We can do better.

Dr. Robert Naviaux named this response the Cell Danger Response, and proposed that in some chronic conditions, including autism, it never fully switches off. He did the research showing we're always in fight or flight, which matches our experience. What drew me to his work is that he traced the whole cascade of metabolic events that follows when cells sense a threat. Earlier research had shown that stress depletes nutrients. Naviaux mapped the full spread of damage. Calcium shifts inside the cell. Sulfur drains away. Nutrients are depleted, and others, vitamin D among them, become unusable even when they're present. He did it with real specificity, which makes the work worth reading in its own right. At the start of the cascade, purines change the meaning of ATP, turning it from an energy signal into a danger signal.

His research did flag that there are numerous times in development that a human body is supposed to shift out of fight or flight and autism doesn't do that. It brings more evidence to our experience of our baseline being a higher level of stress. He did test a medication called Suramin (which blocks the purine receptors) in mouse models, and then in a small placebo-controlled trial. Lowering the ATP danger signal temporarily in autistic children eased core autism features like being nonverbal, which returned as the effect wore off.

The drug is not the point in me bringing this up.

The point is the principle underneath it: when the danger signal was lowered, features that looked fixed changed. That points toward a metabolic state that can move, not fixed wiring, at least in some of us. I don't believe Suramin impacted all of the children the same way. But essentially, what went down can come back up, which is what I needed to heal burnout. It also highlighted for me that purines can be a powerful influence and one for me to pay attention to.

Note about Suramin: Suramin has toxicity issues with long-term use. Naviaux wants to use a low dose with autism, but I question if there are too many other factors at play to have that be effective longer than the medication is in the system. Genetics in autism range from ones that impact the brain to metabolic genes to mitochondrial damage, not to mention lead being another cause. It feels like he's not appreciating the diversity of genetics and reasons behind autism. I write this for burnout because the more studies I was exposed to, the more I understood how wide our differences could be. I struggle to understand how shutting off purines alone could have the impact he claims on the entirety of autism. I think purines are incredibly relevant to burnout, though, and his work helped me understand that pathway more.

One of the reasons I became so interested in purines is that the same pathway that starts the alarm also contains the brake. Purines get broken down in the body into other molecules as our system metabolizes them. Xanthine, a purine breakdown product, creates reactive oxygen species, fragments mitochondria, and runs high in adults with anxiety. The enzyme that makes uric acid also throws off ROS (reactive oxygen species) as it works, so purine breakdown is itself a direct source of oxidative stress that eats glutathione. This goes back to where I described the body as a kitchen that creates messes as part of its normal function. You want purines, you want them to be broken down, but you also want your body to be able to clean up the mess.

The breakdown of purines goes ATP → ADP → AMP → adenosine → inosine → hypoxanthine → xanthine → uric acid. So two of the steps necessary to handle purines in the body produce oxidative stress (xanthine and uric acid). The more purines the body handles, the more oxidative stress is created that needs to be cleared. More stress = more oxidative stress. Rest removes that burden temporarily, letting the body start to catch up. It does not support the whole process.

The pathway also has a brake called adenosine, which is an inhibitory signal that builds up when a cell has been working hard, promotes sleep, and lowers anxiety. Nudging adenosine up, including through movement, is beneficial metabolically. Adenosine's importance is best explained through its role in seizure conditions. In many seizure conditions, adenosine is broken down too quickly by the body and, when absent, seizures occur. Seizure medication is commonly based around slowing down the enzyme that breaks down adenosine so it stays in the system longer and reduces seizure activity. Keto diets do the same thing through epigenetic changes that happen when you switch from glucose as an energy source to fat.

The alarm and the brake share one system depending on how you support it, which now makes sense when I look at those of us who stay regulated through heavy activity. I wear a Garmin to monitor sleep, and the longer I go without exercise, the more my sleep deteriorates. It's proposed that adenosine rises when ATP is depleted, meaning intense exercise is necessary. However, what counts as intense exercise is directly related to how much ATP you have. I tried to exercise when I was deep in burnout and it was not effective. It's only now that I have spoons that I use exercise to improve sleep, and that was a gradual increase. Burnout greatly damaged my ability to exercise and required pretty extensive rebuilding.

The theory is that intense exercise depletes cerebral energy stores (ATP), and the resulting adenosine/inosine accumulation is part of what builds sleep pressure afterward, with sleep then serving to replenish those high-energy compounds. So the causal chain is: intense exertion → energy store depletion → adenosine rises → binds A1 receptors in the basal forebrain → suppresses wake-promoting neuronal activity → increases sleep pressure. This was helpful to understand for when and what exercise to introduce during different stages of healing.

Purines got personal once I learned the fight-or-flight response starts with purine signaling, and anyone autistic knows how high our anxiety runs. I have wondered how much my daily anxiety was producing purines that strained my nervous system, which fed more anxiety, which then strained it further.

The friends I have known who were people-pleasers or performers, myself included, struggled most with burnout or crashed even further. People-pleasing is a fear response that would add to our metabolic load. I'm not sure if those of us who perform instead of people please would be included in the fear response, but it is nonetheless a deviation from our true selves. I imagine there are elements of fear involved. It's not beneficial.

My answer to burnout leaned heavily on supporting certain pathways and enzymes and eventually addressing the gut dysbiosis to further lower the metabolic load. However, it showed me whole new degrees of nervous system regulation and being able to influence myself somatically. When I wasn't overstimulated, I could direct my energy much better. When I could direct my energy better, I could focus my life instead of surviving it. When I wasn't so emotionally dysregulated, I could achieve much more emotional movement in therapy. When I could find more balance emotionally, I began growing again.

Purines eventually turn into uric acid, making uric acid one way to look at how this pathway is functioning.

Low uric acid is the more common autism finding and has been found low enough in studies to be proposed as a biomarker of autism. Low uric acid means the antioxidant end is thin and the buffer depleted.

A distinct subset runs high, a hyperuricosuric group, with sharply increased synthesis from scratch. So this is not everyone-is-low or everyone-is-high, and it is not the whole picture in autism, but I found it interested in a group that is known to be low glutathione. It's another antioxidant side that is commonly depleted, which is what I kept seeing in the studies. We consistently got hit from multiple directions, not just one.

The pathway that coordinates energy, danger signaling, antioxidant production, and stress recovery is also one of the pathways most consistently disrupted in autism. I have no way of knowing if it is in me. I want to be very honest that I operated from the assumption that it applied to me before I assumed that it didn't. I've seen burnout spiral into more extreme consequences, and I am my favorite guinea pig. I don't expect others to do the same, but I want to be honest about that and supply the information I've learned so people can ask questions that might be more helpful in their healing. I believed I had seen how much worse it could get, and I went all in. I wasn't interested in being conservative. I was interested in surviving.

Supporting those pathways nutritionally was vital to me. Nutrient depletion on top of a disrupted pathway means even more disruption. I also put more weight on learning modalities like breathwork to give me tools to use in the moment. I'm still learning to stay regulated in a wider range of situations, and it's turning into an incredible way to explore life.

This doesn't prove purines cause burnout. It does ask the question of if autistic people may begin with a system that is already less efficient at processing the metabolic demands of stress, and there is currently no specific-enough research on burnout or Level 1 to know. Level 1 would theoretically have the least dysregulated pathways, which is why it takes years of accumulation until we finally start breaking. It also leaves room for nutritional and somatic support to have the most impact. As I'll go into later, the support I needed was not at all what I thought it was before I did this work.

The simple version: purine metabolism, the same system behind energy, danger signaling, and antioxidant production, is one of the most consistently disrupted pathways in autism. That doesn't prove it causes burnout, but it's a strong sign autistic bodies may start out less efficient at handling the metabolic cost of stress.

