Our bodies do not have a maintenance mode. Every system that keeps us functional — muscle mass, bone density, mitochondrial capacity, neuronal connectivity, immune calibration — is maintained only in response to demand. Remove the demand signals, and the biology interprets absence as permission to downregulate. What we call aging is substantially the accumulation of unrepaired degradation in systems that require demand signals to repair themselves. The mitochondrion is where this convergence becomes visible: it is the common downstream target of protein deficiency, mineral depletion, movement deprivation, circadian disruption, and chronic inflammation.
Most of what follows is explanation. If you want the prescription first:
The Short Version
Do these without testing:
| Intervention | Specifics |
|---|---|
| Protein | 1.6–2.2g per kg lean body mass; diverse animal sources plus collagen; past 60, prioritize ~2.5–3g leucine per meal |
| Magnesium | Glycinate or malate, 200–400mg, evening; threonate if cognitive support is the primary goal |
| Vitamin D3 + K2 | 2000–4000 IU D3 daily; K2 (MK-7 form) with it |
| EPA/DHA | 1–2g combined daily; fish oil concentrate for anti-inflammatory emphasis; algae-based for neurological emphasis |
| Methylated B complex | Methylcobalamin, methylfolate (5-MTHF), P5P — one capsule covers it |
| Fiber | Diverse plants and fermented foods; psyllium husk as a practical supplement |
| CoQ10 + NAD+ precursor | Ubiquinol (CoQ10), 100–200mg; NMN or NR, 250–500mg |
| Resistance training | Primary trigger for mitochondrial biogenesis; eccentric emphasis for disproportionate signal |
| End-range movement | Yoga, Pilates, or equivalent — positional diversity under load |
| Sleep and light | Consistent timing, eight hours, dark after sunset; cool room (65°F/18°C) |
The essay below explains why these converge on the same underlying problem.
The Central Argument
This is not primarily a story about aging. It is a story about signal ecology — the degradation of the biological signal environment our physiology evolved to require. When mitochondrial function declines, neurons run short on energy before they die — and recent research has established this energy failure as a cause of cognitive decline, not merely a consequence of it. The mitochondrion is the convergence point: every upstream deprivation named above hits the same downstream target.
The framework that follows is organized into what can be done universally and what requires individual information to optimize.
Tier 1: The Universal Stack
These interventions have wide therapeutic windows, meaningful upside for almost everyone, and no significant downside risk for healthy adults. No testing required. Start here.
Protein
The dietary reference intake of 0.8g per kilogram of bodyweight is a floor against deficiency, not an optimum for function. Most adults — especially past forty, when anabolic resistance increases — are chronically under-consuming protein relative to what is needed to maintain muscle mass, support mitochondrial biogenesis, and buffer the metabolic stress of modern life.
A more functional target is 1.6–2.2g per kilogram of lean body mass. Composition matters as much as quantity: diverse animal sources provide complete amino acid profiles, and collagen-containing foods (bone broth, slow-cooked cuts) supply glycine and proline that most modern diets have largely eliminated. Past sixty, meal distribution matters as much as daily total — muscle protein synthesis becomes harder to trigger, and hitting roughly 2.5–3g of leucine per meal appears necessary to fully stimulate it. Higher protein intake also substantially improves insulin signaling by blunting postprandial glucose response, increasing the glucagon-to-insulin ratio, and preserving the muscle mass that is the body’s primary glucose sink.
Magnesium
Magnesium is a cofactor in over three hundred enzymatic reactions, including ATP synthesis — the energy currency that mitochondria produce. ATP exists in the body primarily as magnesium-ATP. Soil depletion, food processing, and the renal magnesium wasting driven by chronic stress have created widespread functional insufficiency that standard serum testing fails to detect, because serum magnesium is a poor proxy for total body stores.
Form and timing both matter. Magnesium glycinate is the best-absorbed general form; the glycine it is bound to has independent sleep-supportive properties — it lowers core body temperature and improves sleep quality — which makes evening dosing the practical default. Magnesium threonate is worth considering for those whose primary goal is cognitive support: it is formulated to cross the blood-brain barrier and raise brain magnesium levels specifically, at higher cost and with a still-growing evidence base. Magnesium malate is a reasonable alternative for muscle fatigue. Oxide, the most common form in cheap supplements, is poorly absorbed and primarily a laxative. The main side effect at higher doses of any absorbable form is loose stools — a self-limiting signal to reduce dose. A fuller guide to forms and dosing is here.
Vitamin D3 and K2
Vitamin D deficiency is widespread and its effects run deeper than bone health — it regulates immune function, inflammatory tone, and gene expression in ways that affect mitochondrial health indirectly. Supplementation in the 2000–4000 IU range is a reasonable starting floor for most adults.
D and K2 function as a system: higher vitamin D increases calcium absorption, and K2 (specifically the MK-7 form, which has a longer half-life than MK-4) directs that calcium into bone rather than soft tissue. Taking them together is a sensible default. Neither requires testing before starting at moderate doses, though testing vitamin D levels (25-OH) is inexpensive and useful for calibration.
