing the anti-inflammatory cytokine IL-10. It also downregulated stress-related signaling pathways (PI3K/Akt, p75NTRECD, MAPKs) associated with neurodegeneration. These molecular effects translated into meaningful functional recovery, as evidenced by improvements in grip strength, locomotor performance, spatial memory, and depressive-like behavior.

Histopathological evaluation demonstrated attenuation of demyelination and preservation of neuronal architecture including the cerebral cortex, hippocampus, striatum, midbrain, and cerebellum. Beyond central neuroprotection, SUFP exerted systemic benefits by normalizing hepatic enzymes, improving skeletal muscle integrity, restoring redox balance, stabilizing neurofilament and myelin-associated proteins, and correcting hematological alterations.

Despite limitations related to study duration and animal sex, this work strongly positions SUFP as a promising, multi-target therapeutic candidate for ALS with both neural and systemic protective efficacy.”

https://link.springer.com/article/10.1007/s12035-026-05683-5 “Sulforaphane-Mediated Multitarget Therapeutic Effects in Methylmercury-Induced ALS-Like Pathology: Comparative Analysis and Multifaceted Approach to Neuroprotection and Systemic Recovery” (not freely available) Thanks to Dr. Sidharth Mehan for providing a copy.


Unlike A Nrf2 treatment for ALS?, this study didn’t present evidence that its treatment compound was effective for preventing ALS. For one thing, currently-known disease factors involving heat shock proteins and associated genes, some of which are Nrf2 targets, weren’t investigated.

Two Nrf2 activators were used in both studies as comparators of Nrf2 activation effects. Neither omaveloxolone nor dimethyl fumarate are ALS causal treatments, though, and have undesirable side effects.

A human equivalent of this study’s higher sulforaphane dose is ((4 mg x .162) x 70 kg) = 45 mg. 45 mg of sulforaphane might be too much to consistently take at one time because of unpalatability. But I documented taking an estimated 52 mg for a year during 2020-2021 by eating microwaved 3-day-old broccoli sprouts twice a day.


Nrf2 and stem cells

A 2026 review subject was mechanisms and therapeutic potential for Nrf2 activators in combination with mesenchymal stem cells:

“Mesenchymal stromal/stem cells (MSCs) are multipotent stem cells that can be isolated from various tissues – such as bone marrow (BM), umbilical cord (UC), adipose tissue (AD), dental pulp (DP), hair follicle (HF), and placenta – and differentiated into multiple lineages under appropriate conditions. Their functional repertoire includes immunomodulation, homing, and differentiation, which collectively help establish a balanced inflammatory and regenerative niche within damaged tissues during severe inflammation. MSCs-derived extracellular vesicles (MSCs-EVs) and conditioned medium (MSCs-CM) play remarkable roles, exhibiting potent anti-inflammatory and antioxidant properties that offer novel therapeutic alternatives for inflammatory diseases.

Therapeutic capacity of MSCs in inflammatory conditions is increasingly attributed to their potent paracrine activity rather than solely to their differentiation potential. A key mechanism underlying this paracrine effect is activation of the Nrf2 antioxidant pathway.

MSCs and their secreted products including exosomes (Exos), EVs, and CM, activate Nrf2 through multi-dimensional/target mechanisms, thereby enhancing cellular antioxidant defenses, modulating immune responses, and promoting tissue repair. It is noteworthy that therapeutic efficacy of MSCs and their derivatives can be enhanced through external modulation, including pretreatment with natural compounds.

Preconditioning refers to brief treatment of MSCs or their derivatives with physical, chemical, or biological factors prior to application, aiming to enhance their ability to counteract oxidative stress and improve their therapeutic efficacy. Flavonoids precondition and prime MSCs via the direct Keap1-Nrf2 pathway or indirect PI3K-Akt pathway, which enhances cellular resilience to adverse conditions by reducing apoptosis and promoting survival. Primed MSCs, in turn, remodel the microenvironment through an altered secretory profile, releasing bioactive factors that create more favorable conditions for their own persistence.

The core logic of these strategies lies in simulating or inducing adaptive stress, such as employing specific chemical molecules or drug stimuli, or utilizing physical / microenvironmental preconditioning to mimic specific physical conditions of the in vivo injury environment. The most straightforward strategy is overexpression of Nrf2 or its key downstream effector molecules.

The majority of existing studies remain at the level of observing correlations with Nrf2 upregulation, and there is still a lack of precise causal validation regarding key upstream signals – such as specific cytokines, miRNAs, or proteins – through which MSCs or derivatives initiate Nrf2 activation. Mechanistic insights are predominantly derived from in vivo or rodent (mouse/rat) model experiments, with a notable absence of clinical validation, insufficient long-term safety and pharmacokinetic data, and a lack of standardization in administration routes and dosages, all of which hinder clinical translation.

The essential role of the Nrf2 pathway has not been rigorously confirmed, as most studies have not employed reverse genetic validation using Nrf2-knockout animals or specific inhibitors. Consequently, it remains unclear whether therapeutic effects are necessarily and exclusively dependent on Nrf2, and potential synergistic contributions from other pathways may have been overlooked.

Most natural flavonoids face challenges such as low oral bioavailability, rapid metabolism, and poor targeting. Numerous challenges remain to be addressed in order to translate these promising preclinical findings into clinical practice. Future research should focus on the following aspects:

  1. Elucidating precise upstream molecular mechanisms by which MSCs activate Nrf2;
  2. Employing more clinically relevant chronic disorder models;
  3. Systematically evaluating long-term safety, optimal delivery strategies (including dosage and route of administration), and immunogenicity of MSCs-based therapies;
  4. Validating selection criteria (optimal source), quality control, batch-to-batch consistency of MSCs, and addressing regulatory and ethical barriers to clinical translation; and
  5. Integrating molecular docking, ADMET (Absorption, Distribution, Metabolism, Excretion, Toxicity) prediction, and in vitro and in vivo validation to further elucidate regulatory effects of flavonoids and enhance understanding of their mechanisms of action.”

https://link.springer.com/article/10.1186/s13287-026-04925-6 “Activation of Nrf2 with natural flavonoids and mesenchymal stromal/stem cells: mechanisms and therapeutic potential for inflammatory diseases” (click pdf)


This paper was overly long at 127 pages, so I focused on the later sections. None of these treatments are currently ready for clinical trials.