One of the things that surprised me was how closely purine metabolism and folate metabolism overlap. I only found this when I started digging into details, trying to make a better system for me to follow for my health. I hadn't expected the two pathways to meet here at all. Two steps of the purine assembly line require folate. What that means is a purine problem and a folate problem can look remarkably similar to the cell.

There was, in fact, a study that modeled exactly this: knocking out the purine genes produced nearly the same fingerprint as a simulated folate deficiency. (Geryk et al. (2020), "The Key Role of Purine Metabolism in the Folate-Dependent Phenotype of Autism Spectrum Disorders: An In Silico Analysis.") They modeled knockout of six purine-synthesis genes (GART, PFAS, PPAT, PAICS, ATIC, ADSL) and found it produced nearly the same metabolic blockage pattern as simulated folate depletion.

I'm not implying that a folate deficiency causes autism. I'm saying that a folate deficiency can make our ability to handle stress through purine metabolism worse. I go into folate further down, because that's not the simplest question either.

Purines aren't simply one molecule. As ATP is used, it moves through a series of different purine compounds, each with its own biological role.

ATP → ADP → AMP → adenosine → inosine → hypoxanthine → xanthine → uric acid

Most of us only learn that ATP is our energy. What fascinated me was what happened when I followed the molecule after that point: each step along the pathway does something different.

• ATP powers the cell.
• Adenosine becomes one of the body's primary inhibitory signals, promoting sleep and helping calm the nervous system.
• Uric acid becomes one of the body's major circulating antioxidants.

Looking at the pathway this way changed how I thought about burnout. Purines aren't simply producing energy. They are participating in nearly every phase of the stress response, from creating energy to signaling danger to supporting recovery afterward.

Before learning about purines, I thought burnout was mostly about running out of energy. Now I think it is better understood as placing repeated demands on a metabolic system responsible for much more than energy production.

Purines help coordinate energy production, danger signaling, DNA and RNA synthesis, folate metabolism, antioxidant production, sleep, and recovery. When that system is functioning well, the body can respond to stress and then return to normal physiology. There's also a real mechanistic basis for a theoretical vicious cycle here. Folate cofactors are a shared, limited pool competed for by three processes at once: de novo purine synthesis, thymidylate synthesis, and homocysteine remethylation. If purine demand rises, because recycling isn't keeping up and more has to be built from scratch, it pulls folate away from the other two processes, straining methylation and DNA synthesis capacity from a different direction. That's a genuine feedback structure, not just two separate problems running in parallel.

When it struggles, every stress response theoretically becomes metabolically more expensive. Right now our narrative around folate has to do with helping anxiety and that in some of us, folate to the brain is blocked. This broadened folate for me dramatically. I took all this as I really didn't want to be folate deficient and be autistic. There's too much at stake, at least for me.

The simple version: the body prefers to recycle purines because building new ones is expensive. When recycling falls behind, the body has to spend more energy and folate making new ones from scratch, which is what turns an ordinary stress response into a bigger metabolic bill.

The common narrative is MTHFR does not cause autism but we all have it so take methylfolate, it improves neurotransmitters, anxiety, and you've done well. I started there, then went deeper. What I think is important to know is that once folate is in a usable form, it can be used in two directions, and both are commonly depleted in autism.

• Toward SAMe, the body's primary methyl donor, used in hundreds of reactions: neurotransmitter synthesis, gene expression, inflammation control. Without enough SAMe, methylation falters and a lot of downstream processes feel it.
• Toward glutathione, the master antioxidant and detox molecule that handles oxidative stress, supports immunity, and is critical for repair.

One way to hold it: SAMe helps the body run its processes, and glutathione helps it survive the damage done while running them. Supporting them takes more than folate. SAMe in full needs folate, B2, B6, B12, zinc, and magnesium. Glutathione needs glutamine, cysteine, glycine, B6, B2, and selenium. It was never as simple as one pill. That being said, every study I've seen giving multivitamins to autism have shown improvement. The bigger picture of what I needed to support myself did untangle with grit, patience, and time.

Folic acid started being added into our food in 1998. It is also what is commonly found in most multivitamins. Folinic acid can be bought over the counter or given in prescription form called Leuvoceryn and is given specifically to replete folate levels in the brain because it's better at crossing the blood brain barrier. Methylfolate is what we seem to hear about the most, and it has its use but does not cross the blood brain barrier as well.

Folic acid, folinic acid, and methylfolate aren't interchangeable, and the difference comes down to which enzyme each one needs to become usable. Folic acid has to pass through DHFR (dihydrofolate reductase) before the body can use it. DHFR is slow and easily saturated, so a high dose of folic acid just leaves the extra circulating in the blood, unmetabolized (UMFA). Folinic acid skips DHFR entirely and enters the cycle further downstream, which is likely why it works more reliably for people whose DHFR conversion is sluggish. Methylfolate is already the finished, active form, so it doesn't need DHFR or MTHFR at all, but it does need B12 to hand its methyl group off; without enough B12, it can get stuck in that form instead of cycling through.

Both folinic acid and methylfolate bypass the MTHFR variant, so they are equal there, but not elsewhere. In one comparison, people with the MTHFR 677CT genotype lowered homocysteine more on folinic acid than on methylfolate, and folinic acid may support DNA synthesis and rapidly dividing tissues differently because of where it enters the cycle. I've met people that say they can't handle folinic acid but handle methylfolate better, but some people don't handle methylfolate well. It does a good job at adding methyl groups for the body to use, which can cause anxiety for some.

The critical detail: folinic acid alone did not meaningfully improve methylation or glutathione markers. It only moved them when combined with methylcobalamin, a specific form of B12. If you are taking folinic acid hoping to feel better, the combination is what moves the needle, not folinic acid by itself.

When I dug further, I think there are important questions to be asked when it comes to giving and taking any form of folate. Studies have found that greater intake of folic acid has been connected with dementia. This is where the dementia question comes in. A 2024 UK Biobank study of over 466,000 people found folic acid supplementation linked to higher Alzheimer's and vascular dementia risk. The proposed mechanisms are specific to folic acid: UMFA buildup, and folic acid masking an underlying B12 deficiency that then goes uncorrected. None of the dementia research I found implicates folinic acid or methylfolate specifically. So this looks like a folic acid problem, not a folate problem, but we don't know. Low folate still raises dementia risk generally. Since we're commonly given high levels of folinic acid and many of us take methylfolate, are those versions raising or correcting a dementia risk? Or something else?

What doesn't get argued about is the starting material. Methionine and cysteine are the raw inputs for this entire cycle. If protein intake is low, whether from a restricted diet, poor appetite, or GI issues that make eating difficult, the cycle is constrained at the very first step, before folate, B12, B6, or any enzyme even gets a chance to matter. You can have every cofactor in the world and still run short if there isn't enough methionine coming in to begin with. Methionine gets converted to cysteine. And with autism being a condition of oxidative stress, N-Acetyl-Cysteine is a supplement you hear about a lot in the world of autism, and I hope this explains some of why. Between needing it for glutathione and then every other process in the body that needs cysteine, of which there are many, we tend to benefit for more sulfur amino acids. Protein isn't about muscle building for us. It has more to do with sulfur amino acids and handling basic metabolic processes correctly.

The simple version: this whole cycle, purines, folate, glutathione, homocysteine, all of it, needs methionine to start it. Methionine comes from protein. No amount of supplementing downstream fixes a shortage upstream.

I still eat sugar, but I have to account for what it costs me. And now I eat much less and avoid high fructose corn syrup.

High-carbohydrate, high-sugar diets increase thiamine (B1) demand because thiamine is the required cofactor for the enzyme that processes glucose into cellular energy. The more sugar going through that pathway, the more thiamine gets used up, and refined sugar contains none of the thiamine needed to replace it. Rice is another one people miss. Rice is pretty easy to make a safe food, but white rice will deplete thiamine, and that has led to documented thiamine deficiencies in some populations. A thiamine deficiency is called beriberi.

