← Sage hub · Surfer

Sage — Serotonin: Bioenergetic Flux Manager

The Metabolic Hypothesis of Serotonin

Any methylene-blue milligram language below is information from the essay — not a recommendation to self-administer and not a clinical protocol.

Bioenergetic Flux Management in a Biology Tuned for Ancestral Entrainment

By Val Zudans, MD EnSurf Frugal Philosophy Essay • September 2026

Disclaimer. This essay is a scientific and philosophical hypothesis for education and discussion. It is not medical advice, not a diagnosis, and not a treatment protocol. It does not establish a doctor–patient relationship. Methylene blue, serotonergic drugs, psychedelics, and metabolic interventions can cause serious harm, including serotonin toxicity, hemolysis in G6PD deficiency, withdrawal syndromes, and pregnancy-related risk. Do not start, stop, combine, or taper any medication or supplement based on this text. Decisions about pregnancy, antidepressants, methylene blue, and metabolic care require a qualified physician who knows the individual case. Statements about methylene blue amounts (Window A, 8–25 mg) are information discussed herein; they are not a recommendation to self-administer. The author and EnSurf assume no liability for actions taken from this essay.

Serotonin is not a happiness molecule. That sentence is short enough to be dismissed as contrarian branding. It is not. It is the only framing that survives contact with the molecule's actual distribution, evolutionary history, and pharmacology.

Most of the body's serotonin is in the gut. Platelets take up and store serotonin from the blood. The molecule was first isolated as enteramine—an amine of the intestine—and then renamed serotonin because it was a serum factor that changed vascular tone. Neither name points at mood. Both point at coordination of flow: motility of lumen contents, tone of vessels, partitioning of blood, timing of oxygen delivery. The later popular idea that serotonin is a "happy chemical" was not a discovery. In this essay's reading, it grew up alongside a class of drugs whose main action is to block the serotonin transporter.

I respect the clinical utility of serotonergic medicines in selected patients. Some people get better. Some get worse. Many get a delayed, incomplete, and state-dependent change that does not look like replacement of a missing neurotransmitter. After years of clinical observation, mitochondrial work, and the EnSurf project of treating ATP-generating capacity as the upstream variable in chronic disease, I find the mood-chemical story incomplete in the same way the purely genetic story of cancer is incomplete. It describes a dashboard reading and a useful lever. It does not describe the engine.

I propose an upstream reframing: serotonin is a whole-body bioenergetic flux manager. Its evolved job is to coordinate the production, storage, mobilization, distribution, and utilization of energetic resources—especially oxygen and oxidizable carbon—so that mitochondrial ATP generation can be matched to local and systemic load. When oxygen supply, fuel quality, or mitochondrial throughput is threatened, serotonin may rise or be redistributed as an acute compensatory signal. When the mismatch becomes chronic, I propose that the same signal can lock the organism into a compensatory mode we later name depression, anxiety, fatigue, gut dysmotility, pulmonary hypertension, platelet dysfunction, fetal behavioral immaturity, SSRI withdrawal, or "brain zaps."

This essay presents that hypothesis, shows why it is more coherent and more parsimonious than the serotonin-deficiency theory of mood, and places the molecule inside the same EnSurf frame used for cancer, atherosclerosis, aging, and diabetes: disease as chronic compensatory mode in a biology tuned for ancestral entrainment. It does not reject synapses, receptors, or psychiatry. It contextualizes them inside flux.

The Standard Story and Why It Cannot Carry the Weight Placed on It

The dominant public and much of the clinical story runs as follows. Depression is caused by low synaptic serotonin. SSRIs block the serotonin transporter (SERT), raise synaptic 5-HT, and thereby restore mood. The delay of two to six weeks is inconvenient but is treated as a black box of "receptor adaptation." Side effects, sexual dysfunction, emotional blunting, weight gain, platelet serotonin collapse, neonatal adaptation syndromes, and withdrawal phenomena are treated as separate problems, each with its own mini-explanation.

This story has real institutional power. It organized decades of drug development, medical education, and patient explanation. It is also empirically thin in the place it claims to be strongest.

A 2022 umbrella review in Molecular Psychiatry (Moncrieff and colleagues) found no consistent evidence that depression is associated with lowered serotonin concentration or activity. Measures of the metabolite 5-HIAA, plasma serotonin, receptor binding, transporter binding, and tryptophan depletion do not add up to a simple deficiency. High-quality genetic work does not support a causal SERT-to-depression pathway of the kind the marketing implied. Meanwhile, the direct pharmacological action of an SSRI—occupying SERT throughout the body within hours—does not produce antidepressant effects on that same timescale. If the problem were a missing synaptic drop of 5-HT, the temporal mismatch would be surprising. It is not surprising if SERT occupancy is a whole-body perturbation of flux management and the later "therapeutic" change is a compensatory reorganization of energy allocation.

Paul Andrews and colleagues stated the paradox cleanly in 2015: serotonin transmission appears elevated in several depressive phenotypes, including melancholia; SSRIs further raise extracellular 5-HT and often worsen symptoms acutely; delayed improvement, when it occurs, is better read as the brain's attempt to restore energy homeostasis after that perturbation. Their energy-regulation hypothesis is the closest published ancestor of what follows. The EnSurf contribution is to push the same logic one layer deeper: not only energy allocation among organs and behaviors, but energy generation at the mitochondrion, and oxygen as the rate-limiting electron acceptor for that generation.

The standard story also cannot explain the geography of the molecule. A brain-first happiness chemical should not be stored primarily in enterochromaffin cells and platelets. It should not be the precursor of mitochondrial melatonin. It should not constrict poorly ventilated lung segments, drive gut peristalsis after a meal, or empty from platelets when SERT is blocked. Those facts are not side quests. They are the main plot.

The Engine, Not the Dashboard

EnSurf essays start from a single physical constraint. Multicellular life is an ATP problem. Food electrons must be passed down the electron transport chain, protons pumped across the inner mitochondrial membrane, and ATP synthase paid in the resulting chemiosmotic force. Oxygen is the terminal acceptor. When oxygen, fuel, cofactors, membrane integrity, or ETC complexes fail, ATP-generating capacity falls relative to load. The cell then does what ancestral cells have always done: it shifts toward faster, less efficient cytosolic ATP production, changes vascular and behavioral allocation, and broadcasts stress signals so that the organism can re-partition scarce redox capacity.

In this hypothesis, much of what modern medicine calls "disease" is this compensatory mode left on. Glycolytic rescue in a tumor is one face. Insulin resistance that keeps glucose in the blood while mitochondria starve for usable fuel is another. Atherosclerotic inflammation in a macrophage that cannot meet ATP demand is another. Depression, in this frame, is not a missing smile molecule. It is a whole-organism reallocation when the engine cannot cover the current invoice.

Serotonin sits at the switchboard of that invoice.

It is uniquely placed to do this work. No other small molecule simultaneously:

• is synthesized from an essential, diet-limited amino acid (tryptophan), so its production is already a statement about nutritional and inflammatory state;
• is made in the gut in response to mechanical and chemical load, where postprandial oxygen consumption is fierce;
• is packed into platelets, which patrol the circulation and release 5-HT at sites of activation, hypoxia, and endothelial distress;
• is cleared by SERT in gut, lung, placenta, adrenal, brown fat, and synapse;
• acts on a receptor family diverse enough to constrict or dilate, speed or slow gut, suppress or permit appetite, drive or restrain thermogenesis, and change mitochondrial membrane potential;
• is taken into cells and mitochondria, where it may support respiration under relative hypoxia and serve as the precursor to melatonin, which neurons can make inside mitochondria;
• is oxidized on the outer mitochondrial membrane by MAO, tying its disposal directly to mitochondrial geography.

That is not the anatomy of a mood droplet. That is the anatomy of a flux manager.

Flux, Defined

By bioenergetic flux management I mean the real-time matching of four streams:

1. Oxygen delivery to the mitochondrial inner membrane.
2. Carbon fuel delivery (fatty acids, ketones, glucose, amino acids) in a form the ETC can use.
3. Electron throughput and proton-motive force, including the option to uncouple when heat or ROS control is more valuable than ATP.
4. ATP expenditure across competing programs: motility, immunity, growth, reproduction, thermogenesis, rumination, social vigilance, fetal movement, platelet activation.

Serotonin participates at every layer. The same molecule can look like a vasoconstrictor in one bed, a motility hormone in another, an insulin secretagogue in a third, a suppressor of brown-fat thermogenesis in a fourth, and a synaptic modulator in a fifth, because those are not separate jobs. They are local solutions to one problem: where should scarce oxidative capacity go right now?