I also didn’t mention specific flavonoids as Nrf2 activators. It’s beyond a reviewer’s task to rank Nrf2 activators, and a study’s researchers seldom address why they used a poorly-activating flavonoid instead of a higher-ranked natural plant compound such as sulforaphane.

No hero will be rescuing your and your children’s neurodegeneration for you

Starting this blog’s twelfth year by curating a poorly-done 2026 review of Nrf2 and its capability to change a person’s development of Parkinson’s disease. I’ll emphasize precedent conditions that if not effectively dealt with in youth, can’t prevent PD from occurring at some later life stage.

“This review explicitly examines how age-associated decline in NRF2 responsiveness intersects with redox imbalance, mitochondrial dysfunction, proteostatic failure, and neuroinflammation, core mechanisms shared between aging and PD. PD unfolds through a complex interplay of cellular stress and immune responses. Oxidative stress, mitochondrial dysfunction, and chronic neuroinflammation converge to damage dopaminergic neurons, with microglia playing a central role in amplifying this injury.

NRF2 emerges as a key regulator of antioxidant defenses, inflammatory balance, and mitochondrial protection, offering a promising target for clinical intervention. NRF2 activity favors the anti-inflammatory microglial over the pro-inflammatory phenotype. Decline in NRF2 inducibility with age impairs microglial clearance, promotes neuroinflammation, and reduces antioxidant defenses, while NRF2 activation restores protective functions and offers a promising therapeutic target.

Strategies aimed at restoring or enhancing NRF2 activity hold significant promise as disease-modifying interventions, not only to slow PD progression but also to promote resilience against the broader spectrum of age-associated neurodegenerative and inflammatory conditions.”

https://www.sciencedirect.com/science/article/pii/S0891584926000316 “NRF2 AT THE CROSSROADS OF PARKINSON’S DISEASE AND AGING: MECHANISTIC INSIGHTS AND TRANSLATIONAL PERSPECTIVES”


This review only gave lip service to PD progression outside of the brain, as if the importance of prodromal factors to a person’s neurodegeneration such as dysfunction in gut, eyes, skin, and olfactory systems can be minimized. But failure to recognize early what will doom a person to be unable to recover health in later decades is disingenuous. Since these reviewers omitted early interventions into PD prodromal factors, the best they came up with was interventions to “slow PD progression.”

Maybe these reviewers felt it would be outside the scope of this review to discuss early non-brain PD factors for more than one sentence? However, while PD is defined by striatal brain neurons, Nrf2 activity is much less in brain and central nervous system neurons than elsewhere in the body per Nrf2 Week #2: Neurons.

I disagree with these reviewers’ self-imposed emphasis on aging. Repeating ‘age-associated’ numerous times seemed as if they wanted to influence the reader into thinking age in and of itself was a cause for PD, rather than an imputed mathematical correlation. Their emphasis led to dumb mentions such as senolytics for no apparent reason than senescence is a ‘hallmark of aging’, and to meaningless ‘diseasome of aging’ characterizations, and to ignoring the existence of early non-age-associated PD diagnoses in 20- and 30-year-olds.

Whatever it takes to get published, I’d guess. Or maybe it’s that the number of omissions and useless points a review paper makes increases with the number of reviewers and their sponsors’ agendas.

For example, why was it permissible to dedicate lip service to ‘exposome’ factors like microplastics, environmental pollution, and viruses, but it’s still not permitted in 2026 to discuss research into the impacts on vascular disease and neurodegeneration of lipid nanoparticles and DNA contamination in what a large number of humans were exposed to by injected pharmaceuticals starting in late 2020? Not to mention two studies published in 2024 of over 2.5 million people whose incidences of neurologic issues, mild cognitive impairment, and Alzheimer’s disease rapidly increased after ‘vaccination’?

I’ve mentioned in this blog many times how it’s every human’s choice whether or not we take responsibility for our own one precious life. I suggest, if it’s not too late, do that for your children’s lives, too.

A Nrf2 / NAD+ connection?

Here are two 2025 papers, starting with a rodent study that investigated interactions between the Nrf2 and kynurenine pathways:

“Exposure to the tryptophan metabolite kynurenine and its electrophilic derivative kynurenine-carboxyketoalkene (Kyn-CKA) leads to an increase in the abundance of transcription factor Nrf2 and induction of Nrf2-target genes. The Keap1/Nrf2 system is the main orchestrator of cellular defence against environmental stress, most notably oxidative and inflammatory stress.

Nrf2 can be activated pharmacologically by small molecules, the majority of which are electrophiles and oxidants that modify specific cysteine-based sensors in Keap1. C151 in Keap1 is the target of the isothiocyanate sulforaphane, a classical Nrf2 activator that has been employed in ∼90 clinical trials, as well as for the two Nrf2 activators that are clinically in use: dimethyl fumarate, for relapsing remitting multiple sclerosis, and omaveloxolone, for Friedreich’s ataxia.

Kynurenine is an endogenous metabolite derived from the essential amino acid tryptophan. Kynurenine and its metabolites, such as the electrophilic kynurenine-carboxyketoalkene (Kyn-CKA), have been demonstrated to activate Nrf2 in other pathologies, including sickle cell disease, attenuating inflammation. Moreover, identification of the gene encoding the kynurenine-metabolising enzyme kynureninase as a gene transcriptionally upregulated by Nrf2, provides a plausible negative feedback regulatory mechanism.

Because kynurenine is not electrophilic, whereas its metabolite Kyn-CKA is, we considered the possibility that Kyn-CKA is the actual Nrf2 activator. Using biochemical and cell-based assays, we found that Kyn-CKA reacts with C151 in the BTB domain of Keap1 and increases the thermostability of Keap1, indicating target engagement. Consequently, Nrf2 accumulates and induces transcription of antioxidant/electrophile-responsive element (ARE/EpRE)-driven genes.

These findings demonstrate that Kyn-CKA targets C151 in Keap1 to derepress Nrf2, and reveal that Nrf2 is a main contributor to the anti-inflammatory activity of Kyn-CKA in macrophages.”

https://www.sciencedirect.com/science/article/pii/S2213231726000078 “The electrophilic metabolite of kynurenine, kynurenine-CKA, targets C151 in Keap1 to derepress Nrf2”


A review subject was targeting nicotinamide adenine dinucleotide, oxidized form (NAD+) for clinical use:

“Mammalian NAD+ biosynthesis includes four known pathways, primarily occurring in cytoplasm:

  • (a) the NRH pathway;
  • (b) the salvage pathway;
  • (c) the Preiss–Handler pathway; and
  • (d) the kynurenine pathway.