Fructose specifically also hits the purine system I've already described. The enzyme that processes fructose in the liver burns through ATP rapidly, with almost no built-in brake on how fast it does this. That rapid ATP depletion pushes the purine breakdown pathway into overdrive, producing more uric acid as a byproduct. This is the same ATP-to-purine pathway turning "spoons" into breakdown products that I described earlier, except HFCS is actively accelerating it rather than stress alone.

Oxalates are a signaling molecule between bacteria and fungi, and they're present in every microbiome, from the forest floor to our own gut. Since autism is partly defined by gut dysbiosis, all of us are dealing with an oxalate burden to some degree. Oxalates matter for a second reason too: they share the sulfur pathway, and sulfur should be the third most abundant mineral in the body. Oxalates bind to minerals, especially calcium, which can create a functional deficiency even when calcium intake itself is adequate. And they disrupt enzymes, potentially more broadly than sulfites, purines, or aldehydes do.

Oxalate can be produced internally or absorbed from food. I removed the biggest dietary offenders and have been slowly lowering my level further as it's felt reasonable. I also made sure my B6 status was solid, because the AGXT enzyme depends on B6 directly. AGXT converts a compound called glyoxylate into glycine, an amino acid, and that conversion is the critical detoxifying step that prevents glyoxylate from turning into more oxalate and building up in tissue. Genetic AGXT problems are rare and cause a condition called primary hyperoxaluria, but oxalate levels found in autistic children have actually been measured higher than levels found in children with that genetic condition. Known consequences of oxalate buildup include kidney stones, which are frequently oxalate-related.

For me, the highest-offending foods were spinach, potatoes, celery, nuts, and carrots, and I removed those from my regular diet. If I have potatoes on Thanksgiving or eat a couple fries off someone's plate occasionally, that's not an issue for me personally, though it could be for someone else. I don't crave or miss these foods anymore, which surprised me. My general experience has been that once a diet change genuinely improves how I feel, I become self-regulating about maintaining it, because any regression back to being dysregulated doesn't feel good.

Where to learn more: the TLO (Trying Low Oxalates) group on Facebook is the most in-depth community resource I've found on this specifically.

I've mentioned that the molybdenum studies were done poorly. Here's specifically why: the current RDA for molybdenum is based on a total of four young men in 1995. Four. That's the entire human evidence we base the current adult RDA for molyb on, and none of those four were autistic, none were dealing with a heavier MoCo enzyme burden, and none were living with chronic gut dysbiosis. Out of those four, one of the men did show signs of his system struggling with the low amount. They also tried a higher amount, over 1 mg, with none of the men struggling. They chose to double the lower amount and call it a day, which is how we got the 45 mcg RDA. It's also not widely known that high protein intake raises your molybdenum requirement. Molybdneum is also not available in processed or canned foods. Sulfur is complex in autism, so I don't want to encourage taking high amounts, but it's a mineral I don't widely hear considered for us. There is a study on sulfur in autism that gave molybdenum to its participants with 1/3 of them showing improved handling of sulfur after supplementation, but they don't list how much molybdenum they gave.

There is no RDA for sulfate/sulfur at all. The Institute of Medicine looked at the evidence and explicitly declined to set one, reasoning that sulfur amino acid intake alone should cover it. But the researchers who actually study sulfur balance directly have pointed out a real gap in that reasoning: the existing amino acid requirements were only ever tested for their ability to maintain nitrogen balance, never sulfur balance specifically, and sulfur-balance testing "has never been done, neither in humans nor animals." Even the original studies behind the methionine RDA were run on groups of three to six people per amino acid. Studies in autism have shown we are often sulfur deficient and transsulfuration pathways are often an issue in us. Sulfur is complex in autism, but it's another nutrient that many of us experiencing burnout need to be aware of to heal well.

The simple version: the numbers we're told are "enough" were tested on a small handful of healthy young men decades ago, and nobody has ever gone back to check if they hold up for a population with a heavier metabolic load.

Most amino acids fall into two categories: essential ones the body can't make and has to get from food, and non-essential ones the body makes on its own. There's also a third category: conditionally essential amino acids, ones the body can normally make in adequate amounts, but not fast enough to keep up under illness, chronic inflammation, oxidative stress, or metabolic stress. Under those conditions, the body's own production can't meet demand, and the amino acid effectively becomes essential, something that has to come from the diet because synthesis alone won't cover it.

The list includes arginine, cysteine, glutamine, glycine, proline, and tyrosine. Look at that list next to what I've already described as chronically elevated in autism: oxidative stress, inflammation from LPS and gut dysbiosis, and a baseline metabolic load that's higher than average. Autism isn't a single acute illness, but it does check the boxes the conditionally-essential framework is built around: sustained oxidative stress and sustained metabolic demand.

This connects directly to glutathione. Glutathione is built from cysteine, glutamine, and glycine, three of the conditionally essential amino acids on that list. So the idea of "just take glutathione" is missing some important nuances. If the raw materials themselves are running short because the body's baseline synthesis can't keep pace with baseline demand, supplementing the end product without addressing the amino acid supply is treating the symptom of a supply problem, not the supply problem itself. It's also not addressing why we are depleting our glutathione.

The simple version: the amino acids autism needs more of, under stress, are exactly the ones standard nutrition guidance assumes the body always makes enough of on its own.

What changed my thinking was realizing that my body, via the gut dysbiosis, was producing a tremendous workload just by existing. When I learned about MoCo, that raised a new question: what happens when all of that work lands on the same enzyme? And what happens when that enzyme is down regulated in a population? Every day I was producing purines through stress, sulfites via the gut bacteria, aldehydes via gut bacteria, and all of those compounds had to be processed. I couldn't change how much down regulation this enzyme system had, but I could try to reduce what it had to process. The amount of stress we carry as a baseline is incredibly underestimated by a population that hasn't experienced it. Said another way, neurotypicals haven't experienced our level of dysregulation, and people generally do a very poor job of considering a reality they haven't experienced. The metabolic load being created is not the same as a population that has a regulated nervous system. It's so very important to keep that in mind.

And what's important to understand is that strain on any of the enzymes that MoCo controls strains the other enzymes. Purine overload? The aldehyde, sulfite, and mARC sides are going to suffer also. The nutrients needed to fuel MoCo get shared by the purine, sulfite, aldehyde, and mARC side. The only deviation from this is sulfur because only purines and aldehydes share this. Considering the majority of participants had a down regulation of this enzyme, I took into consideration that I potentially did also. I have no way of knowing, but looking back, there came a tipping point in my late 30s where I started to see sulfite and MCAS symptoms, and that's when my cognition started struggling. I question if that's the point where this enzyme became totally overwhelmed, depleted, whichever or both, and it started impacting my cognition and health.

This enzyme wasn't the whole picture. It was simply the bridge I had to understand to cross. Instead of trying to force my body to work harder, I started looking for ways to reduce the burden I was placing on it. Like using rest or unmasking to heal burnout, but on a bigger scale. That became the philosophy behind everything that follows.

The original study compared olfactory stem cells from eleven autistic adults with eleven neurotypical controls. Olfactory stem cells are unusual because they reflect very early development, which is the window where autism is thought to begin. Since you can't biopsy a developing brain, they're one of the closest windows researchers have into early neurodevelopment.

What they found was that MOCOS, the enzyme responsible for sulfurating the aldehyde and purine side of molybdenum cofactor, was downregulated in eight of the eleven autistic participants, roughly 80%. The gene itself wasn't mutated or missing. It was simply being expressed at lower levels. For a study this small, that stood out. Most autism-associated genes are found in relatively small subsets of people, so 8 out of 11 is hard to find, even in a small study. There's one article calling it the first gene strongly associated with autism.