Oxygen as the master constraint

Oxidative phosphorylation is oxygen-limited in more of ordinary life than the word "hypoxia" suggests. The gut after a meal may become relatively hypoxic because enterocytes burn oxygen to absorb and process fuel. Exercising muscle creates regional oxygen debt. The pregnant uterus and placenta live near the edge of fetal oxygen supply. Poorly ventilated lung units must not steal blood from well-ventilated ones. The newborn must switch from placental gas exchange to pulmonary gas exchange in minutes. Aging tissues accumulate mitochondrial inefficiency and call for more perfusion than their vessels can politely supply.

In this setting, free serotonin may act as a traffic cop. In the lung it may help constrict poorly oxygenated segments so blood is sent to better-oxygenated ones. Hypoxia may activate platelets and raise free 5-HT; that signal could reach carotid body and brainstem circuits that drive ventilation and sympathetic outflow. In the gut, enterochromaffin release of 5-HT speeds transit and secretion, which may be a way of not letting a stagnant, oxygen-hungry bolus sit against mucosa. In the fetus, platelet and circulating 5-HT may help coordinate movement and tone in a body that does not yet breathe.

Chris Masterjohn has called serotonin the firefighter of the mitochondrion and melatonin its guardian angel. The metaphor is useful if kept precise. Serotonin is not a chronic tonic that should bathe the system. It is an emergency and load-matching signal. It arrives when oxygen or throughput is threatened. Melatonin, synthesized from the same tryptophan spine inside mitochondria, is the standing watch: antioxidant, circadian, mitophagy-adjacent. Confusing the firefighter with a daily vitamin is how one turns a useful acute signal into a chronic compensatory lock.

Carbon and the two faces of serotonin

Peripheral and central serotonin often look as if they disagree. Central 5-HT tends to suppress appetite and raise energy expenditure. Peripheral 5-HT, mostly gut-derived, tends in animal studies to promote nutrient absorption and storage: pancreatic 5-HT supports β-cell function and insulin release; hepatocyte signaling can favor lipogenesis and, during fasting, gluconeogenesis; adipocyte 5-HT can suppress UCP1-dependent thermogenesis. This is not a contradiction. After a meal, the correct flux decision is to accept carbon, store what the engine cannot burn now, and not waste heat in brown fat while the gut is paying an oxygen tax. In a fasted or cold or high-demand state, the correct decision is the opposite.

SERT is the valve between those compartments. Block SERT everywhere and you do not "boost happiness." You freeze the valve. Platelets can no longer pack 5-HT. Free 5-HT rises in some compartments and collapses in the storage pool. Gut signaling is altered. Lung handling of 5-HT is altered. Brown adipose tissue in humans, which expresses SERT and is suppressed by extracellular 5-HT at 5-HT2B, loses a protection against serotonergic braking of thermogenesis. That last point is no longer theoretical: sertraline has been shown to reduce human brown-fat activation, and SSRI-treated patients can lack the usual room-temperature FDG uptake in BAT. Reduced brown-fat activity may contribute to weight gain and metabolic flattening on SSRIs. In this frame, they are flux decisions being made with a jammed valve.

Mitochondria: outside-in and inside-in

Serotonin has two mitochondrial faces, and they must not be collapsed into one slogan.

Outside the cell, 5-HT2A receptor signaling via SIRT1–PGC-1α can drive mitochondrial biogenesis and raise respiratory capacity in rodent neurons. An activating extracellular pulse says: this cell is being asked to work; build or spin up engines.

Inside the cell, serotonin and its metabolite melatonin may support respiration when oxygen is relatively scarce. MAO on the outer membrane ties disposal to that geography. SSRIs, being SERT ligands that also enter cells, can stimulate some of the stress-adaptation machinery while later acting as mitochondrial stressors in their own right. That combination—chronic occupancy of an intermittent system, plus possible direct organelle toxicity at sufficient exposure—is, in this hypothesis, the shape of dependence. It is why withdrawal is not the mirror image of a missing happy drop. It is an engine that has been running on a jammed compensatory circuit and then suddenly loses the jam.

Platelets: The Circulating Flux Buffer

Most circulating serotonin is not free. It is taken up and stored by platelets via SERT. Platelets do not synthesize 5-HT; they harvest it. That architecture is the point. The circulation carries a buffered reservoir that can be released where endothelium is activated, where shear is wrong, where oxygen is low, where clotting is beginning. Release is local. Reuptake refills the buffer.

Maternal platelet serotonin falls on the order of eight-fold compared with non-users. Infant cord-blood platelet serotonin falls on the order of four-fold. These are not subtle biomarker wiggles. They are emptying of the circulating buffer. Depression without medication does not produce the same drop. That dissociation matters. It means the platelet signal is a drug bioeffect, not a depression signature.

Salisbury and colleagues (2024, Neuropsychopharmacology) showed that lower maternal and infant platelet 5-HT tracked more immature, fidgety fetal movement on ultrasound. The study is small and observational. It does not measure placental or fetal oxygen directly. It should not be over-read as proof that SSRIs asphyxiate fetuses. It should be read as what it is: a bioeffect marker of SERT inhibition that co-travels with a change in fetal motor quality, in an organism whose entire energy economy is still placental and therefore hypoxia-adjacent.

In the flux frame this is expected. Fetal movement is expensive. Coordinated, mature movement is a sign that oxygen and ATP can be spent on patterned work rather than on jitter. Emptying the 5-HT buffer removes a coordinator of vascular tone, platelet behavior, and possibly direct mitochondrial support during a period when the organism cannot yet recruit lungs. Whether the dominant pathway is vascular, neural, platelet, or mitochondrial is an empirical question. The hypothesis predicts that the relevant axis is oxygen-ATP matching, not "fetal mood."

Aspirin is the control experiment the thread asked for. Low-dose aspirin has mixed, modest signals for mood and a much stronger anti-inflammatory story. It does not empty platelet 5-HT stores the way SSRIs do. It acetylates COX-1 irreversibly and reduces thromboxane-dependent release during aggregation. Parent aspirin has a plasma half-life of minutes; the antiplatelet effect lasts the platelet's life. If mood were a function of dumping platelet serotonin, aspirin should look like a dirty SSRI. It does not. That is consistent with keeping anti-inflammatory and flux-buffer functions distinct, and argues against a platelet-dump theory of antidepressant action.

Gut: The First Hypoxic Reactor

The gut is where serotonin belongs if flux is the job. Eating may be an oxygen event. Mucosa and smooth muscle increase metabolic rate. Local PO2 falls. Enterochromaffin cells release 5-HT in response to stretch and nutrients, and possibly to low oxygen. Motility and secretion rise. The bolus moves. Stasis—an invitation to local anoxia and bacterial overgrowth—is avoided.

This may be why SSRI gut effects are early and common: nausea, diarrhea or constipation, cramping. The drug hits the largest 5-HT compartment first. It is also why "the gut-brain axis" is often described backwards. In this hypothesis, the gut is the primary serotonergic organ because it is the primary daily site of regional oxygen-fuel conflict, and the brain borrowed a coordinator that already existed for that conflict.

Tryptophan routing makes the same point. Most tryptophan does not become serotonin. It becomes protein or it enters the kynurenine pathway, which is induced by interferon and other inflammatory signals. Infection and chronic inflammation therefore steal precursor from the 5-HT route while increasing the need for flux management. Long COVID work from Maayan Levy and colleagues found reduced serotonin with impaired tryptophan uptake, lower platelet counts, and increased turnover under type I interferon—then linked that reduction to vagal and cognitive features. In the flux frame this is not "viral sadness." It is an inflammatory tax on the precursor and the platelet buffer, levied at the exact moment mitochondria and oxygen delivery are already under load.

Brain: Expensive Tissue, Secondary User

The brain is a small fraction of body mass and a large fraction of resting ATP use. It is metabolically spoiled: it expects glucose or ketones on time, oxygen without interruption, and the right to commandeer supply when threatened. Serotonergic nuclei in the raphe project widely not because mood is everywhere, but because a flux manager must be able to change the spending pattern of an expensive organ.

Andrews's account of melancholia is useful here. Sustained rumination is metabolically costly. It may be paid, in part, by aerobic glycolysis in astrocytes and by reallocation away from growth and reproduction. Elevated serotonergic tone in that state is not a failed attempt at happiness. It is the coordinator of the reallocation. Acute SSRI treatment, by further raising extracellular 5-HT, can intensify the very processes—rumination, emotional flattening, sleep disruption, gut distress—that patients experience as worsening. Later improvement, when it happens, tracks compensatory receptor and network changes that try to restore a livable energy budget, not a restored "level" of a happy chemical.