The de novo kynurenine pathway metabolizes tryptophan (Trp) to NAD+, producing various intermediates that serve as biomarkers for different diseases. These intermediates show alterations in various pathological conditions.

While kynurenine and its metabolic derivatives are intermediates in the de novo NAD+ biosynthesis pathway, these are also produced independently in various physiological contexts, particularly in immune cells, where they act as immunomodulatory compounds.”

https://www.nature.com/articles/s43587-025-00947-6 “Emerging strategies, applications and challenges of targeting NAD+ in the clinic” (not freely available) Thanks to Dr. Jianying Zhang for providing a copy.


This second paper above showed a graphic of the Nrf2 and kynurenine pathways together in a diagram showing relationships between NAD+ augmentation and the hallmarks of aging, but didn’t elaborate other than labeling their box Dysbiosis. So how these two pathways interact is better outlined in the first paper above with explaining how a kynurenine-metabolizing enzyme is one of the hundreds of Nrf2 target genes, creating a natural feedback loop between Nrf2 activation and the kynurenine pathway.

These reviewers also lumped SIRT1 in their Dysbiosis box, and into several other boxes, probably due to the penultimate coauthor’s influence:

However, repeating something over and over doesn’t make it scientifically valid regardless of the number of citations. Or, as a 2022 review Sirtuins are not conserved longevity genes concluded:

“A global pursuit of longevity phenotypes was driven by a mixture of framing bias, confirmation bias, and hype. Review articles that propagate these biases are so rampant that few investigators have considered how weak the case ever was for sirtuins as longevity genes.

Acknowledging that a few positive associations between sirtuins and longevity have been identified after thousands of person-years and billions of dollars of effort, we review the data and suggest rejection of the notions that sirtuins (i) have any specific connection to lifespan in animals and (ii) are primary mediators of the beneficial effects of NAD repletion.”

Plasmalogens Week #7 – Genes

Continuing Plasmalogens Week with three 2025 papers, starting with a rodent study of genetically deleting a plasmalogen catabolizing enzyme:

“In this study, we investigated the impact of global and tissue-specific loss-of-function of a plasmalogen catabolizing enzyme, lysoplasmalogenase (TMEM86B), on circulatory and tissue lipidomes. Mice with homozygous global inactivation of Tmem86b (Tmem86b KO mice) were viable and did not display any marked phenotypic abnormalities.

Tmem86b KO mice demonstrated significantly elevated levels of plasmalogens alkenyl phosphatidylethanolamine (PE(P)) and alkenyl phosphatidylcholine (PC(P)), as well as lysoplasmalogens, in the plasma, liver, and natural killer cells compared to their wild-type counterparts. The endogenous alkenyl chain composition of plasmalogens remained unaltered in Tmem86b KO mice. Consistent with the global knockout findings, hepatocyte-specific Tmem86b knockout mice also exhibited increased plasmalogen levels in the plasma and liver compared to their floxed control counterparts.

Plasmalogens may be synthesized locally within various tissues, with each organ possessing the necessary enzymatic machinery to regulate its own plasmalogen levels. Plasmalogens are important structural constituents of the biological membranes of animals and certain anaerobic bacteria, and have several well-described functions, including regulating membrane dynamics and vesicular cholesterol transport and homeostasis.

  • One of the most interesting features of plasmalogens is their endogenous antioxidant activity, which is mostly due to the vinyl ether bond, which can scavenge reactive oxygen species and thereby protect other biomolecules from oxidative damage.
  • They increase the gene expression of multiple antioxidant enzymes to protect against chemically induced cytotoxicity and lipid peroxidation in cultured hepatocytes.
  • Plasmalogen derivatives such as polyunsaturated fatty acids (AA or DHA) and lysoplasmalogens can act as lipid mediators for multiple cellular signaling activities.
  • Plasmalogens are important for phagocytosis of macrophages, lipid droplet formation, and development and function of neuromuscular junctions.
  • They play vital roles in mediating immune responses, and mitochondrial fission to regulate adipose tissue thermogenesis, and protecting neuronal cells against cell death and inflammation.

All of these are suggestive of a critical role played by plasmalogens in maintaining cellular homeostasis.

While plasmalogen anabolism is well defined, its catabolism has been less studied. During catabolism, plasmalogens are deacylated by the action of a calcium-independent phospholipase A2 enzyme (iPLA2) to produce lysoplasmalogens. However, cytochrome C has also been shown to act as a plasmalogenase under certain circumstances.

The amount of lysoplasmalogens in cells is tightly regulated either by reacylation into plasmalogens through a coenzyme A-independent transacylase, or by degradation into fatty aldehydes and glycerophospholipids by an alkenyl ether hydrolase commonly known as lysoplasmalogenase. Lysoplasmalogenase is a microsomal transmembrane enzyme highly specific for lysoplasmalogens, and has no activity against plasmalogens.

While research on the distinct biological functions of lysoplasmalogens and plasmalogens is lacking, some reports indicate potential toxic effects of lysoplasmalogens. Degradation products of lysoplasmalogens, such as fatty aldehydes, are highly reactive electrophilic compounds that can form toxic adducts with cellular proteins and lipids. These interactions can lead to cellular dysfunction and contribute to various pathological conditions. Their accumulation in ischemic/reperfused tissues has been associated with cellular damage.

However, we observed that the amount of lysoplasmalogens as a proportion of total plasmalogens in the liver of Tmem86b KO mice was only ∼3.5%, indicating that elevated lysoplasmalogens are rapidly converted into plasmalogens within the liver. In adipose tissue-specific Tmem86a KO mice, which also exhibited higher lysoplasmalogens, no toxic effects were observed. Instead, these mice showed elevated mitochondrial oxidative metabolism and energy expenditure, offering protection from high-fat diet-induced metabolic dysfunction. These findings suggest that any potential toxic effects of lysoplasmalogens are largely mitigated by their rapid reacylation into plasmalogens.