Three out of 11 participants showed disruption of AOX1, the gene that encodes aldehyde oxidase, another MoCo-dependent enzyme. When the researchers combined the two findings, ten of the eleven autistic participants had either reduced MOCOS expression or disrupted AOX1. Based on their interpretation, one participant had neither and one participant had both. Two participants were considered Asperger's, and one of those was the participant that had no significant findings. This is an area I want to revisit as I go through this information more, because there are still open questions for me about the one who had no findings.

Sulfites, aldehydes, and oxalates aren't simply things the body has to process. If they aren't handled effectively, there are consequences from the buildup. Each one blocks or damages a specific enzyme directly, and that distinction matters, because it changes what happens when the matching enzyme is in short supply. With an ordinary nutrient deficiency, you have less of something. With an enzyme blockage, the substrate that enzyme was supposed to clear has nowhere to go, so it builds up. Sufficient nutrition isn't going to bypass a blocked enzyme. They're two different mechanisms happening.

There's a real hospital case that shows exactly this. A patient with Crohn's disease was kept on long-term IV nutrition (total parenteral nutrition) for over a year, a feeding method that bypassed his gut entirely and, as it turned out, they gave him essentially no molybdenum. Without molybdenum, his SUOX enzyme couldn't convert sulfite to sulfate. Sulfite built up. His xanthine dehydrogenase couldn't finish converting purines either, so xanthine built up while uric acid dropped. He developed a racing heart, rapid breathing, headaches, nausea, and vomiting, and it progressed to a coma. Blood and urine tests showed exactly what you'd expect: high sulfite, high xanthine, low sulfate, low uric acid. Doctors gave him 300 mcg (not milligrams, micrograms) molybdenum directly into his IV, and he recovered.

For the molybdenum cofactor specifically, running short doesn't just mean you're a little low on something. It means sulfite and other substrates the enzyme was supposed to clear start accumulating in the blood, and that accumulation is what actually does the damage.

Aldehydes work the same way. When AOX1, the MoCo-dependent enzyme that clears aldehydes, is underperforming, aldehydes don't just fail to clear, they accumulate and bind to proteins and DNA, doing more damage the longer they sit there than they would if cleared quickly.

This is the piece I think gets lost when people talk about "supporting detox pathways." These enzymes aren't optional extras. When they're underperforming and the matching compound keeps arriving faster than it can be cleared, the buildup itself is what causes the damage, sometimes severely and quickly, as the hospital case shows. There's no binder for aldehydes or sulfites or purines. Detoxing them for me had a very different route.

Pyridoxal kinase. Sulfite doesn't inhibit this enzyme directly, but it reduces the available pool of PLP (active B6) it depends on, which can indirectly slow the conversion of PL to PLP.

Glutamate dehydrogenase (GDH). Sulfite binds and inhibits GDH activity in a dose-dependent manner, meaning the more sulfite present, the more this enzyme is suppressed. Suppression of GDH certainly wouldn't be helping our glutamate excitoxicity.

Pyruvate dehydrogenase complex (PDH) and α-ketoglutarate dehydrogenase (KGDH). Sulfite doesn't bind these directly, but both require thiamine pyrophosphate (active B1) as a cofactor, and sulfite destroys thiamine by cleaving its methylene bridge. Losing the cofactor supply shuts these enzymes down just as effectively as a direct block would.

Enzymes with a critical disulfide bond or cysteine residue. Any enzyme that depends on a disulfide bond or an oxidized cysteine to function is vulnerable to sulfite directly breaking that bond. Two examples: thyroid peroxidase (TPO), which synthesizes thyroid hormone and depends on multiple disulfide bonds for its structure, and protein disulfide isomerase (PDI), which helps fold new proteins correctly in the first place.

Thioredoxin reductase (TrxR). Regenerates thioredoxin, a small protein central to keeping the cell's redox balance in check. Disrupted thioredoxin/TrxR signaling has specifically been found in autistic neutrophils.

CYP1A2. A liver enzyme involved in drug and toxin metabolism. In an animal model of sulfite oxidase deficiency, meaning excess sulfite exposure, CYP1A2 activity dropped by roughly 28%.

Reactive aldehydes, whether internally produced (acetaldehyde from normal metabolism, 4-HNE from lipid peroxidation) or from outside sources (acrolein, formaldehyde), damage proteins by forming permanent bonds at specific amino acid sites: Schiff base adducts at lysine or arginine, or Michael addition products at cysteine or histidine. When those sites sit in an enzyme's active center, the enzyme stops working.

Dopamine β-hydroxylase (DBH). Certain aromatic aldehydes act as mechanism-based inhibitors of DBH, the enzyme that converts dopamine to norepinephrine. Increased norepinephrine alongside decreased DBH activity has been specifically documented in primary autism.

Glucose-6-phosphate dehydrogenase (G6PD). The rate-limiting enzyme of the pentose phosphate pathway, which produces NADPH for antioxidant regeneration. Acetaldehyde blocks its catalytic activity directly. G6PD deficiency itself has also been studied for a possible association with autism.

Pancreatic ribonuclease (RNase A). A digestive enzyme that depends on lysine for catalysis. Acetaldehyde selectively inactivates lysine-dependent enzymes like this one.

Angiotensin-converting enzyme (ACE). Critical for blood pressure regulation. Acetaldehyde rapidly and directly inhibits ACE activity. Genetic variants of ACE have separately been linked to autism.

Glutathione reductase (GR). The enzyme that regenerates active glutathione from its spent, oxidized form. 4-HNE and acrolein irreversibly inactivate GR by attacking cysteine and histidine residues at its active site, first forming a reversible complex and then a permanent one. Other aldehydes, including acrolein, crotonaldehyde, and methylglyoxal, inactivate it through related mechanisms. Over time, these adducts can even cross-link GR's subunits together.

Arylamine N-acetyltransferase 1 (NAT1). Handles phase II detoxification of aromatic amines. Acrolein irreversibly inhibits it. The gene is separately linked to autism.

Aspartate aminotransferase (AST, cytosolic). A B6-dependent (PLP) enzyme involved in amino acid metabolism. Acrolein modifies and inhibits it directly.

PDH and KGDH, through a second route. I already mentioned these two under sulfites, because they need thiamine. They're also vulnerable to aldehydes through a completely separate mechanism: both depends on lipoic acid, and 4-HNE and acrolein preferentially attack the reduced (active) form of that lipoic acid, disabling the complex. One study found this caused near-complete inactivation of both PDH and KGDH while sparing other TCA cycle enzymes, meaning aldehydes hit this specific point in mitochondrial energy production especially hard.

A few of these enzymes sit right at the intersection of mast cell activation and dysautonomia, which is worth naming since both show up so often alongside autism and burnout.

Dopamine β-hydroxylase (DBH), again. Also needs copper as a cofactor, so a copper shortfall compounds the aldehyde-driven inhibition above.

Renin. Acetaldehyde triggers renin release directly from mast cells; in human mast cell experiments, acetaldehyde exposure caused dose-dependent degranulation and renin release together, which could activate the local renin-angiotensin system.

Tryptase. Both sulfite and oxalate can trigger this one. Sulfite causes non-IgE mast cell degranulation and tryptase release through an oxidative stress mechanism (NADPH oxidase activation). Separately, oxalate crystals have been shown to increase mast cell infiltration and activation in kidney tissue in animal models, meaning a high oxalate load is its own mast cell trigger.

Cyclooxygenase-2 (COX-2). Sulfites have been implicated in prostaglandin release in humans, and there's evidence for a prostaglandin role in sulfite-triggered asthma specifically.