This may be why SSRI effects seem state-dependent. A flux manager cannot have one sign. In a well-fed, well-slept, hormonally intact organism with spare mitochondrial capacity, raising extracellular 5-HT is a different intervention than raising it in a sleep-deprived, inflamed, insulin-resistant, hypoxic-at-the-margins organism. The first may look like a mild stress that the system absorbs. The second may look like locking the firefighter in the hallway.

The brain also reveals the oxygen special case. Altitude and hypoxia may affect mood and serotonergic physiology. Serotonergic neurons take part in the control of breathing. If the molecule's deep job is matching oxidative capacity to demand, the brain's serotonergic system should care about breathing.

Onset, Tolerance, Withdrawal, and Brain Zaps

The temporal profile of SSRIs is the hypothesis's strongest clinical test.

Hours to days. SERT occupancy is high. Platelet 5-HT begins to fall. Gut symptoms appear. Sleep architecture changes. Anxiety or flattening can rise. This is the direct flux perturbation: valve jammed, buffer emptying, extracellular 5-HT up in some compartments, intracellular packaging down.

Weeks. Autoreceptors desensitize. Gene expression shifts. Mitochondrial and network compensations accumulate. Some patients report improvement. In the flux frame this is not "the serotonin has finally arrived." It may be the system accepting a new, less flexible set point and turning down the alarms that the initial jam produced. Improvement that depends on a jammed valve remaining jammed is the definition of a compensatory lock.

Tolerance and tardive phenomena. A system whose job is intermittent matching cannot be held at constant occupancy without losing dynamic range. Sexual function, which is an expensive, optional program, is among the first to be deprioritized. Emotional blunting is reduced allocation to costly affective processing. Weight gain is the peripheral storage program running with thermogenesis braked. In this frame, these are not random toxidromes; they may be the predictable budget cuts of a flux manager stuck in the on position.

Withdrawal. In this frame, the valve is suddenly unjammed. The storage pool is still depleted. Receptor and mitochondrial adaptations are still those of the jammed state. The result may be a transient mismatch between signal and engine, felt as dizziness, electric-shock sensations, instability, flu-like feelings, rebound anxiety, sensory hypersensitivity. "Brain zaps" read, in this frame, as a nervous system that has been running on altered volume transmission and altered mitochondrial calcium/ATP handling, now asked to resume synaptic timing without the buffer or the set point it adapted to. Hyperbolic tapering exists because occupancy-to-effect is nonlinear; flux systems usually are.

None of this requires denying that some people need and benefit from these drugs. It requires telling the truth about what the drugs do. They are not serotonin replacements. In this hypothesis, they are chronic occupancy of a whole-body flux valve, with delayed compensatory remodeling, possible mitochondrial cost, and a withdrawal physiology that looks like engine instability.

Ancestral Entrainment and Why the Signal Misbehaves Now

We are born with biology optimized for intermittent fuel, solar and seasonal light, obligatory movement, and a microbial and thermal world that no longer exists in the packaged form we inhabit. Serotonin and melatonin are part of that entrainment. Daylight, movement, and feeding pulses should raise and lower serotonergic tone. Night should convert the same tryptophan spine into mitochondrial melatonin. Fasting and cold should change peripheral 5-HT's permission for thermogenesis and lipolysis. Social hierarchy and physical threat should reallocate, through the same system, between vigilance and growth.

Modern mismatch holds the firefighter in a smoky building.

• Continuous insulinogenic feeding keeps peripheral storage programs on and rarely asks BAT or muscle mitochondria to run their preferred fat/ketone base load.
• Indoor light and delayed sleep may disrupt the serotonin-to-melatonin handoff.
• Physical inactivity removes hormetic ROS and AMPK signals that would have increased mitochondrial density, so ATP-generating capacity may fall while the flux manager is asked to coordinate a weaker engine.
• Chronic low-grade inflammation can divert tryptophan into kynurenine.
• Pharmacologic SERT occupancy, when added on top, converts an intermittent matcher into a constant.

In this hypothesis, depression in this ecology is often the experience of an organism whose engine cannot cover the cognitive and social invoice, broadcasting a serotonergic compensatory state that psychiatry then treats as a primary chemical deficit. Sometimes the compensatory state is worth interrupting with a drug. Often the more coherent move is to restore the conditions the matcher evolved to serve: fuel the mitochondria were trained on, circadian alignment, strength and aerobic hormesis, sleep as anabolic mitochondrial night shift.

That is not romantic ancestral theater. It is matching the assembly line to the product it was selected to make. In other EnSurf essays I have argued that function determines form more than the reverse: the bioenergetic persistence of the line determines the body, and the dashboard readings follow. Serotonin is a dashboard and a line supervisor. Occupying SERT changes the supervisor's clipboard. It does not rebuild the line.

Aspirin, Uncoupling, and Other Apparent Puzzles

Several loose ends in the originating thread fall out of the same frame.

Aspirin and mood. Anti-inflammatory lowering of IL-6, TNF-α, and prostaglandin drive might reduce the kynurenine tax and the mitochondrial inflammatory load. That could improve energy capacity and, secondarily, mood. It is not an SSRI-like emptying of platelet 5-HT. Different lever, overlapping invoice.

Uncoupling. Mild mitochondrial uncoupling trades ATP yield for heat and ROS control. Serotonin's relationship to BAT and UCP1 is the organism-level version of that trade. Peripheral 5-HT that suppresses UCP1 is a decision to keep coupling tight and store carbon after a meal. Central and sympathetic programs that permit thermogenesis are the opposite decision. Pharmacologic uncouplers and serotonergic drugs both move this trade; they should not be narrated as unrelated.

Psychedelics and intracellular receptors. Some 5-HT2A agonists reach intracellular receptor pools that circulating serotonin may not. If extracellular 5-HT is the ordinary flux broadcast and intracellular access is a different key, psychedelics may reset network and mitochondrial programs without chronically jamming SERT. That would explain a different temporal profile: intense occupancy, then release, rather than a jammed valve for years. The mitochondrial literature on serotonergic psychedelics is early and should be held as a testable branch of the hypothesis, not a slogan.

Methylene blue. The EnSurf methylene blue hypothesis discusses low-dose methylene blue as an ETC-supportive tool: 8–25 mg (Window A), not a daily tonic; information, not a recommendation, and any use is physician-guided. It is also a potent reversible MAO-A inhibitor. That dual binding is not an embarrassment to the hypothesis. It is evidence that the ETC, MAO, and 5-HT sit in one neighborhood. You still do not occupy that neighborhood casually, and you do not stack it with SERT drugs. Engine tools and flux-valve tools interact because they were never separate systems. Details are in the methylene-blue addendum.

What the Hypothesis Explains That the Mood Story Does Not

A good upstream frame should collapse many "side effects" into one mechanism.

Parsimony is not a decoration. It is the reason to prefer this frame. One upstream job—match oxidative capacity to load—generates the geography, the pharmacology, the timeline, and the comorbidities. The mood story requires a primary psychiatric molecule plus a long list of unrelated peripheral accidents.

Falsifiable Predictions

A hypothesis that cannot die is not a hypothesis. Several predictions follow.

1. Bioenergetic state will moderate SSRI response more than baseline mood scores. Patients with low mitochondrial reserve (poor exercise capacity, insulin resistance, sleep apnea, high inflammation, low BAT activity) should show more early worsening, more metabolic side effects, and more withdrawal instability than metabolically flexible patients at the same SERT occupancy.

1. Platelet 5-HT will track flux bioeffect, not depression per se. Untreated depression should not empty the buffer; SERT occupancy should. Restoring the buffer after taper should lag symptom change if mitochondrial and receptor adaptations are the slow variables.

1. Oxygen challenges will move serotonergic physiology in a coordinated way. Hypoxia, apnea, anemia, and placental insufficiency should look like high demand on the firefighter. Supplemental oxygen, treatment of apnea, iron repletion, and improved placental function should lower that demand without an antidepressant.

1. Interventions that raise ATP-generating capacity without occupying SERT—resistance training, circadian light, ketogenic or low-insulinogenic fuel in appropriate patients, sleep restoration, reduction of interferon-driven kynurenine flux—should improve a subset of "serotonergic" syndromes. Where they fail, the remaining cases are the proper target for pharmacologic valve tools.

1. Human BAT and thermogenic flexibility will fall on SSRIs and recover after washout, proportional to peripheral 5-HT exposure, not to mood change.