Omega-3 Fatty Acids (EPA and DHA)
The ratio of omega-3 to omega-6 fatty acids in the modern diet has shifted dramatically from historical norms, driven by the proliferation of seed oils and reduced consumption of fatty fish. Omega-6 fatty acids are pro-inflammatory substrates; omega-3s are anti-inflammatory. The balance shapes baseline inflammatory tone, which has direct downstream effects on mitochondrial function. The two relevant fatty acids are not interchangeable: EPA drives the anti-inflammatory effect; DHA supports neuronal membrane integrity and myelin maintenance. Fish oil concentrate, specifying 1–2g combined EPA+DHA per serving, provides both. Algae-based omega-3 is the original source — fish accumulate DHA by eating algae — and is DHA-dominant, which suits neurological support well but delivers less of the anti-inflammatory EPA punch. For those using algae oil as a direct fish oil swap, this distinction is worth knowing. Quality varies significantly in cheap fish oil; Nordic Naturals and Carlson are reliable.
Fiber and the Gut-Brain Axis
Adequate fermentable fiber supports colonic production of butyrate, a short-chain fatty acid that serves as a direct energy source for colonocytes and is associated with improved mitochondrial function and biogenesis. This pathway matters beyond digestion: the gut and brain communicate bidirectionally through the vagus nerve, and disruption of the microbiome — through low-fiber diets, antibiotics, or chronic stress — impairs that signaling channel.
The practical prescription is diverse plants and fermented foods as the foundation. Psyllium husk is the most practical supplement form: primarily soluble fiber that ferments into butyrate and other short-chain fatty acids, inexpensive, widely available, and well-tolerated at 5–10g daily in water. It is not a substitute for dietary diversity but a useful gap-filler, and its fermentation profile makes it a better choice than insoluble fiber supplements for this purpose.
Methylated B Vitamins
B12, folate, and B6 support methylation — a biochemical process involved in DNA repair, neurotransmitter synthesis, myelin maintenance, and mitochondrial function via CoQ10 synthesis. Many people have genetic variants (particularly MTHFR) that impair their ability to convert synthetic folic acid into active methylfolate. High-dose folic acid in these individuals can actually block the pathway by competing for the enzyme. The B6 conversion issue is less discussed but matters: pyridoxine, the synthetic form, can be neurotoxic at high doses or when conversion to the active P5P form is impaired.
A single methylated B-complex covers the requirement without needing to genotype first. Look for: methylcobalamin (not cyanocobalamin) for B12, methylfolate or 5-MTHF (not folic acid) for folate, and pyridoxal-5-phosphate or P5P (not pyridoxine) for B6. Thorne, Pure Encapsulations, and Jarrow make reliable formulations. These forms work for everyone; synthetic forms work only for those without conversion variants.
Mitochondrial Support: CoQ10 and NAD+ Precursors
Two supplements target mitochondrial function directly, and both have meaningful human evidence.
CoQ10 is a required component of the mitochondrial electron transport chain — the machinery that produces ATP. Production declines significantly with age and is further depleted by statin medications. The ubiquinol form is substantially better absorbed than ubiquinone. A dose of 100–200mg daily is the practical range; people on statins should consider the higher end.
NAD+ (nicotinamide adenine dinucleotide) is essential for mitochondrial energy production and declines by roughly half between early adulthood and age sixty. NMN and NR are the two available precursors; both raise NAD+ levels in humans, with functional outcome evidence still catching up to the mechanistic rationale — promising but not yet settled. A dose of 250–500mg daily is the common range. An emerging parallel strategy is CD38 inhibition — CD38 is an enzyme that consumes NAD+, and compounds that suppress it (including apigenin, found in parsley and chamomile) may help preserve NAD+ levels; this is an area to watch rather than a current prescription.
Movement Diversity and End-Range Loading
The problem with modern movement is not only quantity but variety. Repetitive movement patterns — even active ones — leave entire ranges of motion unloaded, and tissue that is never loaded at length undergoes adaptive shortening while losing the fibroblast activity that keeps connective tissue healthy.
The prescription is positional diversity under load: end-range positions held and moved through, eccentric and isometric demands that differ from habitual patterns. This is why yoga and Pilates produce results that walking alone does not — they load tissues at lengths the body otherwise never visits. The signal being sent is: this range is required, maintain it.
Resistance training serves a distinct function: it is the primary trigger for PGC-1α, the master regulator of mitochondrial biogenesis. Within resistance training, eccentric-focused work — the lowering phase, slow and controlled — provides a disproportionately strong mitochondrial and hypertrophic signal relative to concentric effort alone. This is the most direct demand signal in the entire framework.
Breath-integrated movement also supports autonomic regulation through vagal tone. The vagus nerve is the primary pathway through which the brain modulates peripheral inflammation and receives signals from the gut. Practices that combine slow movement with deliberate breathing patterns appear to influence this axis in ways that pure cardio does not.