Purine toxicity works differently from sulfites and aldehydes. Purine metabolites are normal cellular molecules, not foreign compounds, so "toxicity" here means the normal molecule building up past the level the body can handle.

Ribonucleotide reductase (RNR). Converts ribonucleotides into the deoxyribonucleotides needed for DNA replication, and is allosterically shut down by high levels of certain purine breakdown products (dATP, dGTP). In purine nucleoside phosphorylase (PNP) deficiency specifically, accumulated deoxyguanosine converts to dGTP, which inhibits RNR, hitting T-cells especially hard.

S-adenosylhomocysteine hydrolase (AdoHcy hydrolase). The enzyme that irreversibly breaks SAH down into adenosine and homocysteine, which is what drives the whole methylation cycle forward. High intracellular adenosine or deoxyadenosine locks this enzyme in an adenosine-bound state, effectively shutting it down, meaning a purine buildup can jam the methylation cycle at this specific step.

This one interests me personally. In 2015 I was diagnosed with Congenital Adrenal Hyperplasia (3β-Hydroxysteroid Dehydrogenase), based on an ACTH stimulation test, though it was always described to me as a "partial" enzyme deficiency. Every year since I started working out what nutrition my autism actually needed, my bloodwork on this has improved. My endocrinologist now runs things like a Renin test because she suspects the original diagnosis was wrong. I don't think it was wrong. As I've laid out through this whole section, enzymes can be disrupted by genetic mutations, but also by nutrient deficiencies or toxin interference. The functional result looks the same either way. The enzymes below are all part of the CAH family, and each has a documented toxin-driven vulnerability, separate from the genetic story.

CYP21A2 (21-hydroxylase). The classic CAH-associated enzyme.

CYP1A2. Already covered under sulfites above, listed here because it's also part of steroid-adjacent drug metabolism.

CYP11B1 (11β-hydroxylase). In a 2023 study, hyperuricemic (high-purine) patients and mice showed a blunted adrenal response to ACTH, with significantly decreased CYP11B1 expression and reduced cortisol synthesis as a result.

CYP17A1 (17α-hydroxylase/17,20-lyase). Highly sensitive to aldehyde-driven oxidative damage. In vitamin-E-depleted adrenal tissue, lipid peroxidation dropped 17α-hydroxylase activity to about 13% of normal and 17,20-lyase activity to about 18%, within five minutes.

HSD3B2 (3β-hydroxysteroid dehydrogenase type II). The enzyme in my own diagnosis. Converts pregnenolone-type steroids into progesterone-type steroids in the adrenal glands and gonads. In hyperuricemic mice, adrenal HSD3B2 expression was significantly downregulated alongside CYP11B1.

StAR (steroidogenic acute regulatory protein). Transports cholesterol into mitochondria as the first step of steroid production. In an animal study, acrolein (aldehyde) exposure during pregnancy significantly reduced fetal StAR expression, along with reduced HSD3B2 and testosterone, showing broad suppression of steroid-making capacity.

Aldehydes bind rather than float past. They form irreversible adducts with proteins, inactivating them, and with DNA, causing strand breaks and mutations. The AOX1 enzyme (part of MoCo) helps clear them. One hypothesis paper on aldehydes made the strong claim that the many symptoms and divergent theories of autism are consistent with aldehyde toxicity, and laid out four ways they damage us: micronutrient depletion, oxidative stress, protein inactivation, and DNA damage. Whether they explain all of autism or not wasn't my concern. I looked at them like another burden to make sure I wasn't adding to my system any more than necessary.

• Outside the body: vehicle exhaust, smoke, cooking fumes, browned and fried food (especially reused oil, like fast food), off-gassing from furniture, carpet, and cosmetics, and many foods and natural flavorings such as vanilla, cinnamon, and citrus. Indoor levels run four to ten times outdoor. Alcohol and smoking produce them too. But this wasn't my biggest concern because the studies suggest that the larger problem might be internal production, or at least that was my read of it. All the same, I don't want to work at a factory breathing in paint fumes all day. I stopped eating fried food pretty immediately, was careful to not burn my food, and simplified my cosmetics. I eventually was able to quit smoking weed altogether and alcohol as a weekly habit. I quit marijuana altogether, but edibles may not provide aldehydes the way smoking does. I haven't looked into this yet.
• Inside the body: reactive oxygen species hitting fats, called lipid peroxidation, throws off more than 200 different reactive aldehydes such as 4-HNE, 4-HHE, and MDA. ROS are unstable oxygen molecules from normal metabolism. In small amounts, antioxidants handle them, but when they outpace defenses you get a burden that contributes to oxidative stress, one of the most consistently replicated findings in autism.
• The gut: yeast and bacteria, Candida especially, generate them directly. Candida thrive on purines, but they offput aldehydes as a part of them living in our body. I did address candida in my gut.

They consume the body's defenses as they go, depleting sulfur antioxidants (glutathione, cysteine, taurine) and locking up B1, B6, folate, zinc, and magnesium, and they strip zinc out of proteins, including the very enzymes meant to clear aldehydes and ROS. The lipid-peroxidation ones travel across membranes and damage neighboring cells far from where they formed. The paper described the result as a downward spiral: aldehydes deplete the local antioxidants, which lets more survive, which disables protection further out, until repair machinery goes offline and DNA damage starts to stick. The study suggests earlier intervention rather than later.

The SUOX enzyme converts sulfite, one of the body's most reactive sulfur compounds, into sulfate, a stable nutrient the body can use. Autism has been found to be consistently sulfate deficient in studies. (There are separate genetic reasons that complicate this in some of us, but not for the majority of us. I do want to note that.) That conversion helps supply us with necessary sulfur to use for numerous processes, and without the conversion happening, it may be impacting areas we don't realize. I haven't found a good nutritional supply of sulfur that matches the conversion that happens in our body. The closest our body can use chemically is Epsom salt, which isn't meant to be ingested. One study specifically suggests it, but unfortunately there's no form meant to be ingested. Gallbladder groups drink it for their gallbladder flushes, so it has been ingested by people. It's another gap that could be addressed. San Peligrino water contains sulfur, which is why groups that like alkaline water hate that brand. You can't make sulfur water alkaline. If you have the funds, drinking that water is an option, but I don't know if it's sufficient to fill a gap.

Sulfate is not waste. It regulates dopamine, epinephrine, and norepinephrine through sulfation, maintains the gut's protective mucus lining, sulfates bile acids for fat digestion, and helps detoxify phenols. When sulfite is produced faster than SUOX can convert it, sulfite accumulates and sulfate falls: neurotransmitters linger, mast cells release more histamine, the gut barrier weakens, oxidative stress rises, and active B6 and thiamine get damaged. The autism literature consistently reports elevated urinary sulfite and reduced sulfate findings in autism.

I do want to note that the SUOX gene is not directly linked to autism, whereas numerous purine and aldehyde genetics are. However, there are genetics that cause sulfate wasting that have been found and studies on prenatal sulfur levels suggest that higher sulfur intake led to lower levels of autistic symptoms in autistic children. Giving molybdenum improved sulfate levels in 1/3 of autistic children in one study. There are many gaps in the research with sulfur and sulfite which would answer a lot of questions on what we need for our health.

When I say detoxify phenols, this is where the research seemed to stop and I wish it hadn't. This was studied in the 90s over several studies that were wondering how well a subgroup of autistic children handled phenols in foods (grapes, chocolate, etc. contain phenols). When the whole Tylenol debacle came out, I was immediately reminded of those studies because they used paracetamol, aka Tylenol, because it has the same phenol ring that the foods the children were reacting to had. What they found was that those children did not handle paracetamol in the body the correct way. They didn't have sufficient sulfur to handle Tylenol/paracetamol with the appropriate pathway, so their body relied on alternate pathways that are much harder on the body. In short, they didn't process Tylenol properly, and thus the studies concluded didn't handle foods with the same phenol amines properly.