1. Withdrawal intensity will correlate with mitochondrial and sensory-timing markers, not only with receptor occupancy curves. If zaps are engine-timing instability, ETC-supportive and sleep-circadian measures should blunt them better than mood explanation alone. (This is a research prediction, not a treatment protocol.)

1. Cybrid and cell-autonomous work should show that intracellular 5-HT/melatonin availability changes hypoxic respiration and ROS handling, while chronic SSRI exposure reduces spare respiratory capacity in a washout-reversible then washout-fragile pattern.

If platelet 5-HT collapse is irrelevant to fetal motor quality in larger samples with direct oxygen and outcome measures, that branch dies. If SSRI benefit is shown to be independent of any measurable change in energy allocation or mitochondrial physiology, the upstream claim dies. The hypothesis is willing to die. The mood story has been less willing.

Therapeutic Order, Not Anti-Psychiatry

If serotonin is a flux manager, the order of operations changes.

First, restore the conditions the manager evolved to serve. Proper human diet: fuel mitochondria with the fats, proteins, and micronutrient cofactors they were selected on, and stop flooding a weak engine with continuous glucose-insulin traffic. Circadian alignment: morning light, darkness at night, feeding windows that permit a melatonin-mitochondrial night shift. Strength and aerobic work: one of the most reliable, cheap ways to raise mitochondrial density and quality in living people. Sleep and airway: oxygen is the terminal acceptor; in this frame, treating apnea may act as serotonergic therapy by another name. Inflammation: lower the kynurenine tax.

Second, use serotonergic drugs as what they are—chronic occupancy of a whole-body valve—when the invoice is still unpaid and the person cannot wait for the engine. Consent should include platelet buffer collapse, BAT and metabolic risk, sexual and emotional allocation costs, delayed onset as compensation rather than arrival, and withdrawal as engine instability. Tapering should respect nonlinearity.

Third, do not stack ETC tools and SERT tools without understanding the neighborhood. Methylene blue at the Window A amount discussed in EnSurf, MAO inhibitors, high-dose tryptophan or 5-HTP, and SSRIs are not interchangeable wellness accessories. MB is still a MAOI.

This ordering is the same one EnSurf uses for cancer and atherosclerosis. Treat the engine. Read the dashboard. Do not confuse a jammed gauge for a repaired motor.

Conclusion

In this hypothesis, serotonin is a coordinator of bioenergetic flux. It matches oxygen, carbon, electron throughput, and ATP expenditure across gut, blood, lung, placenta, adipose, and brain. It is stored in platelets because a circulating buffer is how a flux manager acts locally without bathing the whole organism. It is abundant in the gut because eating is a hypoxic-adjacent event. It is sparse in the brain because the brain is a customer of the system, not its factory. It becomes melatonin inside mitochondria because the firefighter and the night watch are the same lineage.

The happiness story inverted this geography. It took a few percent of the molecule, a transporter that is not primarily neuronal, and a drug class that jams the valve everywhere, and taught a generation that the problem was a missing drop in a synapse. That teaching organized useful prescriptions and a great deal of confusion. It cannot explain why the drug works late, harms early, empties platelets, quiets brown fat, changes fetal movement quality, and lets go with zaps.

The metabolic hypothesis does not ask psychiatry to disappear. It asks it to take its place downstream of ATP-generating capacity, as oncology and cardiology are being asked to do in other EnSurf essays. Depression, in many of its common forms, may be a chronic compensatory mode: an expensive organ and a tired body trying to reallocate a budget that no longer covers the modern invoice. SSRIs can force a new allocation. They cannot mint new ATP.

Restore the line—fuel, light, movement, oxygen, sleep—and the supervisor often becomes quiet without being handcuffed. When a cuff is still required, put it on with an accurate label.

This essay draws on mitochondrial physiology, serotonergic pharmacology, platelet and fetal bioeffect data, and the energy-regulation tradition in behavioral biology as of 2026. EnSurf methylene blue, where mentioned, refers to 8–25 mg (Window A) discussed as a hypothesis—not a daily mood drug and not a self-treatment protocol; any use is physician-guided. Do not combine with serotonergic drugs. Talk with your own physician before changing medications or supplements.

Disclaimer (repeated). This work is for education and discussion only. It is not medical advice, not a prescription, and not a substitute for in-person care. Do not start, stop, combine, or taper medications or supplements—including SSRIs, SNRIs, MAOIs, 5-HTP, psychedelics, or methylene blue—on the basis of this essay. Serotonin toxicity, G6PD-related hemolysis, and drug–drug interactions are real. Individual decisions about pregnancy, psychiatry, and metabolic interventions require a qualified physician. No doctor–patient relationship is created by reading this text. The author and EnSurf assume no liability for use or misuse of these ideas.

EnSurf: mitochondrial optimization as wellness enhancement. ensurfer.com

Addendum: Psychedelic Bioenergetic Flux, Neuroplasticity, and Serotonin Hormesis

By Val Zudans, MD EnSurf Frugal Philosophy Essay • September 2026

Same disclaimer as the parent essay: hypothesis only; not medical advice; not a protocol to use psychedelics.

The parent essay treated psychedelics as a testable branch: lipophilic 5-HT2A agonists that reach intracellular receptor pools serotonin itself may not reach, then release the system rather than jamming the valve for years. This addendum specifies the flux mechanism. The claim is not that psychedelics are magic mitochondria. The claim is that a brief, high-amplitude occupancy of an intracellular construction program couples a demand spike to a capacity spike, and that neuroplasticity is what that coupling looks like when the invoice is paid in ATP.

Hormesis is the same idea written as a dose-time curve. In this model, exercise, cold, heat, ischemia-reperfusion, and a psychedelic pulse may sit on one family of curves: a transient mismatch between load and capacity that, if survived and then unloaded, raises the set point of the engine. Chronic SERT occupancy sits on the other family: a jammed compensatory lock with no unloading phase.

Two keys, two rooms

Vargas, Olson, and colleagues showed in 2023 that membrane-permeable psychedelics promote cortical neuron growth by activating intracellular 5-HT2A receptors. Polar serotonin, the endogenous ligand everyone assumed was sufficient, does not engage the same plasticity program at ordinary extracellular concentrations. Location bias, not merely receptor identity, is the difference.

That finding is the hinge of a flux reading.

Surface 5-HT2A is the traffic cop of the parent essay. It sits on the membrane, sees circulating or synaptic 5-HT, and participates in moment-to-moment matching: vascular tone, cortical gain, ventilatory drive, the ordinary business of allocation. Intracellular 5-HT2A is a different room. It is reached by lipophilic ligands that cross the bilayer—psilocin, DMT, LSD, DOI—not by the charged amine that platelets and SERT spend their lives packaging and clearing. The intracellular pool looks less like a traffic light and more like a building permit. It is not asking where oxygen should go this minute. It is asking whether this cell should grow new membrane, new spines, and new mitochondria.

If that distinction holds, serotonin is not "the endogenous psychedelic that we fail to take." Serotonin is the endogenous flux matcher. Psychedelics are synthetic (or plant) keys to a remodeling program that evolution may have reserved for other ligands, other concentrations, or other compartments—or that evolution rarely opened because opening it is expensive and destabilizing.

The parent essay's warning still applies: we do not know that intracellular 5-HT2A's true ligand is not a locally synthesized tryptamine, a melatonin-adjacent species, or a lipid-soluble metabolite. Vargas raised the possibility that serotonin is not the endogenous ligand for the intracellular cortical pool. Flux language can live with that uncertainty. It only requires that the intracellular pool be a capacity switch and the surface pool a allocation switch.

The invoice of a spine

Neuroplasticity is discussed as if dendrites were software. They are hardware. Growing a dendritic branch, inserting AMPA receptors, cycling vesicles, and maintaining ionic gradients against leak are among the most ATP-expensive things a neuron does. Local translation at a synapse is gated by mTOR, and mTOR is not a mood kinase. It is an energy-and-nutrient kinase. It listens to ATP, amino acids, insulin/IGF, and stress. A plasticity cascade that ignores bioenergetics is a blueprint without a contractor.

The published psychoplastogen cascade can be rewritten as a flux sequence.

1. Occupancy of the intracellular permit. Lipophilic 5-HT2A agonists cross the membrane. Gq–PLC signaling generates IP3 and diacylglycerol. IP3 releases endoplasmic-reticulum calcium. That calcium has two jobs at once. It is a synaptic coincidence signal. It is also the ion that mitochondria take up through the uniporter to accelerate dehydrogenases and match ATP production to a sudden workload. A calcium pulse that never reaches mitochondria is a demand without payment. A calcium pulse that does is a brief, lawful overdraft.