This study enhances our understanding of regulatory mechanisms governing plasmalogen metabolism, and highlights the potential of targeting Tmem86b to therapeutically raise plasmalogen levels.”

https://www.jlr.org/article/S0022-2275(25)00068-9/fulltext “Modulation of endogenous plasmalogens by genetic ablation of lysoplasmalogenase (Tmem86b) in mice”


An independent researcher published a commentary on the above study:

“While the biosynthesis of this particular lipid subclass, starting in the peroxisomes and ending at the endoplasmic reticulum, has been the subject of extensive research, the degradation pathway of these compounds remains to be further elucidated. Plasmalogen breakdown is a complex process involving enzymatic hydrolysis, oxidative cleavage, and possibly also a recycling mechanism.

A fundamental unresolved question in the field of plasmalogen catabolism is which of the two possible reaction routes is actually the more important one. Either 1) directly via plasmalogenase or 2) via a deacylation step by a plasmalogen-specific phospholipase A2 (cPLA2, PLA2G4A), yielding a lysoplasmalogen as the first degradation product, and subsequent hydrolysis of the ether bond by a lysoplasmalogenase such as TMEM86A and TMEM86B. It is also unclear how these pathways interact or compensate for each other, how they are regulated, and whether they are tissue- or cell type–specific.

To make the story even more complex, a CoA-independent transacylase activity was described that reacylates lysoplasmalogen intermediates back to plasmalogens by transferring polyunsaturated fatty acids to the vacant sn-2 position of ether lysophospholipids. But no gene for this enzyme has so far been identified.

Why is plasmalogen breakdown so important? Disturbances in plasmalogen metabolism are associated with several human disorders. Neurodegenerative diseases such as Alzheimer’s disease, Parkinson’s disease, and multiple sclerosis have been shown to be associated with reduced levels of plasmalogens.

Unfortunately, it is still too early to draw conclusions about the individual roles of TMEM86A and TMEM86B, as their cellular localisation and function are not sufficiently studied, and reliable antibodies for these proteins are not yet available. Localization of the two TMEM86 homologs overlaps to some extent, as shown, for example, by their gene expression in small intestine. However, whether one isoform is able to compensate for a deficiency in the other is uncertain, and was not found in small intestine of Tmem86b knockout mice [in the above study].

In contrast to the two proteins TMEM86A and TMEM86B, cytochrome c is much better studied. It is associated with the inner mitochondrial membrane, and can be released into the cytosol during apoptosis. It has a wide tissue distribution with most abundant gene expression levels in the digestive tract and heart.

https://www.jlr.org/article/S0022-2275(25)00074-4/fulltext “Plasmalogen. Quo vadis?”

The statement “no gene for this enzyme has so far been identified” revealed a paradigm. But maybe what’s being observed evolved before genes?

One example of this principle is from the 1966 https://www.science.org/doi/10.1126/science.152.3720.363 “Evolution of the Structure of Ferredoxin Based on Living Relics of Primitive Amino Acid Sequences” which provided evidence pointing to heme protein evolution beginning before gene evolution. Its abstract included this statement:

“We explain the persistence of living relics of this primordial structure by invoking a conservative principle in evolutionary biochemistry: The processes of natural selection severely inhibit any change in a well-adapted system on which several other essential components depend.”

Maybe the process of reassembling plasmalogen breakdown products back into plasmalogens without involving a specific gene likewise became essential?


A role of plasmalogens in diabetic kidney disease was found in a third study that investigated a genetic rodent model of diabetes:

“Diabetic nephropathy (DN) represents a frequent cardiovascular complication of diabetes, affecting about 20–50% of individuals with the disease. Globally, it constitutes a primary etiology for end-stage kidney disease (ESKD) and chronic kidney disease (CKD), while also serving as a significant independent risk factor for cardiovascular morbidity and mortality.

Although intensive management strategies targeting blood pressure and glucose levels demonstrably attenuate the risk of DN development, they do not confer complete protection. This residual risk strongly implicates pathogenic factors beyond impaired glucose metabolism and hemodynamic alterations in DN pathogenesis.

In the present study, we employed the db/db mice as the DN model. When compared to other diabetes models, such as those induced by streptozotocin (STZ) or high-fat diet combined with STZ, the db/db model more accurately recapitulates the pathological features of human type 2 diabetes mellitus (T2DM). It also possesses a stable genetic background, making it particularly well-suited for the investigation of diabetes complications.

Transcriptomics revealed extensive dysregulation of metabolic and lipid regulatory pathways in db/db. Lipidomics uncovered pronounced abnormalities in cardiolipin species composition and plasmalogen profiles. Transcriptome-lipidome integration demonstrated impaired phosphatidylcholine (PC) biosynthesis, mechanistically linked to dysregulation of choline phosphotransferase 1 (chpt1), which correlated significantly with compromised tissue regeneration capacity.

Volcano plot analysis delineated specific lipid alterations, particularly in plasmalogen species in plasmalogen lipids. Plasmenylcholines (plas-PC) and plasmenylethanolamine (plas-PE) containing n-3 polyunsaturated fatty acids (PUFAs) were significantly decreased in the kidneys of db/db mice. Conversely, plas-PCs and plas-PEs esterified with n-6 PUFAs showed substantial accumulation in diabetic kidneys.

In conclusion, the highly sensitive and extensively targeted UHPLC-MS/MS methodology enabled a more in-depth characterization of renal metabolic and lipid perturbations in db/db mice. These alterations principally reflect the sustained inflammatory milieu and compromised antioxidant defenses characteristic of DN renal tissues.”

https://www.csbj.org/article/S2001-0370(25)00301-0/fulltext “Multi-omics characterization of diabetic nephropathy in the db/db mouse model of type 2 diabetes”

Plasmalogens Week #5 – Health and Diseases, Part 1

Continuing Plasmalogens Week with three 2025 papers, starting with a human study that included plasmalogen biomarkers of non-communicable disease fatigue symptoms:

“This study explored the biological mechanisms underlying fatigue in patients with NCDs using a multi-omics approach. Our findings indicate that distinct metabolic pathways, salivary microbiota, and genetic factors may contribute to different dimensions of fatigue, including general, physical, and mental fatigue.

  • General fatigue is associated with unsaturated fatty acid biosynthesis, indicating its role in lipid metabolism.
  • Physical fatigue was associated with plasmalogen synthesis, mitochondrial beta-oxidation of long-chain fatty acids, and selenoamino acid metabolism, suggesting a potential contribution of impaired energy production.
  • Mental fatigue is associated with homocysteine degradation and catecholamine biosynthesis, which may influence cognitive fatigue.

This exploratory study suggests that fatigue in patients with NCDs may involve disruptions in lipid metabolism, neurotransmitter pathways, microbial composition, and genetic variations. Blood-based biomarkers showed better predictive potential for physical fatigue, whereas salivary-based models were more indicative of mental fatigue.