Does Tylenol cause autism? Absolutely not. But Tylenol needs sulfur, and we're known to be sulfur deficient.

The studies back then speculated it was possibly an issue with the PST enzyme (now renamed SULT enzyme). It would be real helpful to know if this is only an issue in a subgroup or if all of us share this, because there are common pain killers that don't require sulfur, and this is where our general health would be impacted. When the people who did the study on Tylenol recently concluded that Tylenol could cause autism, they excluded all research that was not prenatal or natal. They never saw these studies done in the 90s. There are other studies that show that transsulfuration patterns are also present in the parents as well as the children. Their methodology in the study was sound. The question they asked lacked even a basic knowledge of austistic metabolism.

It should not be a surprise that a drug requiring sulfur aggravates a population that is consistently found to have sulfur issues.

The down regulation was in MOCOS (aldehydes and purine side), so this is where I have to bridge why I'm bringing sulfites into the picture. That bridge is lipopolysaccharides.

LPS sits in the outer membrane of gram-negative gut bacteria. Behind a healthy gut wall, it is normal physiology. The moment the barrier becomes permeable, LPS enters the bloodstream and becomes inflammatory.

LPS is one of the best-studied links between a leaky gut, behavior, and neuroinflammation. Everyone has it. The difference in autism is degree: autistic microbiomes consistently carry a higher proportion of gram-negative bacteria, so more LPS is theoretically produced to begin with, and the dysbiosis-related barrier damage means a more permeable wall for it to cross. I say theoretically because we haven't tested for LPS except one study on a profound population. I felt that filling that gap was necessary to heal. We have a dysbiosis. I did testing and know that I am very proteobacteria heavy. Those bacteria produce significant LPS, and I have a lot of the bacteria. More recent testing, now that I feel healed, shows my microbiome has shifted and I have significantly less proteobacteria, although still more than the average population or is ideal. My gut health has tracked with my symptoms.

There are numerous studies showing that our gut is permeable, but they are often testing something like zonulin. There are also numerous studies finding we have an increased immune response when given LPS, with that being more extreme in one subgroup. There is only one study I found measuring LPS directly in profound autism, and it found a direct correlation. No other tests have tested LPS levels. So we know LPS has neurological, metabolic, and immune inflammatory effects, we know we have a gut dysbiosis that would create more LPS as a byproduct, and we know that we react poorly to it. No one has tested the biggest question - do all of us have it?

This was the gap that caught my attention. It was a genuine gap: the studies were measuring the response, but what if we asked whether the trigger was already there? The studies seem to come at LPS to see how our immune systems respond instead of measuring if it already exists, so we know our immune systems overreact to it. This feels like such a foundational question that is not being asked directly.

Where this comes into burnout is that studies do show that stress increases LPS, and studies also show that LPS releases ATP. On top of that, there is a study showing if you block two certain purinurgic receptors, LPS-created inflammation is reduced. So the inflammation is, at least in part, showing itself through purine receptors. That raises another question about Navuix's work - Suramin blocks purine receptors. How much of the response to Suramin was from LPS? Suramin also is exceptional at killing candida because purines are energy, including for fungus. I looked at these connections and went, yup, I think addressing all of this will improve my health.

One of the larger disruptions to my decline was binding LPS specifically because it stopped a cycle that seemed to be happening. And why I'm pairing it with sulfites is because sulfites produce an immune response that produce more lipopolysaccharides. The solution I found here was to make sure I had the nutrients the MoCo enzyme needed and ate whole food so I reduced the sulfite load while supporting its conversion to sulfate as much as possible. Later on, when I started playing with ways to address sulfites further, I implemented chitosan. Chitosan binds very well to LPS and was a game changer. Spending a few weeks removing it gave my system a tremendous break, especially paired with the nutrition my body needed to catch up.

There's a paper that came out in 2025 that encompasses a big chunk of what I found with LPS. It's titled A Systems Hypothesis of Lipopolysaccharide-Induced Vitamin Transport Suppression and Metabolic Reprogramming in Autism Spectrum Disorders: An Open Call for Validation and Therapeutic Translation. What I would add is that LPS are doing damage on a body that already has weaknesses that I've talked about - purine and methionine pathways, among others. Fatty acid issues have been found, carnitine, more extreme folate genetics, etc. But the part doing damage that I can control are LPS, and this paper does a beautiful job of laying that out.

I want to make sure I explain this well. The interesting part is what happens when LPS meets the weaknesses already described.

The purine side is already potentially a problem, especially if you've experienced burnout. LPS-driven ATP release adds more purine turnover on top of that, straining the exact enzyme, xanthine dehydrogenase, that's already working at a deficit from the MOCOS down regulation. The aldehyde side is already carrying a MOCOS-related burden in a majority of autistic adults tested. LPS-driven lipid peroxidation adds more aldehydes for that same weakened AOX1 enzyme to clear. The sulfite (SUOX) side may be the only part of MoCo working well, but immune cells responding to LPS produce more sulfite. If the base nutrients that MoCo needs are lacking, now we risk SUOX not working well with an extra sulfite burden.

LPS is not introducing three new problems. It is loading three pathways that were already running at or over capacity before it showed up. That reframing mattered to me, because it changes what "removing LPS" is actually doing. It's not treating a fourth, independent issue. It's taking weight off the exact three systems the rest of this site is about.

Up to this point I've described stress affecting LPS through permeability: cortisol loosens the gut wall, so more LPS gets through. That's real, but it's only half of it. Stress also directly reshapes which bacteria are living in the gut in the first place, independent of the wall.

Stress hormones act on gut bacteria directly. This is called microbial endocrinology: catecholamines released under stress have been shown to directly affect the growth and behavior of gut bacteria, including several gram-negative species. Separately, cortisol itself correlates with gut microbial diversity. Higher post-stress cortisol has been associated with lower microbial diversity, and specific genera have been shown to track inversely with cortisol levels.

There's also a host-side mechanism. Stress triggers the gut's own epithelial cells to produce reactive oxygen species through specific enzymes (Duox2 and Nos2), and in animal studies, that host-driven oxidative response tracked closely with the resulting shift in microbiome composition, meaning the stress response itself, not just permeability, is reshaping who lives there.

Put together, stress raises LPS translocation two ways at once: it loosens the barrier, and it changes the gut population toward more of the bacteria that produce LPS to begin with. That is the piece I hadn't connected clearly before. The dysbiosis I keep describing as a baseline autistic trait is also something ongoing stress is actively feeding, not just something stress passes through.

This is part of why burnout is hard to climb out of: the elements it produces in your body become what spirals you further. The HPA axis, the body's central stress system, directly compromises the gut barrier when it is chronically active. Cortisol loosens the tight junctions, the wall turns permeable, and LPS gets into the blood. That aggravates the HPA axis, which makes more cortisol, which opens the gut further, which lets in more LPS. Addressing LPS felt like interrupting a loop that stress had created and LPS was sustaining, if not spiraling me further. I was not lowering inflammation in general. I was lowering the inflammation specific to what is found in autism. I don't agree with taking generic binders. I agree with testing for any heavy metals, but those often have specific chelation therapy that can be supervised by a doctor. I used chitosan when I knew I wasn't nutrient deficient. Activated charcoal is an option for vegans and vegetarians. Binders can also bind nutrients, and should be taken with that in mind. I don't agree with heavy herb detoxes or nonspecific binders like zeolite. That is total guesswork and can do damage if you don't know what you're binding and don't support the nutrient side.