2. Demand burst. Cortical 5-HT2A activation is reported to produce a glutamate surge onto pyramidal cells. AMPA receptors open. The cell depolarizes. Ion pumps run. Glucose use rises. Vollenweider's FDG-PET work already showed this as physiology, not metaphor: psychotomimetic psilocybin raised global cerebral metabolic rate of glucose, most in frontomedial and frontolateral cortex, anterior cingulate, and temporomedial cortex—on the order of 20–25% in those territories—and the metabolic rise tracked the intensity of the state. That is an invoice. Carhart-Harris later saw reduced CBF on a shorter fMRI timescale. The discrepancy is itself flux-like: phasic vascular and network reorganization versus a longer integrated glucose bill. Either way the acute state is not energetically quiet.

3. Capacity response. 5-HT2A stimulation, including by the tool agonist DOI, recruits SIRT1–PGC-1α in rodent cortical neurons (Fanibunda et al., 2019). Mitochondrial biogenesis increases. Basal and maximal respiration rise. Spare respiratory capacity rises. Cellular ATP rises. Antioxidant enzymes rise. ROS fall relative to the challenge. SIRT1 is required; knockout abolishes the mitochondrial and neuroprotective effects. This is the engine answering the invoice: not only spending more, but installing more generators and raising the ceiling.

The same axis may explain why a pulse of agonism can look neuroprotective against excitotoxic and oxidative insult in rodent neurons if the cell is given hours to build. Protection here is not sedation. It is hormetic preconditioning.

4. Permission to build. BDNF is released. TrkB is engaged. mTOR turns on. Dendrites branch. Spines appear. Shao and colleagues watched this in living mouse cortex after one dose of psilocybin: spine density and head width up within 24 hours, still visible a month later. Ly, Olson, and others showed that TrkB blockade or rapamycin kills the structural phenotype. mTOR is the clerk who will not stamp the permit unless fuel and energy look adequate. That is why, in this frame, a starved, inflamed, insulin-resistant neuron is likely a poor candidate for durable remodeling no matter how romantic the 5-HT2A story.

Moliner et al. (2023) reported that LSD and psilocin bind the TrkB transmembrane dimer as high-affinity positive allosteric modulators, amplifying endogenous BDNF, with plasticity in their assays independent of 5-HT2A and head-twitch dependent on it. That result is not uniformly reproduced. Hold it as a possible second key to the same construction office, not as settled doctrine. Even if TrkB binding is context-dependent, the flux point stands: plasticity drugs converge on a growth program that is energetically gated.

5. Placement of engines. New synapses are useless if ATP is generated three cell-body diameters away. Mitochondrial transport and docking at active spines is part of the phenotype. Serotonin, acting through 5-HT1A receptors, increases mitochondrial movement in rat hippocampal neurons in culture. Psychedelic literature on transport is thinner than the biogenesis literature, but the prediction is obvious: a successful pulse should increase not only mitochondrial mass but mitochondrial presence at the new work sites. Fissler and colleagues framed this as one of three mitochondrial effects of serotonergic psychedelics—biogenesis, function, transport—and treated mitochondrial state as a transdiagnostic moderator of benefit. That is EnSurf language in another dialect.

6. Unloading. The ligand clears. Occupancy falls. Calcium returns to baseline. The glutamate burst ends. What remains is hardware: more spines, more local translational capacity, more spare respiratory capacity, a network that can write new associations because the physical substrate can afford them. This is the step SSRIs do not take. They hold occupancy. There is no unloading phase. Hormesis without unloading is not hormesis. It is chronic load.

Sigma-1 and the ER–mitochondria synapse

DMT binds sigma-1 receptors at the endoplasmic reticulum–mitochondria interface. Sigma-1 is a chaperone at mitochondria-associated membranes. It may help deliver IP3-gated calcium into mitochondria, and it supports survival under hypoxia. In human iPSC cortical neurons and in a rat ischemia model, DMT's protection against hypoxic or ischemic injury has been reported as sigma-1 dependent. Hippocampal neurogenesis after DMT has been blocked by a sigma-1 antagonist in mice.

Flux reading: Gq–PLC at intracellular 5-HT2A opens the calcium store; sigma-1 aims that calcium at the mitochondrion rather than letting it become a cytosolic accident. The same pulse that looks "psychedelic" at the network level looks like coordinated calcium-to-ATP coupling at the organelle level. That coupling is how a cell survives a demand spike instead of entering permeability transition. It is also how a brief hypoxic-adjacent state—the firefighter condition of the parent essay—can end as adaptation rather than injury.

Not all classical psychedelics are equally sigma-1 ligands. The prediction is graded: tryptamines with sigma-1 activity should show more hypoxic resilience and more mitochondrial calcium handling than 5-HT2A-only tools, holding lipophilicity constant.

Hormesis, stated as a curve rather than a vibe

Hormesis is an inverted U relating stressor dose-duration to later capacity.

On the left limb, too little signal: no Ca2+ pulse, no glutamate burst, no SIRT1 recruitment, no reason to build. Ordinary extracellular serotonin lives here for the intracellular pool. So does a subthreshold "microdose" that never crosses the energetic and transcriptional threshold, if such a threshold exists.

On the peak, a pulse large enough to stress ATP, ROS, and calcium handling, short enough that quality-control programs (mitophagy, antioxidant transcription, BDNF) complete before the next insult. The cell is briefly in firefighter mode. Then the firefighter leaves. Spare capacity is higher than before. This is exercise. This is ischemic preconditioning. This is a single well-timed 5-HT2A pulse in a neuron that still has membrane integrity and substrate.

On the right limb, too much or too long: calcium overload, permeability transition, excitotoxicity, receptor downregulation without rebuilding, glycolytic lock, loss of dynamic range. Chronic high-dose occupancy, repeated pulses without recovery, or a pulse delivered into an already failing engine live here. So does the SSRI pattern if one insists on calling constant SERT occupancy a "serotonergic therapy." It is the wrong region of the same curve.

The serotonin hormesis model in flux terms is therefore not "more 5-HT is better until it isn't." It is:

• Amplitude. Intracellular occupancy must be high enough to open the construction program. Surface occupancy alone allocates; it does not reliably rebuild.
• Duration. Hours, not months. The signal must end so mTOR, PGC-1α, and spine stabilization can run in a lower-demand night shift—circadian language is not decorative here. Mitochondrial quality control may be largely nocturnal in a properly entrained animal.
• Substrate. The pulse spends glucose and oxygen now (Vollenweider's bill) and needs fat/ketone flexibility, amino acids, and cofactors later to finish the protein and membrane work. A ketogenic or low-insulinogenic background is not required for the acute state, but it is a plausible moderator of whether the new hardware is paid off or scavenged.
• Baseline engine quality. Spare respiratory capacity before the pulse determines whether the overdraft is hormetic or injurious. This is why Fissler's claim that benefit depends on patient-specific mitochondrial dysregulation is the right kind of claim. It is also why the parent essay's order of operations—fuel, light, movement, oxygen, sleep before exotic ligands—is not lifestyle garnish. It is setting the left and right bounds of the U.
• Unloading and writing. New spines are options, not outcomes. Metaplasticity means the network is more writable. Therapy, light, social repair, and strength work are the programs that write. A pulse without a subsequent ancestral-aligned life is a construction site that weathers.

SSRIs fail this curve on duration and on compartment. They raise extracellular 5-HT continuously, empty the platelet buffer, brake brown fat, and, in this model, do not open the intracellular permit the way a lipophilic agonist does. Fluoxetine can still interact with TrkB at much lower affinity and may promote plasticity slowly, over weeks. That is a different, flatter curve: chronic occupancy plus slow growth, purchased with loss of sexual allocation, emotional range, and metabolic flexibility. Calling both "serotonergic neuroplasticity" is like calling a marathon and a prison treadmill the same stimulus because both involve legs.

Why the subjective state may still matter without being the mechanism

A flux model does not need the hallucination to be the medicine. It needs the occupancy pattern that happens to produce the hallucination in a human cortex. Gq efficacy at 5-HT2A predicts head-twitch and, in many assays, the acute state. Plasticity assays sometimes dissociate: non-hallucinogenic 5-HT2A agonists and biased ligands can grow neurites; some TrkB-dependent effects may not need 5-HT2A at all. The honest position is dual.