Although our findings support the role of lipid metabolism, the contribution of plasmalogen synthesis remains underexplored. Further studies are needed to validate these findings and understand their mechanisms of action.”

https://link.springer.com/article/10.1186/s12911-025-03034-3 “Visualizing fatigue mechanisms in non-communicable diseases: an integrative approach with multi-omics and machine learning”


A human study of metabolic dysfunction-associated steatotic liver disease (MASLD) included investigating plasmalogens:

“In this study, we applied untargeted metabolomic profiling to serum samples from individuals with and without MASLD, classified by the Fatty Liver Index, with the goal of identifying characteristic metabolic signatures and pathways that may underlie disease presence and progression. Individuals in the MASLD group displayed significantly higher levels of ALT, AST, ALP, and GGT, reflecting ongoing hepatic injury, cholestasis, and oxidative stress. However, albumin and bilirubin levels remained within normal limits, indicating early to intermediate disease stages rather than advanced fibrosis or cirrhosis.

A consistent and highly significant lipidomic pattern in the MASLD group is the depletion of plasmalogens and sphingomyelins. Depletion of these lipid classes was identified as a hallmark of insulin resistance as defined by the triglyceride-glucose index. In contrast, phosphatidylcholine, phosphatidylethanolamine, and phosphatidylinositol species were elevated in MASLD, pointing toward broader lipid remodeling events.

Reduced plasmalogen and sphingomyelin levels positions their depletion as a core feature of metabolic dysfunction. Plasmalogens are ether phospholipids with strong antioxidant capacity, and their reduction suggests a loss of protective buffering against oxidative stress, one of the main drivers of MASLD progression. Similarly, sphingomyelin depletion implicates altered membrane dynamics and signaling disturbances, further contributing to metabolic dysfunction.

Depletion of plasmalogens 1-(1-enyl-palmitoyl)-2-oleoyl-GPC (P-16:0/18:1), 1-(1-enyl-palmitoyl)-2-linoleoyl-GPC (P-16:0/18:2), 1-(1-enyl-palmitoyl)-2-palmitoyl-GPC (P-16:0/16:0), 1-(1-enyl-palmitoyl)-2-palmitoleoyl-GPC (P-16:0/16:1), 1-(1-enyl-palmitoyl)-2-oleoyl-GPE (P-16:0/18:1), 1-(1-enyl-palmitoyl)-2-linoleoyl-GPE (P-16:0/18:2), and disruption of the glutamate–gamma-glutamyl pathway stand out as central features of metabolic dysfunction in MASLD, with clear potential to inform biomarker discovery, disease classification, and the design of targeted therapeutic strategies.”

https://www.mdpi.com/2218-1989/15/11/687 “Metabolomic Signatures of MASLD Identified by the Fatty Liver Index Reveal Gamma-Glutamyl Cycle Disruption and Lipid Remodeling”


A rodent study investigated dietary sea squirt (AM) plasmalogen ethanolamine (PlsEtn) extract’s and dietary pig liver (PL) phosphatidyl ethanolamine (PtdEtn) extract’s effects on acetaminophen liver injury:

“We investigated dietary effects of PlsEtn from ascidian on chronic hepatic injury in acetaminophen (APAP)-treated mice. Five-week-old male mice were divided into four groups (n = 12), which were treated with experimental diets for two weeks and then the respective APAP-containing diet for five weeks.

Ingested PlsEtn is digested into lysoPlsEtn and free fatty acid in the small intestine. PlsEtn digests are absorbed and are subsequently resynthesized into PlsEtn preferentially with PUFA.

Acetaminophen is a frequently used analgesic and antipyretic. Approximately 90% of APAP is metabolized by UDP-glucuronosyltransferase and sulfotransferase into glucuronic acid and sulfate conjugates, respectively.

5–9% of APAP is metabolized into the highly reactive intermediate N-acetyl-p-benzoquinone imine (NAPQI). This metabolite is considered a pivotal molecule in APAP-induced hepatotoxicity and is conjugated by glutathione (GSH). Excessive NAPQI levels deplete GSH and covalently bind to cellular proteins, resulting in organelle dysfunction, such as mitochondria dysfunction. These impairments induce oxidative stress, cell malfunctions, and subsequently, cell death, such as ferroptosis and apoptosis.

Mice were treated with continuous APAP consumption to induce oxidative stress and impaired lipid metabolism in the liver. Effects of diets were evaluated based on levels of malondialdehyde (MDA), a marker of lipid oxidation, on fatty acid content, and on expression of apoptosis-related proteins in the liver.

The PlsEtn-rich diet effectively suppressed APAP-induced decrease in body and liver weights of mice. However, this suppressive effect was not observed in mice fed a PtdEtn-rich diet. APAP administration decreased the total fatty acid content in the liver, whereas a PlsEtn-rich diet alleviated this decrease and increased the hepatic content of docosahexaenoic acid (DHA).

Owing to the alkenyl linkage, which exhibits antioxidant properties, PlsEtn was expected to markedly suppress hepatic lipid oxidation. However, its suppressive effect was the same extent as that by PtdEtn. Both PlsEtn and PtdEtn contain an ethanolamine base in their structures, and free ethanolamine and its metabolite choline suppress lipid peroxidation. Dietary PlsEtn and PtdEtn may be metabolized into free ethanolamine and its further metabolites, which may alleviate APAP-induced hepatic lipid oxidation.

Dietary ethanolamine glycerophospholipids (EtnGpls) rich in PlsEtn or PtdEtn suppressed APAP-induced lipid oxidation in the liver. Protein expression results revealed that dietary EtnGpls reduced expression of certain apoptosis-related proteins compared to the APAP group. This reduction was more effective in mice fed the PlsEtn-rich diet than in those on the PtdEtn-rich diet.”

https://www.mdpi.com/2076-3417/15/11/5968 “Dietary Ethanolamine Plasmalogen from Ascidian Alleviates Chronic Hepatic Injury in Mice Treated with Continuous Acetaminophen”

This study neither demonstrated nor provided citations for its dietary plasmalogen recycling statements.


Three more plasmalogen health and disease papers are curated in Part 2.

A Nrf2 treatment for ALS?