MoCo is important because LPS doesn't just activate the immune system. It increases the metabolic work the body has to perform. LPS increases oxidative stress and stimulates nitric oxide production. In small amounts, nitric oxide is beneficial, helping regulate blood flow and support immune defense. During chronic inflammation, however, excess nitric oxide reacts with superoxide to form peroxynitrite, which damages mitochondria and oxidizes BH4. At the same time, LPS causes cells to release ATP as a danger signal, activating purinergic signaling. That ATP is eventually broken down through the purine degradation pathway, increasing demand on xanthine dehydrogenase, one of the body's molybdenum cofactor (MoCo)-dependent enzymes. LPS also promotes lipid peroxidation, generating reactive aldehydes that must be detoxified by aldehyde-metabolizing enzymes. In other words, one trigger is increasing the workload across several pathways that depend directly or indirectly on MoCo.

The documented effects of LPS are not small. LPS depletes zinc, which matters because zinc runs the enzyme that disarms LPS at the gut barrier, so it strips the body's defense when we need it most. It depletes that enzyme directly. It hits the liver first, since gut blood drains there. It crosses the blood-brain barrier. It suppresses oxytocin, a source of pleasure that we're known to be depleted of. It depletes serotonin by diverting tryptophan into quinolinic acid, an excitotoxin that overstimulates NMDA receptors. It is linked to depression, metabolic and cardiovascular disease, and endocrine disease, and it can trigger autoimmune-looking responses through molecular mimicry, where anti-LPS antibodies cross-react with human tissue. It also is connected with temperature regulation issues, which is one of the first symptoms I noticed changing. The guy I was dating noticed a dramatic body temperature change via disappointment that I wouldn't be an oven next to him through the winter.

The symptoms of chronic LPS activation are similar to burnout symptoms: lethargy and fatigue, brain fog and cognitive dulling, social withdrawal, reduced motivation and anhedonia, and then it also has altered thermoregulation, where you stop sensing temperature correctly.

Two routes besides a leaky wall.

The chylomicron route: when fat is digested it is packaged into chylomicrons, and any LPS present hitchhikes across the epithelium with them, bypassing the barrier entirely. A single high-fat meal spikes plasma LPS even in healthy people; and on a dysbiotic, already-permeable gut, the effect multiplies.

The bacterial route: high-sugar and high-fructose diets preferentially feed the gram-negative bacteria that make LPS, and processed-food additives directly increase permeability. A four-week Western-style diet raised plasma LPS by 71 percent.

Intestinal alkaline phosphatase (IAP) is the gut lining's built-in firewall: it dephosphorylates LPS, defusing it right at the wall. The catch is that IAP needs two zinc ions per active site, and LPS depletes zinc. Less zinc means less IAP, less IAP means more LPS through, more LPS means more zinc lost. A Western diet alone cut IAP activity by about 75 percent in mice with a fourfold rise in plasma LPS. Restoring zinc, fiber, butyrate, and vitamin D all raise IAP.

The link is direct. A heavy LPS burden produces sulfite because immune cells respond to LPS by produce sulfite. LPS and sulfite drain the same supplies, zinc and glutathione among them, and the zinc loss is its own trap through IAP. And it doesn't stop at sulfite: the ATP that LPS releases from macrophages increases nitric oxide production through the same purinergic receptors covered earlier in this site, which is one more way an LPS burden is not just a gut or immune issue, it's feeding the purine danger-signal cascade at the same time it's feeding the sulfite problem. Focusing on reducing the LPS burden was a dramatic shift in my comfort level because it impacted the other areas simultaneously. That reduced my stress further, and it became a positive cycle.

The mast cell connection I described earlier came through sulfite specifically. LPS has its own separate route into mast cell activation that doesn't depend on sulfite at all.

Mast cells carry TLR4, the same receptor that detects LPS everywhere else in the body. When LPS binds TLR4 on a mast cell, it triggers production of inflammatory cytokines directly: IL-1 beta, TNF-alpha, IL-6, and IL-13. Separately, LPS also primes the mast cell to degranulate harder when it does get triggered by something else, by increasing calcium entry through store-operated calcium channels. In one study, LPS exposure alone didn't necessarily cause degranulation on its own, but it dramatically increased degranulation once the mast cell was then exposed to an allergen or antigen on top of it. The more my health declined, the more mast cell symptoms I had. While I still feel a little puffier than I was when I was younger, I don't have mast cell symptoms anymore.

My goal was never to cure autism or repair a gene. It was to restore my health, which I began to realize meant to reduce the amount of work my body had to do every day metabolically. If stress increased purines, I reduced stress with habits like breathwork and boundaries. If gut dysbiosis increased sulfites and aldehydes, I started figuring out what addressing my diet looked like with that purpose in mind. If nutrient deficiencies made those enzymes work less efficiently, I corrected the deficiencies. Instead of trying to force the kitchen to work harder, I tried to stop overwhelming it and provided resources so it could catch up.

A population whose most common metabolic snag sits on the body's energy pathways being prone to burnout makes sense to me. Whether those purine problems are from the MOCOS down regulation or independent of it, they exist for a chunk of us. More profound genetic conditions are often much more complex, so please understand that when I say this information is geared towards burnout. Do I think this applies to a wider audience? Potentially. But I think it gets more complicated the more profound someone is. My goal was different. I was looking for what we all had in common in hopes of stopping a spiralling decline. When you ask a different question, you find a different solution. I care about redox and energy pathways now. I care about gut dysbiosis and eating to improve my diversity. I care building a life I don't have to survive.

Here is the most important part when it comes to stress. When you have energy, life is less stressful. That doesn't mean a bad job or mean people feel good. It means you have the reserves to handle it and potentially address the situation differently. The closer you are to regulation, the easier life is, even when it's hard.

Because the MOCOS study was so small, I had to go on side quests to figure out how much weight I was going to give MoCo. Initially that was seeing if the markers of each of those enzyme dysfunctions showed up in autism, which they did, at least enough to satisfy me. Later on, I compared it with the wider literature. MoCo seemed to have its toe in numerous areas. That being said, it was not 100% of the participants that had the down regulation. There's several paths to Rome, but this one seems to be involved in a portion of the journey. This is not every idea out there, but it was enough for me to take MOCOS seriously considering how small of a study it was. For funsies, here they are:

BH4

The oxidative stress from a reduced MoCo enzyme capacity oxidizes BH4 to BH2. That uncouples the four BH4-dependent conversions, stalling dopamine, serotonin, and melatonin synthesis and flipping nitric oxide synthase into a superoxide source. Oxidative stress can come from multiple directions, but this is one of the large ones. Damaging purine production also damages BH4 production because BH4 comes from GTP, part of the purine de novo pathway.

Microglia

Microglial activation runs on purinergic (ATP and adenosine) signaling and produces an oxidative and aldehyde load, the same currencies the MoCo enzymes handle. So you need this enzyme working well for microglia activation.

Excitatory / inhibitory imbalance

Two routes converge: SUOX-limited sulfite forms S-sulfocysteine, an NMDA agonist that drives excitotoxicity and calpain-mediated gephyrin destruction (loss of the GABA and glycine brakes), and gephyrin's dual role links the MoCo build itself to inhibitory-synapse scaffolding. Gephyrin is the single protein where the cofactor system and E/I balance meet.

Aldehyde toxicity

AOX1, which needs MOCOS sulfuration, is a primary clearance route for reactive aldehydes, so reduced AOX1 leaves aldehydes to damage proteins, membranes, and DNA. This is one of the two airtight MOCOS-downstream branches, alongside purines. There is a fantastic paper talking about the role of aldehydes in autism and claiming it fits every known theory up to the publishing of that paper. A large number of the genes needed to handle aldehydes in the body are also thought to be connected with autism.

Mast cell activation

A MoCo bottleneck reduces the body's ability to convert sulfites, and sulfite triggers non-IgE degranulation through NADPH-oxidase-driven ROS.