The construction program can be triggered without a full psychedelic state. The allocation of that new capacity across circuits—what gets wired to what—is likelier to depend on the network regime during and after the pulse. High-entropy, high-glucose-use cortex with loosened default-mode coupling is a different writing environment than sleep or ordinary wake. Set and setting, in flux language, are the load profile during the construction window. They do not mint ATP. They decide which synapses keep the ATP they just earned.

This is why integration is not aftercare fluff. Hardware without a program is waste. It is also why combining a pulse with sleep deprivation, alcohol, or continuous insulinogenic feeding is a way to land on the right limb of the U.

Predictions that would kill or carry this branch

1. Spare respiratory capacity and spine persistence will co-vary. If spines last a month only in cells whose spare capacity rose, plasticity is an energy phenotype with a morphological shadow. If spines persist while mitochondria do not change, the flux reading of plasticity is too strong.

1. Intracellular access will predict the construction program better than surface Ki. Membrane-impermeant 5-HT2A agonists should allocate without rebuilding; permeant ones should rebuild. Chemical series that vary logP at constant surface potency are the experiment.

1. Sigma-1 contribution will show up as hypoxic resilience and mitochondrial calcium, not as head-twitch. Strip sigma-1 and DMT should remain visually "psychedelic" in some models and lose organelle protection.

1. Metabolic preconditioning will widen the hormetic window. Resistance training, circadian alignment, and adequate protein before a pulse should increase durability and decrease adverse physiologic events. Severe insulin resistance, apnea, and active inflammation should do the opposite.

1. Chronic SSRI pretreatment will flatten the pulse. Downregulated 5-HT2A, depleted platelet buffer, braked BAT, and a mitochondrion already adapted to jammed extracellular 5-HT should shrink both the subjective amplitude and the biogenesis response. Washout time should restore the U. If SSRIs potentiate psychedelic plasticity, this branch is wrong in an interesting way.

1. Rapamycin and severe caloric-plus-protein restriction during the building window will spare the acute state more than they spare the spines. If mTOR is the clerk of the invoice, starving the clerk prevents construction even when the permit was stamped.

1. Regional FDG or equivalent flux imaging days to weeks later should show a lower, more flexible bill if the pulse worked as hormesis—more capacity, less compensatory hypermetabolism—rather than a permanently hyperfrontal engine.

Therapeutic order, restated for this ligand class

Psychedelics, in this frame, are not antidepressants that happen to grow dendrites. They are brief, high-amplitude keys to an intracellular remodeling program whose output is more writable cortex sitting on more mitochondrial capacity—if the engine can pay the acute bill and if life after the pulse writes something worth keeping.

That still puts them downstream of EnSurf's ordinary work. You do not hormetically shock a failing ETC and call the shock a foundation. You restore fuel matching, circadian melatonin-mitochondrial night shift, strength as the cheap biogenesis drug, and oxygen as the terminal acceptor. Then, if a construction permit is still required, you use a pulse with an honest label: this will raise the invoice for a day, attempt to raise capacity for weeks to months, and leave the writing to what the person does with the new hardware.

Stacking methylene blue or other ETC tools with a serotonergic pulse is the same neighborhood problem named in the parent essay. MAO, sigma-1, 5-HT2A, and complex I live close together. Curiosity is not a protocol.

The serotonin hormesis model is not a permission slip for chronic serotonergic occupancy. It is the opposite. Hormesis is the case for rarity, amplitude, unloading, and a competent engine. The firefighter may open the construction site. He is not supposed to move in.

Addendum: Methylene Blue — ETC Tool and MAO Valve in the Same Neighborhood

By Val Zudans, MD EnSurf Frugal Philosophy Essay • September 2026

Same disclaimer as the parent essay: hypothesis only; not medical advice; not a protocol to self-administer methylene blue. Do not combine with serotonergic drugs. Do not use in G6PD deficiency.

Methylene blue sits on both sides of the serotonin–mitochondria neighborhood named in the parent essay. It is a catalytic electron cycler in the ETC. It is also a potent, reversible MAO-A inhibitor. Those are not two marketing claims for one pill. They are two direct binding behaviors of one redox dye, with different concentration windows, different directions of effect on flux, and a nasty interaction when someone has already jammed SERT.

They are both affected. They are not affected in the same direction. One changes capacity. The other changes signal disposal. Dose decides which story dominates. The ETC low dose discussed in EnSurf (Window A) is 8–25 mg. At real human exposures the ETC and MAO-A windows still overlap.

On Sage — Methylene Blue.

Geography first

MAO-A lives on the outer mitochondrial membrane, facing the cytosol. That is not a coincidence in a flux model. Serotonin that has entered the cell is oxidized there to 5-HIAA. The same organelle that needs 5-HT/melatonin as an inner-membrane stress adaptation is the organelle that wears the disposal enzyme on its coat.

The ETC lives on the inner membrane: complexes I–IV, cytochrome c, ATP synthase. Low-dose MB cycles between oxidized MB and reduced leucomethylene blue (MBH2). It can accept electrons from NADH-linked sites and donate them to cytochrome c, bypassing bottlenecked or leaky stretches of complexes I and III, and it can raise cytochrome oxidase (complex IV) activity. At those concentrations it is an alternative wire, not a receptor agonist.

So the dye is physically in a position to touch: - the wire that makes ATP from oxygen and fuel, and - the shears that cut serotonin on the way in.

That is why stacking MB with SSRIs, SNRIs, MAOIs, tramadol, triptans, or high-dose 5-HTP is a hard stop. You do not occupy the shears and the valve at the same time.

Direct vs indirect

Both hits are direct.

ETC: redox cycling at the chain. Neuroprotection EC50 in some cell models is nanomolar. High concentration (Window C) reverses the sign: MB may then act as an electron sink and a pro-oxidant, and can uncouple or swell mitochondria. Gonzalez-Lima's group has called this what it is—a hormetic autoxidizable dye, not a linear "mito vitamin."

MAO-A: tight-binding reversible inhibition. Ramsay's kinetics show tight-binding inhibition. IC50 values cluster around 70 nM for MB; the metabolite azure B is tighter still (~11 nM). MAO-B is far weaker (high nM to ~1 µM). After 0.75 mg/kg IV, peak plasma is about 1.6 µM—enough, in Gillman's analysis, to inhibit MAO-A completely and MAO-B partially. Definite serotonin toxicity has been documented at 1 mg/kg IV on an SSRI background, not only at the old "5–10 mg/kg" warning band.

Indirect effects exist and matter for flux, but they are downstream of those two bindings:

• Better ETC → higher ATP-generating capacity → less hypoxic-adjacent signaling → less reason to deploy the firefighter (platelet release, ventilatory 5-HT drive, gut distress signaling).
• MAO-A block → more 5-HT survives in gut, platelet-adjacent plasma, cytosol, and synapse → more surface 5-HT2A allocation signaling, and possibly more substrate for intracellular 5-HT and melatonin if the cell can still acetylate what MAO would have burned.
• MAO-A block in the gut after an oral dose is first-pass flux: the largest 5-HT organ sees the dye before the brain does.
• NOS / soluble guanylate cyclase inhibition appears at higher micromolar levels than MAO-A—another oxygen-delivery lever, later on the curve.

Do not flatten this into "MB raises serotonin and energy." One lever raises the engine's ability to pay the invoice. The other slows destruction of the invoice clerk.

Direction: not the same

Same-direction only in the narrow sense that both can feel like "more energy / more alertness" in a person whose bottleneck was either a weak ETC or a high 5-HT clearance. Opposite-direction in the sense that matters for the hypothesis: ETC support should reduce dependence on compensatory 5-HT; MAO-A block increases 5-HT persistence whether or not the engine needed it.

That is why MB can look like a clean mito tool in a metabolically constrained, drug-naive person and like a dirty MAOI in a person on sertraline. Same molecule. Different occupancy of the two sites relative to the background valve state.

Dose is the whole argument

Approximate windows—not a protocol, a map. Tissue and mitochondrial accumulation of this cationic dye may make plasma an underestimate of organelle concentration. Oral bioavailability is incomplete; in rats, the gut wall sees a high local dose.

There is no clean human dose at which ETC is on and MAO-A is guaranteed off on a dosing day. In vitro, MAO-A is the more potent of the two bindings. The reason low oral pulses can still look like a mito drug is pharmacokinetics and context: oral MB concentrated in intestinal wall and liver in rats (Peter et al., 2000), and gave lower blood exposure than the same milligrams IV in people; partitioning into mitochondria makes cycling catalytic (one molecule shuttles many electrons); and a person whose limiting reagent was complex I/III leak rather than 5-HT clearance. Catalytic ETC cycling can matter at amounts that only partially inhibit MAO. Partial MAO-A inhibition is still a flux event in the gut on the day of the capsule.