A 2025 rodent in vivo / human cell ex vivo study investigated effects of a Nrf2 activator on ALS rodent models and ALS human nervous system cells:

“M102 is a central nervous system (CNS) penetrant small molecule electrophile which activates in vivo the NF-E2 p45-related factor 2-antioxidant response element (NRF2-ARE) pathway, as well as transcription of heat-shock element (HSE) associated genes. Apart from the recent promising emergence of tofersen as a disease modifying therapy for the 2% of ALS patients who harbor mutations in the SOD1 gene, other approved drugs have only marginal effects on life expectancy (riluzole) or indices of disease progression (edaravone).

Data from disease model systems and from human biosamples provide strong evidence for a role of redox imbalance, inflammation, mitochondrial dysfunction, and altered proteostasis, including autophagy and mitophagy, as four key drivers in the pathobiology of ALS. We demonstrate that M102 is a dual activator of NRF2 and HSF1 transcription factor pathways, two upstream master regulators of neuroprotective mechanisms, with the potential to modulate all four of these key drivers of neurodegeneration and with excellent penetration across the blood brain barrier.

Stress response of the KEAP1-Nrf2-ARE system is stronger in astrocytes compared to neurons. A body of evidence from in vitro and in vivo model systems and from post-mortem CNS tissue from ALS patients has indicated that the NRF2 response is impaired in ALS, and has also been shown to decline with age.

HSF1 is a stress-inducible transcription factor that is the key driver for expression of multiple heat shock proteins which act as chaperones responsible for correct folding of newly synthesized proteins, refolding of denatured proteins, and prevention of aggregation of misfolded proteins. However, to date, many small molecule activators of HSF-1 have shown undesirable properties e.g. by acting as Hsp90 inhibitors or by exerting direct proteotoxic effects.

M102 (S-apomorphine hydrochloride hemihydrate) is a proprietary new chemical entity (NCE) and the S-enantiomer of the marketed R-apomorphine (Apokyn®; pure R-enantiomer). The R-enantiomer is a dopamine agonist administered subcutaneously for management of advanced Parkinson’s disease. M102 is a very weak dopamine antagonist and does not show the adverse effects associated with dopamine agonism.

M102 treatment rescues motor neuron (MN) survival in co-cultures with C9, SOD1 and sporadic ALS patient-derived astrocytes. Other NRF2 activators have been investigated in clinical trials or have been approved for medical use. These include dimethylfumarate (DMF) (Tecfidera®, Biogen) and omaveloxolone (Reata, Biogen).

  • DMF was originally approved for the treatment of psoriasis (Fumaderm®) and was later repurposed for the treatment of relapsing-remitting multiple sclerosis (Tecfidera®). A phase 2 trial of DMF in ALS provided Class 1 evidence of safety at a dose of 480 mg/day and lack of disease-modifying efficacy. DMF treatment is associated with dose-limiting lymphopenia and flushing (Tecfidera® Prescribing Information).
  • Omaveloxolone (Skyclarys®) is a potent NRF2 activator that has been approved by the FDA and EMA for the treatment of Friedreich’s ataxia. By activating the NRF2 pathway, omaveloxolone ameliorates oxidative stress and improves mitochondrial function. As a potent NRF2 activator, omaveloxolone exhibited significant liver toxicity with elevated AST/ALT levels in 37% of patients exposed to a dose of 150 mg.
  • Toxicity has also been reported with other potent NRF2 activators, such as bardoxolone methyl (EC50: 53 nM) which showed significant heart, liver, and renal toxicity in humans.

In contrast, our preclinical toxicological studies indicate that M102 has a much higher safety margin in relation to liver toxicity. M102 has the potential to modulate multiple key drivers of neurodegeneration, increasing the chances of achieving impactful neuroprotection and disease modifying effects in ALS.

This comprehensive package of preclinical efficacy data across two mouse models as well as patient-derived astrocyte toxicity assays, provides a strong rationale for clinical evaluation of M102 in ALS patients. Combined with the development of target engagement biomarkers and the completed preclinical toxicology package, a clear translational pathway to testing in ALS patients has been developed.”

https://molecularneurodegeneration.biomedcentral.com/articles/10.1186/s13024-025-00908-y “M102 activates both NRF2 and HSF1 transcription factor pathways and is neuroprotective in cell and animal models of amyotrophic lateral sclerosis”


Combining exercise with sulforaphane

A 2025 clinical trial with old people compared NRF2 effects of acute exercise with pre- and post-exercise sulforaphane treatment:

“This study tested the hypothesis that combining acute exercise (in vivo stimulus) with ex vivo sulforaphane (SFN) treatment would induce greater NRF2 activation and signaling in older adults compared to either treatment alone. This approach was used to bypass the potential issue of inter-individual variability in metabolism and bioavailability of SFN supplementation through oral consumption and thereby provide more rigorous biological control to establish mechanistic feasibility.

Twenty-five older adults (12 men, 13 women; mean age: 67 ± 5 years) performed 30-min cycling exercise. Blood was drawn before and immediately after exercise to isolate peripheral blood mononuclear cells (PBMCs) and incubate with and without SFN (5 μM) treatment.

Acute exercise induced modest transcriptional changes across the four tested transcripts compared to the robust upregulation elicited by SFN. This disparity was notable given the comparable NRF2/ARE binding activity observed between EX and SFN.

Near-significant trends were observed for EX in heme oxygenase-1 (HO-1), NAD(P)H quinone oxidoreductase 1 (NQO1), and glutathione reductase (GR) (after Bonferroni correction), while glutamate-cysteine ligase catalytic subunit (GCLC) was not induced by EX. In contrast, SFN alone robustly induced expression of NQO1, HO-1, GR, and GCLC.

We had chosen 5 μM as the dose based on pilot data from our laboratory and existing literature from in vitro experiments. However, typically, SFN is not combined with another stimulus.

To test this speculation, we ran a post hoc dose–response experiment where we stimulated PBMCs (n = 5) at six different SFN concentrations ranging from 0 to 20 μM (incubated for 5 h) and analyzed responses across the four genes used in the present study. The dose responses displayed hormetic curves for NQO1, GR, and GCLC, with 5 μM eliciting the peak response, suggesting that the lack of difference between SFN and the combined treatment was due to a ceiling effect of the SFN dose. Interestingly, HO-1 displayed a linear/curvilinear response with the maximal observed response at 20 μM.