Essential fatty acids

Membrane PUFAs are especially vulnerable to oxidative lipid peroxidation, producing reactive aldehydes. Reduced AOX1 activity may increase the persistence of some aldehydes, linking membrane integrity indirectly to MoCo-dependent aldehyde metabolism.

Glyphosate

Stephanie Seneff's work connecting glyphosate (Roundup) to autism has received substantial criticism. She proposed that glyphosate takes the place of glycine and contributes to autism, but when I dug more, I found a different potential connection that interested me. One component of her proposed mechanism involves PIN1. PIN1 regulates gephyrin conformation at inhibitory synapses, and gephyrin is also the enzyme responsible for the final steps of molybdenum cofactor (MoCo) biosynthesis. Whether PIN1 also regulates gephyrin's MoCo function has not been investigated. The known biological links exist, but the proposed PIN1-MoCo connection has not yet been experimentally tested. She leans more on the glycine side, but this is a connection that could potentially exist and links to MoCo. I'm not saying it is the cause of the MOCOS down regulation. I'm saying it could mimic a side of what is known to be happening already. There are multiple genetics and multiple paths to Rome in autism.

PANS, PANDAS, and post-viral

Infection and immune activation load all three MoCo queues at once. Xanthine oxidase is upregulated during inflammation, so purine substrate rises while the enzyme handling it also becomes a ROS source. Cell stress and immune cell turnover push ATP out as a danger signal, and immune cells responding to endotoxin generate sulfite. A system already running at its clearance ceiling has no reserve for that surge, which may be part of why the onset is abrupt and the resolution is slow. The overlap here is capacity, not cause. PANS and PANDAS involve anti-neuronal antibodies and basal ganglia targeting that this framework does not explain. What it may explain is why some of us flare harder and recover slower from the same infectious hit. The BH4 entry above compounds this, since PANDAS work implicates dopamine receptor antibodies and BH4 oxidation stalls dopamine synthesis from the other direction.

Candida

Candida produces acetaldehyde directly through its ethanol metabolism, placing it upstream of AOX1 and the aldehyde queue rather than merely alongside it. It also produces arabinose, which forms adducts on lysine residues, the same binding chemistry that makes aldehydes damaging. The loop runs both directions: purines feed candida, stress raises purine turnover, candida raises aldehyde load, and aldehyde load competes for the same sulfurated cofactor that clears purines. That is the mechanism behind the antifungal case reports easing autism features without any of those studies naming a cofactor.

Clostridia

Clostridia has been proposed to be a factor in autism. Clostridia loads two MoCo queues at the same time. On the aldehyde side, clostridial fermentation generates acetaldehyde and related reactive aldehydes directly in the gut lumen, feeding the same AOX1 clearance route that candida burdens and that MOCOS downregulation limits. On the sulfur side, clostridial phenolic metabolites, p-cresol and HPHPA among them, are cleared by sulfation, and every molecule consumes sulfate, the exact pool a limited SUOX is already failing to replenish. So one overgrowth drains sulfate from the demand side while sulfite conversion underdelivers on the supply side, and simultaneously adds to the aldehyde queue that competes for the same sulfurated cofactor. It also competes for the sulfation capacity measured in the 90s paracetamol studies, which means the phenol findings and the clostridia findings may be one finding seen from two angles. Worth distinguishing sulfate-reducing gram-negatives like Desulfovibrio, which sit on the LPS side as well. There's a lot of possible questions here.

I used to think of gut health as "take a probiotic." What actually changed things was realizing that different types of fiber feed different bacteria, so a diverse gut needs a diverse fiber intake, not just more fiber in general.

• Inulin and FOS (found in chicory root, onions, garlic, asparagus): selectively feed Bifidobacterium and Lactobacillus
• GOS (found in legumes): also feeds Bifidobacterium and Lactobacillus, through a different route
• Resistant starch (found in cooked-and-cooled potatoes and rice, green bananas, legumes): feeds a broader range of butyrate producers, and different resistant-starch sources favor different bacterial genera. Beans and lentils in particular were shown in one study to reduce Proteobacteria specifically, which matters given how often Proteobacteria shows up as overgrown in autism gut testing.
• Pectin (found in apples, citrus): tends to preserve overall bacterial diversity better than most single fibers, and feeds Bacteroides
• Beta-glucan (found in oats): supports similar diversity-preserving effects alongside pectin

The pattern that mattered most to me: no single fiber does everything, and some (like plain inulin taken alone) can even narrow diversity if it's the only fiber source, by over-feeding a couple of species at the expense of others. Including multiple fiber sources, not just "eating more fiber," is what actually builds diversity.

Autistic children have been found to have abnormally high total B6 in their blood, not low. That sounds like the opposite of a deficiency, until you look at what's actually being measured. Total B6 includes the inactive forms, pyridoxine and pyridoxal, before they've been converted into PLP, the only form the body can actually use as a cofactor. The same research, and studies before it, found that this conversion step, run by an enzyme called pyridoxal kinase, has unusually low activity in autism, and that PLP itself runs low even while total B6 sits high. So the numbers on a blood panel can look reassuring while the form that actually does anything is the part that's short. This is the same pyridoxal kinase I mentioned earlier gets indirectly slowed by sulfite. That being said, it's not been explored why that conversion isn't happening well in us.

PLP is the required cofactor for glutamate decarboxylase, the enzyme that converts glutamate, an excitatory neurotransmitter, into GABA, the primary inhibitory one. One of the two forms of this enzyme, GAD65, exists mostly in an inactive state until PLP binds and switches it on. Reduced GAD expression has specifically been found in autistic brain tissue, and glutamate decarboxylase is named directly in the autism research as relevant to glutamate excitotoxicity. If PLP is short, this enzyme can't convert glutamate to GABA as fast as it needs to, which tips the excitatory-to-inhibitory balance toward excitatory, toward excitotoxicity, not because glutamate itself is the problem, but because the enzyme meant to convert it away is running on a thin supply of its cofactor.

PLP is also the cofactor for SHMT, the enzyme that pulls one-carbon units out of serine and hands them to folate in the first place. This is the connection I wasn't sure existed when I started asking the question, and it turns out it does: folate can't be used for anything downstream, purine synthesis, methylation, none of it, until SHMT does this handoff, and SHMT doesn't run without PLP. So a B6 shortfall doesn't just sit next to a folate problem. It can cause one, even when folate intake itself is fine.

And PLP is the cofactor for CBS, the enzyme that runs transsulfuration toward cysteine and glutathione, which I've already covered in the folate and glutathione sections.

Three separate systems, glutamate/GABA balance, folate activation, and transsulfuration, all bottleneck at the same single conversion step. That's what made B6 register as a common denominator rather than just one more nutrient on a list.

I did supplement B6 in the form of P5P. P5P is the better form to take because it doesn't require conversion, but having too much can cause neuropathy. I do not take it reguraly, but it was a part of my healing.

There is a study showing that we are what's called functionally deficient in B2. It's one that needs to be followed up on as there could be bias with the original study.

Riboflavin, the form of B2 in food and most supplements, isn't usable on its own. It has to be converted in two steps. The first step, run by an enzyme called riboflavin kinase, turns riboflavin into FMN, and that step specifically requires zinc. The second step, run by FAD synthetase, turns FMN into FAD, the form most flavin-dependent enzymes actually use, and that step specifically requires magnesium. Both steps also cost ATP.

That matches the pattern already all over this site. Zinc is depleted by LPS. Magnesium is depleted by chronic stress. ATP is the exact currency this whole framework keeps returning to as being in short supply. So even a body getting adequate dietary B2 could still be undersupplying active FAD if zinc, magnesium, or ATP are the limiting factor instead, which would look identical to a straightforward riboflavin deficiency on paper while actually being a cofactor problem underneath it.

R5P is the form of B2 that does not need conversion.