Azure B, the demethylated metabolite, is a stronger MAO-A inhibitor than parent MB. Daily oral dosing is not just "more MB." It is a changing ratio of parent to a tighter shears-blocker.

Interval is the occupancy variable

After 24 hours a prior pulse is only partly cleared. After 48–72 hours systemic parent is largely gone.

Window A amounts are not MAO-silent.

SRI, SNRI, MAOI, tramadol, triptan, 5-HTP, and psychedelic stacking remains a hard stop on dosing days and through washout. Reversible does not mean "safe with sertraline if you only dose twice a week."

How this maps onto serotonin's flux job

Recall the four streams: oxygen delivery, carbon delivery, electron throughput, ATP expenditure.

Low-dose ETC cycling is a direct intervention on stream three. If complexes I or III are leaky or blocked, electrons still reach cytochrome c and complex IV. Oxygen can be used. ATP synthase still gets a proton-motive force from the segments that remain coupled. Firefighter signaling should, in theory, quiet because the hypoxic-adjacent and ROS-adjacent cues that deploy 5-HT are reduced. This is engine work. It is the EnSurf rationale.

MAO-A inhibition is a direct intervention on the disposal of the flux manager itself. 5-HT that should have been cut at the mitochondrial coat persists. Consequences:

• Surface receptors see more ligand → more allocation signaling (gut motility, vascular tone, cortical gain, 5-HT2A–SIRT1–PGC-1α biogenesis). That last path can synergize with ETC cycling: more construction-and-biogenesis signal plus a better wire.
• Intracellular 5-HT and melatonin routing change. Less MAO means more 5-HT available to NAT/HIOMT if those enzymes and acetyl-CoA are present. That could support the guardian-angel side. It could also leave an uncleared cytosolic 5-HT load.
• Combined with SERT block: no packaging into platelets, no synaptic clearance, no MAO disposal. The buffer, the valve, and the shears are all offline. That is serotonin toxicity, not hormesis.
• Combined with a psychedelic pulse: more lingering 5-HT plus an intracellular 5-HT2A key. Unpredictable amplitude. Not an experiment to run casually.

So the honest answer to "same direction?" is:

• On 5-HT abundance: both low-dose ETC improvement (indirectly, via less consumption/stress) and MAO-A block (directly) can raise effective serotonergic tone. Same direction on the dashboard.
• On ATP-generating capacity: ETC low-dose up, ETC high-dose down. MAO-A is almost silent on capacity except insofar as extra 5-HT2A signaling drives biogenesis—or insofar as a serotonin crisis wastes ATP on rigidity, fever, and autonomic chaos.
• On the firefighter logic: ETC support should retire the firefighter. MAO-A block keeps him on scene longer. Opposite managerial advice. Useful together only when the engine was weak and 5-HT clearance was not already blocked.

Why the curve is hormetic twice

MB has two stacked inverted U's.

1. Redox U: too little cycling, no bypass; enough cycling, better COX and less I/III leak; too much, electron theft and ROS.
2. Serotonergic U: a little less MAO-A in a high-clearance, low-capacity brain may look like mood or focus; a lot of MAO-A block on top of an SRI is toxicity; even without an SRI, high 5-HT tone with a failing ETC is the parent essay's compensatory lock.

Blue urine is a poor guide to which U you are on. It means dye reached urine. It does not mean you are at 0.5 µM at complex IV.

Practical order, not a recipe

• Treat MB as an ETC tool that happens to be a MAOI, not as a serotonin strategy. The EnSurf indication is capacity, not mood chemistry.
• Window A amount discussed in EnSurf: 8–25 mg; information, not a recommendation, and any use is physician-guided.
• Do not combine with SERT or MAO drugs, 5-HTP, or a psychedelic pulse. Washout exists for a reason.
• Oral 25 mg still sees gut MAO-A on dosing day. That is a feature of the largest 5-HT compartment, not a footnote.
• If the hypothesis is right, the people who benefit most are those whose invoice is an ETC bottleneck (aging heteroplasmy, complex I leak, hypoxic-adjacent tissues) and who are not already running jammed serotonergic valves.
• If someone feels "serotonergic" on a dosing day—jaw tightness, clonus, agitation, sweating, diarrhea, insomnia—that is the shears, not the wire. Stop.

The neighborhood is small. ATP synthase, MAO-A, intracellular 5-HT, and melatonin synthesis share the same organelle. MB walks into that room with a jumper cable in one hand and a pair of shears in the other. Dose decides amplitude. Background drugs decide whether the shears are a tweak or a disaster.

A working hypothesis, open to correction and revised as new evidence arrives. Last revised October 31, 2026.