In future ex vivo studies, a sulforaphane concentration of 1–2 μM in combination with acute exercise is predicted to enhance the expression of these antioxidant genes in the PBMCs of older adults to a greater extent than either treatment alone. Furthermore, lower SFN plasma concentrations are more likely to be achievable with oral supplementation.

To our knowledge, this is the first trial to measure responses to acute exercise combined with sulforaphane stimulation on NRF2 signaling in older men and women. We did not observe any statistically significant differences in any of our outcome variables between men and women.

Our results demonstrate that combining acute exercise with a sulforaphane stimulus elicits a greater response in nuclear NRF2 activity in older adults. While the response in gene expression did not completely mirror the response in NRF2 activation, it is important to note that NRF2 induces hundreds of cytoprotective genes. The four transcripts we measured are among those most commonly used to represent NRF2 signaling but do not capture the full picture. Full transcriptomics in future studies would address this question.”

https://link.springer.com/article/10.1007/s11357-025-01939-5 “Sulforaphane improves exercise-induced NRF2 signaling in older adults: an in vivo-ex vivo approach” (not freely available) Thanks to Dr. Tinna Traustadóttir for providing a copy.


I asked two questions, and will follow up with replies:

  1. Did a second experiment test effects of these subjects eating broccoli sprouts prior to acute exercise? The clinical trial’s NCT04848792 Study Overview section indicated that was the researchers’ intent.
  2. What studies have the data that produced this study’s graphical abstract’s younger vs. older NRF2 response graph?

Sulforaphane and malaria

A 2025 rodent study investigated sulforaphane’s capability as an adjunct with standard treatment to inhibit resistant malaria strains:

“In this study, we performed proteomic analysis on a range of sensitive and artemisinin-resistant parasites, revealing specific dysregulation of PfK13 protein abundance. Reduced PfK13 levels were linked to impaired hemoglobin digestion, decreased free heme levels, and consequently, decreased artemisinin activation. Artemisinin resistant parasites also exhibited elevated thiol levels, indicating a more reduced cellular state.

Modulation of PfK13 levels or localisation modifies glutathione (GSH) levels, and elevated GSH decreases artemisinin potency. Elevated levels of reduced GSH and its precursor γ-glutamyl cysteine (gGlu-Cys) were observed in resistant parasites, while oxidised glutathione (GSSG) was lower.

In mammalian cells, SFN conjugates GSH, either passively or through the activity of glutathione-S-transferases, and the SFN-GSH conjugate causes oxidative stress. In response to this stress, Nrf2 translocates to the nucleus and interacts with the antioxidant response element (ARE) of target genes, resulting in expression of antioxidant genes, which induces an antioxidant response. However, P. falciparum has no identified Nrf2 orthologue and so likely lacks a KEAP1-Nrf2 mediated antioxidant response, which suggests that the SFN-GSH conjugate should only cause oxidative stress in parasites.

SFN has antioxidant properties for the host through activation of Nrf2. Therefore our molecule of choice would not only kill the parasite, but will boost the host antioxidant capacity. This differs from most other available pro-oxidants, which do not have this host antioxidant capacity.

5mg/kg SFN was found to be sufficient to significantly prolong the survival of artesunate-treated mice infected with parasites.

PfK13 mutations drive artemisinin resistance in Plasmodium parasites by enhancing antioxidant defences, which can be targeted by redox modulators such as sulforaphane. By leveraging SFN’s ability to induce oxidative stress and deplete thiol levels in parasites, this approach can enhance the efficacy of artemisinin and potentially restore its effectiveness against resistant strains.”

https://www.biorxiv.org/content/10.1101/2025.10.05.680568v1.full “PfK13-associated artemisinin resistance slows drug activation and enhances antioxidant defence, which can be overcome with sulforaphane”


Get a little stress into your life, Part 2

A 2025 reply to a letter to the editor cited 56 references to elaborate on Part 1 and related topics:

“A positive effect does not necessarily mean benefit, and positive effects on individual organisms may mean adverse effects on other coexisting organisms. However, a vast literature shows that hormetic stimulation can result in benefits depending on the context, for instance, clear growth, yield, and survival improvement.

There is some energetic cost to support hormetic stimulation, with a likely positive energy budget, which might also have negative consequences if there is insufficient energy substrate, especially under concurrent severe environmental challenges. Moreover, hormetic preconditioning could be particularly costly when there is a mismatch between the predicted environment and the actual environment the same individuals or their offspring might face in the future.

Hormesis should not be unilaterally linked to positive and beneficial effects without considering dose levels. For any research to answer the question of whether a stimulation represents hormesis and whether it is beneficial, robust dose–response evaluations are needed, which should be designed a priori for this purpose, meeting the requirements of the proper number, increment, and range of doses.

Both additivity and synergism are possible in the hormetic stimulatory zone, depending also on the duration of exposure and the relative ratio of different components. This might happen, for example, when a chemical primes stress pathways (e.g., heat shock proteins and antioxidants), thus enabling another chemical to trigger hormesis (defense cross-activation) and/or because combined low subtoxicity may modulate receptors (e.g., aryl hydrocarbon receptor and nuclear factor erythroid 2-related factor 2) differently than individual exposures (receptor binding synergy).

Moreover, even when stimulation occurs in the presence of individual components, stimulation may no longer be present when combined, and therefore, effects of mixtures cannot be accurately predicted based on the effects of individual components. There may be hormesis trade-offs; hormesis should be judged based on fitness-critical end points.

While often modeled mathematically, hormesis is fundamentally a dynamic biological process and should not be seen as a purely mathematical function, certainly not a linear one. Much remains to be learned about the role of hormesis in global environmental change, and an open mind is needed to not miss the forest for the trees.”

https://pubs.acs.org/doi/10.1021/acs.est.5c05892 “Correspondence on ‘Hormesis as a Hidden Hand in Global Environmental Change?’ A Reply”


Reference 38 was a 2024 paper cited for:

“Hormetic-based interventions, particularly priming (or preconditioning), do not weaken organisms but strengthen them, enhancing their performance and health under different environmental challenges, which are often more massive than the priming exposure.

The catabolic aspect of hormesis is primarily protective whereas the anabolic aspect promotes growth, and their integration could optimize performance and health. The concept of preconditioning has also gained widespread attention in biomedical sciences.”

https://www.sciencedirect.com/science/article/abs/pii/S1568163724004069 “The catabolic – anabolic cycling hormesis model of health and resilience” (not freely available)


Reference 40 was a 2021 review that characterized hormesis as a hallmark of health:

“Health is usually defined as the absence of pathology. Here, we endeavor to define health as a compendium of organizational and dynamic features that maintain physiology.