References

1. Gershon MD, 2013. 5-Hydroxytryptamine (serotonin) in the gastrointestinal tract. Current Opinion in Endocrinology, Diabetes and Obesity 20(1):14-21. doi:10.1097/MED.0b013e32835bc703
2. Moncrieff J, Cooper RE, Stockmann T, et al., 2023. The serotonin theory of depression: a systematic umbrella review of the evidence. Molecular Psychiatry 28(8):3243-3256. doi:10.1038/s41380-022-01661-0
3. Andrews PW, Bharwani A, Lee KR, et al., 2015. Is serotonin an upper or a downer? The evolution of the serotonergic system and its role in depression and the antidepressant response. Neuroscience & Biobehavioral Reviews 51:164-188. doi:10.1016/j.neubiorev.2015.01.018
4. Suofu Y, Li W, Jean-Alphonse FG, et al., 2017. Dual role of mitochondria in producing melatonin and driving GPCR signaling to block cytochrome c release. Proceedings of the National Academy of Sciences of the USA 114(38):E7997-E8006. doi:10.1073/pnas.1705768114
5. Yabut JM, Crane JD, Green AE, et al., 2019. Emerging Roles for Serotonin in Regulating Metabolism: New Implications for an Ancient Molecule. Endocrine Reviews 40(4):1092-1107. doi:10.1210/er.2018-00283
6. Sumara G, Sumara O, Kim JK, et al., 2012. Gut-derived serotonin is a multifunctional determinant to fasting adaptation. Cell Metabolism 16(5):588-600. doi:10.1016/j.cmet.2012.09.014
7. Suchacki KJ, Ramage LE, Kwok TC, et al., 2023. The serotonin transporter sustains human brown adipose tissue thermogenesis. Nature Metabolism 5(8):1319-1336. doi:10.1038/s42255-023-00839-2
8. Fanibunda SE, Deb S, Maniyadath B, et al., 2019. Serotonin regulates mitochondrial biogenesis and function in rodent cortical neurons via the 5-HT2A receptor and SIRT1-PGC-1alpha axis. Proceedings of the National Academy of Sciences of the USA 116(22):11028-11037. doi:10.1073/pnas.1821332116
9. Salisbury AL, Anderson GM, Yang A, et al., 2024. Fetal behavior and gestational serotonin reuptake inhibitor exposure: relationships between behavior, drug dosage, plasma drug level, and a measure of drug bioeffect. Neuropsychopharmacology 49(13):1968-1977. doi:10.1038/s41386-024-01923-1
10. Wong AC, Devason AS, Umana IC, et al., 2023. Serotonin reduction in post-acute sequelae of viral infection. Cell 186(22):4851-4867.e20. doi:10.1016/j.cell.2023.09.013
11. Horowitz MA, Taylor D, 2019. Tapering of SSRI treatment to mitigate withdrawal symptoms. The Lancet Psychiatry 6(6):538-546. doi:10.1016/S2215-0366(19)30032-X
12. Ramsay RR, Dunford C, Gillman PK, 2007. Methylene blue and serotonin toxicity: inhibition of monoamine oxidase A (MAO A) confirms a theoretical prediction. British Journal of Pharmacology 152(6):946-951. doi:10.1038/sj.bjp.0707430
13. Vargas MV, Dunlap LE, Dong C, et al., 2023. Psychedelics promote neuroplasticity through the activation of intracellular 5-HT2A receptors. Science 379(6633):700-706. doi:10.1126/science.adf0435
14. Vollenweider FX, Leenders KL, Scharfetter C, et al., 1997. Positron emission tomography and fluorodeoxyglucose studies of metabolic hyperfrontality and psychopathology in the psilocybin model of psychosis. Neuropsychopharmacology 16(5):357-372. doi:10.1016/S0893-133X(96)00246-1
15. Carhart-Harris RL, Erritzoe D, Williams T, et al., 2012. Neural correlates of the psychedelic state as determined by fMRI studies with psilocybin. Proceedings of the National Academy of Sciences of the USA 109(6):2138-2143. doi:10.1073/pnas.1119598109
16. Shao LX, Liao C, Gregg I, et al., 2021. Psilocybin induces rapid and persistent growth of dendritic spines in frontal cortex in vivo. Neuron 109(16):2535-2544.e4. doi:10.1016/j.neuron.2021.06.008
17. Ly C, Greb AC, Cameron LP, et al., 2018. Psychedelics Promote Structural and Functional Neural Plasticity. Cell Reports 23(11):3170-3182. doi:10.1016/j.celrep.2018.05.022
18. Moliner R, Girych M, Brunello CA, et al., 2023. Psychedelics promote plasticity by directly binding to BDNF receptor TrkB. Nature Neuroscience 26(6):1032-1041. doi:10.1038/s41593-023-01316-5
19. Chen S, Owens GC, Crossin KL, et al., 2007. Serotonin stimulates mitochondrial transport in hippocampal neurons. Molecular and Cellular Neuroscience 36(4):472-483. doi:10.1016/j.mcn.2007.08.004
20. Fissler P, Vandersmissen A, Filippi M, et al., 2023. Effects of serotonergic psychedelics on mitochondria: Transdiagnostic implications for mitochondria-related pathologies. Journal of Psychopharmacology 37(7):679-686. doi:10.1177/02698811231164707
21. Fontanilla D, Johannessen M, Hajipour AR, et al., 2009. The hallucinogen N,N-dimethyltryptamine (DMT) is an endogenous sigma-1 receptor regulator. Science 323(5916):934-937. doi:10.1126/science.1166127
22. Szabo A, Kovacs A, Riba J, et al., 2016. The Endogenous Hallucinogen and Trace Amine N,N-Dimethyltryptamine (DMT) Displays Potent Protective Effects against Hypoxia via Sigma-1 Receptor Activation in Human Primary iPSC-Derived Cortical Neurons and Microglia-Like Immune Cells. Frontiers in Neuroscience 10:423. doi:10.3389/fnins.2016.00423
23. Szabó Í, Varga VÉ, Dvorácskó S, et al., 2021. N,N-Dimethyltryptamine attenuates spreading depolarization and restrains neurodegeneration by sigma-1 receptor activation in the ischemic rat brain. Neuropharmacology 192:108612. doi:10.1016/j.neuropharm.2021.108612
24. Morales-Garcia JA, Calleja-Conde J, Lopez-Moreno JA, et al., 2020. N,N-dimethyltryptamine compound found in the hallucinogenic tea ayahuasca, regulates adult neurogenesis in vitro and in vivo. Translational Psychiatry 10(1):331. doi:10.1038/s41398-020-01011-0
25. Casarotto PC, Girych M, Fred SM, et al., 2021. Antidepressant drugs act by directly binding to TRKB neurotrophin receptors. Cell 184(5):1299-1313.e19. doi:10.1016/j.cell.2021.01.034
26. Wen Y, Li W, Poteet EC, et al., 2011. Alternative mitochondrial electron transfer as a novel strategy for neuroprotection. Journal of Biological Chemistry 286(18):16504-16515. doi:10.1074/jbc.M110.208447
27. Rojas JC, Bruchey AK, Gonzalez-Lima F, 2012. Neurometabolic mechanisms for memory enhancement and neuroprotection of methylene blue. Progress in Neurobiology 96(1):32-45. doi:10.1016/j.pneurobio.2011.10.007
28. Bruchey AK, Gonzalez-Lima F, 2008. Behavioral, Physiological and Biochemical Hormetic Responses to the Autoxidizable Dye Methylene Blue. American Journal of Pharmacology and Toxicology 3(1):72-79. doi:10.3844/ajptsp.2008.72.79
29. Petzer A, Harvey BH, Wegener G, et al., 2012. Azure B, a metabolite of methylene blue, is a high-potency, reversible inhibitor of monoamine oxidase. Toxicology and Applied Pharmacology 258(3):403-409. doi:10.1016/j.taap.2011.12.005
30. Gillman PK, 2011. CNS toxicity involving methylene blue: the exemplar for understanding and predicting drug interactions that precipitate serotonin toxicity. Journal of Psychopharmacology 25(3):429-436. doi:10.1177/0269881109359098
31. Walter-Sack I, Rengelshausen J, Oberwittler H, et al., 2009. High absolute bioavailability of methylene blue given as an aqueous oral formulation. European Journal of Clinical Pharmacology 65(2):179-189. doi:10.1007/s00228-008-0563-x
32. Peter C, Hongwan D, Küpfer A, et al., 2000. Pharmacokinetics and organ distribution of intravenous and oral methylene blue. European Journal of Clinical Pharmacology 56(3):247-250. doi:10.1007/s002280000124
33. Mylecharane EJ, 2013. Serotonin and the Australian connection: the science and the people. ACS Chemical Neuroscience 4(1):28-32. doi:10.1021/cn4000082
34. Berk M, Woods RL, Nelson MR, et al., 2020. Effect of Aspirin vs Placebo on the Prevention of Depression in Older People: A Randomized Clinical Trial. JAMA Psychiatry 77(10):1012-1020. doi:10.1001/jamapsychiatry.2020.1214
35. Patrono C, 2024. Low-dose aspirin for the prevention of atherosclerotic cardiovascular disease. European Heart Journal 45(27):2362-2376. doi:10.1093/eurheartj/ehae324
36. Badawy AA, 2017. Kynurenine Pathway of Tryptophan Metabolism: Regulatory and Functional Aspects. International Journal of Tryptophan Research 10:1178646917691938. doi:10.1177/1178646917691938
37. Magistretti PJ, Allaman I, 2015. A cellular perspective on brain energy metabolism and functional imaging. Neuron 86(4):883-901. doi:10.1016/j.neuron.2015.03.035
38. Kious BM, Kondo DG, Renshaw PF, 2018. Living High and Feeling Low: Altitude, Suicide, and Depression. Harvard Review of Psychiatry 26(2):43-56. doi:10.1097/HRP.0000000000000158
39. Hodges MR, Richerson GB, 2010. The role of medullary serotonin (5-HT) neurons in respiratory control: contributions to eupneic ventilation, CO2 chemoreception, and thermoregulation. Journal of Applied Physiology 108(5):1425-1432. doi:10.1152/japplphysiol.01270.2009
40. Meyer JH, Wilson AA, Sagrati S, et al., 2004. Serotonin transporter occupancy of five selective serotonin reuptake inhibitors at different doses: an [11C]DASB positron emission tomography study. American Journal of Psychiatry 161(5):826-835. doi:10.1176/appi.ajp.161.5.826
41. Wichniak A, Wierzbicka A, Walęcka M, et al., 2017. Effects of Antidepressants on Sleep. Current Psychiatry Reports 19(9):63. doi:10.1007/s11920-017-0816-4
42. Blier P, Piñeyro G, el Mansari M, et al., 1998. Role of somatodendritic 5-HT autoreceptors in modulating 5-HT neurotransmission. Annals of the New York Academy of Sciences 861:204-216. doi:10.1111/j.1749-6632.1998.tb10192.x
43. Aghajanian GK, Marek GJ, 1999. Serotonin, via 5-HT2A receptors, increases EPSCs in layer V pyramidal cells of prefrontal cortex by an asynchronous mode of glutamate release. Brain Research 825(1-2):161-171. doi:10.1016/s0006-8993(99)01224-x
44. Wallach J, Cao AB, Calkins MM, et al., 2023. Identification of 5-HT(2A) receptor signaling pathways associated with psychedelic potential. Nature Communications 14(1):8221. doi:10.1038/s41467-023-44016-1
45. Cameron LP, Tombari RJ, Lu J, et al., 2021. A non-hallucinogenic psychedelic analogue with therapeutic potential. Nature 589(7842):474-479. doi:10.1038/s41586-020-3008-z
46. Mayer B, Brunner F, Schmidt K, 1993. Inhibition of nitric oxide synthesis by methylene blue. Biochemical Pharmacology 45(2):367-374. doi:10.1016/0006-2952(93)90072-5
47. Shah-Khan MG, Lovely J, Degnim AC, 2012. Safety of methylene blue dye for lymphatic mapping in patients taking selective serotonin reuptake inhibitors. American Journal of Surgery 204(5):798-799. doi:10.1016/j.amjsurg.2012.02.004

Related: Signals · Brain · Thyroid / Temperature · Cardiac · Fuel · Circadian · Methylene Blue · Philosophy of Medicine · Immune

See the notice at the top of this page. Working hypothesis, not medical advice.