Biological causes or hallmarks of health include features of:

  • Spatial compartmentalization (integrity of barriers and containment of local perturbations),
  • Maintenance of homeostasis over time (recycling and turnover, integration of circuitries, and rhythmic oscillations), and
  • An array of adequate responses to stress (homeostatic resilience, hormetic regulation, and repair and regeneration).

Disruption of any of these interlocked features is broadly pathogenic, causing an acute or progressive derailment of the system.

A future ‘medicine of health’ might detect perilous trajectories to intercept them by targeted interventions well before the traditional ‘medicine of disease’ comes into action.”

https://www.sciencedirect.com/science/article/pii/S0092867420316068 “Hallmarks of Health”


Betaine as an exercise mimetic

A 2025 human study investigated effects of long-term exercise:

“Exercise has well-established health benefits, yet its molecular underpinnings remain incompletely understood. We conducted an integrated multi-omics analysis to compare effects of acute vs. long-term exercise in healthy males.

Acute exercise induced transient responses, whereas repeated exercise triggered adaptive changes, notably reducing cellular senescence and inflammation and enhancing betaine metabolism. Exercise-driven betaine enrichment, partly mediated by renal biosynthesis, exerts geroprotective effects and rescues age-related health decline in mice.

Betaine binds to and inhibits TANK-binding kinase 1 (TBK1), retarding the kinetics of aging.

Betaine effectively alleviated senescence phenotypes by reduced senescence-associated β-galactosidase (SA-β-Gal)-positive cells, decreased p21 expression, lowered DNA damage indicator γ-H2A.X, and elevated heterochromatin mark H3K9me3. Betaine treatment also enhanced cellular antioxidant capacity, as evidenced by increased NRF2 phosphorylation and reduced ROS accumulation.

These findings systematically elucidate the molecular benefits of exercise, and position betaine as an exercise mimetic for healthy aging.”

https://doi.org/10.1016/j.cell.2025.06.001 “Systematic profiling reveals betaine as an exercise mimetic for geroprotection” (not freely available) Thanks to Dr. Weimin Ci for providing a copy.


Stay away from NAC

A 2025 rodent study added several reasons to avoid non-emergency use of N-acetylcysteine:

“We previously showed that antioxidants induced an impairment of negative feedback of the hypothalamus-pituitary-adrenal (HPA) axis in rats, in parallel to a down-regulation of glucocorticoid receptor (GR) and nuclear factor erythroid 2-related factor 2 (Nrf2) expression in the pituitary gland. This study evaluated the role of the Nrf2-heme-oxygenase-1 (HO-1) pathway on impairment of negative feedback of the HPA axis induced by N-acetylcysteine (NAC).

Male Swiss-Webster mice were orally supplemented with NAC for 5 consecutive days. The Nrf2-HO-1 pathway activator cobalt protoporphyrin IX (CoPPIX) was injected intraperitoneally on days 2 and 5 after starting NAC supplementation.

NAC reduced expression of Nrf2 in the pituitary of mice. NAC induced adrenal enlargement and hypercorticoidism, along with a decrease in GRα expression and an increase of GRβ expression in the pituitary gland.

We observed that dietary supplementation with NAC ( Figure 4A ) significantly increased plasma corticosterone levels in mice 24h ( Figure 4B ) as well as 15 days ( Figure 4C ) after the last administration of the antioxidant with the same magnitude of response (3.5-fold and 3.4-fold, respectively).

Chronic activation of the HPA axis can have damaging effects on immune, cardiovascular, metabolic, and neural functions, increasing the risk of immune system dysfunction, mood disorders, and metabolic and cardiovascular diseases. To prevent these deleterious effects of chronic hypercortisolism, HPA axis function is controlled by a glucocorticoid-dependent negative feedback system that is essential for ending the stress response.

These findings showed that NAC reduced Nrf2-HO-1 pathway activation in the pituitary gland, in a mechanism probably related to a local downregulation of GRα and an up-regulation of GRβ, leading to a failure of negative feedback of the HPA axis and consequently to the hyperactivity of this neuroendocrine axis.”

https://pmc.ncbi.nlm.nih.gov/articles/PMC11827418/ “Activation of the Nrf2/HO-1 pathway restores N-acetylcysteine-induced impairment of the hypothalamus-pituitary-adrenal axis negative feedback by up-regulating GRα expression and down-regulating GRβ expression into pituitary glands”


A human equivalent to this study’s NAC dose is (150 mg x .081) x 70 kg = 851 mg. Human supplements are sold in 600 mg and 1000 mg doses.

Grok 3 replied to a question: Human equivalent time to 15 days in male Swiss-Webster mice aged between 4 and 6 weeks? by stating: “15 days in male Swiss-Webster mice aged 4 to 6 weeks corresponds to approximately 4.1 human years, advancing their equivalent human age from about 10–12 years to roughly 14–16 years.” Four+ years seems like a long time for NAC to steadily and continuously affect humans’ HPA axes per the above graphic. What do you think?

Previously mentioned reasons to avoid daily use of NAC were in the lower part of A good activity for bad weather days.

Practice what you preach, or shut up

A 2025 review subject was sulforaphane and brain health. This paper was the latest in a sequence where the retired lead author self-aggrandized his career by citing previous research.

He apparently doesn’t personally do what these research findings suggest people do. The lead author is a few weeks older than I am, and has completely white hair per an interview (Week 34 comments). I’ve had dark hair growing in (last week a barber said my dark hair was 90%) since Week 8 of eating broccoli sprouts every day, which is a side effect of ameliorating system-wide inflammation and oxidative stress.

If the lead author followed up with what his research investigated, he’d have dark hair, too. Unpigmented white hair and colored hair are both results of epigenetics.

Contrast this lack of personal follow-through of research findings with Dr. Goodenowe’s protocol where he compared extremely detailed personal brain measurements at 17 months and again at 31 months. He believes enough in his research findings to personally act on them, and demonstrate to others how personal agency can enhance a person’s life.

It’s every human’s choice whether or not we take responsibility for our own one precious life. I’ve read and curated on this blog many of this paper’s references. Five years ago for example:

So do more with their information than just read.

https://www.mdpi.com/2072-6643/17/8/1353 “Sulforaphane and Brain Health: From Pathways of Action to Effects on Specific Disorders”