Sleep and Circadian Alignment
The circadian system governs far more than sleep timing. It coordinates cortisol rhythm, melatonin production, insulin sensitivity, immune function, and DNA repair — all of which have downstream effects on mitochondrial health. Melatonin, produced in darkness, has a direct mitochondrial antioxidant role separate from its sleep function.
The prescription: consistent sleep and wake times, eight hours as the target, and the elimination of blue-shifted artificial light after dark. A cool sleeping environment — around 65°F (18°C) — amplifies sleep entry by supporting the drop in core body temperature that initiates deep sleep; a warm bath before bed triggers the same effect through compensatory cooling afterward. Chronotype varies — some people are constitutionally earlier or later — but the prescription for consistency, darkness, and cool is universal.
Tier 2: Going Further — The $200 Panel
The universal stack captures most of the available gain. The following tests unlock the remaining decisions and are available direct-to-consumer through Quest Diagnostics or LabCorp in most US states without a physician’s order. New York, New Jersey, and Rhode Island require a physician order; in those states, any primary care physician will order these if asked. Total cost out of pocket runs approximately $200–275.
Ferritin and Iron Panel (~$30)
Iron sits on a U-curve with serious consequences at both ends. Deficiency impairs mitochondrial function and oxygen transport; excess drives oxidative stress and organ damage. Roughly one in two hundred people of Northern European descent carry HFE variants associated with hereditary hemochromatosis, which causes iron accumulation. Ferritin and transferrin saturation are the relevant markers. This test should precede any iron supplementation.
Fasting Glucose and Insulin (~$40)
These two together — not glucose alone — reveal insulin sensitivity. Fasting glucose can be normal while fasting insulin is elevated, indicating compensated insulin resistance: the pancreas is working harder than it should to maintain blood sugar. The result can be made immediately actionable by calculating HOMA-IR: multiply fasting insulin (µIU/mL) by fasting glucose (mg/dL) and divide by 405. A result below 1.0 is optimal; above 2.0 indicates meaningful insulin resistance and signals that Tier 1 protein and dietary interventions need to be accompanied by carbohydrate reduction.
Homocysteine (~$40)
Elevated homocysteine is an independent cardiovascular and neurological risk factor, and a functional proxy for methylation capacity that cuts through the complexity of multiple methylation-related gene variants (MTHFR, MTR, MTRR, COMT) without requiring genetic testing. If elevated, methylated B vitamins at higher doses are indicated. If normal, the Tier 1 prescription is sufficient.
hs-CRP (~$20)
High-sensitivity C-reactive protein is a low-cost marker of systemic inflammation that tracks closely with mitochondrial oxidative stress and cardiovascular risk. It adds meaningful context to the panel at minimal cost and can reveal smoldering inflammation that the other markers don’t capture directly.
Uric Acid (~$15)
Often overlooked, rising uric acid is an early signal of metabolic dysfunction and mitochondrial stress — it reflects fructose-driven ATP depletion and tracks with insulin resistance before glucose and insulin values become obviously abnormal. Modifiable with diet (primarily by reducing fructose and alcohol), and cheap to test.
Thyroid Panel with Antibodies (~$80)
Thyroid dysfunction mimics and exacerbates mitochondrial symptoms: fatigue, cognitive fog, weight dysregulation, mood disruption. Thyroid antibodies (TPO and thyroglobulin) reveal autoimmune thyroiditis (Hashimoto’s), which affects iodine supplementation decisions and signals that inflammatory interventions are particularly important. Many cases of subclinical hypothyroidism are missed when only TSH is tested; a full panel includes free T3 and free T4.
Vitamin D (25-OH) (~$50)
Inexpensive confirmation of whether baseline supplementation is sufficient or whether higher doses are warranted. Target range is generally 40–60 ng/mL; many people supplementing at standard doses remain below this.
Tier 3: The Genetic Layer
Two genetic variants change the protocol meaningfully enough to warrant testing. Both are available through 23andMe, AncestryDNA (with third-party interpretation via tools like Genetic Genie), or direct genetic testing services.
APOE4
APOE4 is the strongest known genetic risk factor for late-onset Alzheimer’s, carried by roughly 25% of the population in one copy and 2–3% in two copies. Its effect in this framework is not simply to accelerate the clock — it impairs lipid transport to neurons, directly disrupts mitochondrial function, and increases neuronal oxidative stress. APOE4 carriers appear to require higher inputs to achieve the same mitochondrial output as non-carriers, and evidence suggests they are particularly sensitive to sleep disruption and circadian misalignment — which amplifies the already-strong sleep recommendation for this group.
Specific protocol adjustments: higher DHA supplementation, a lower-glycemic dietary pattern emphasizing unprocessed sources, and more aggressive attention to sleep quality and inflammatory tone. There is ongoing research into whether ketone-supportive eating patterns provide additional benefit by offering neurons an alternative fuel source; the evidence is mixed and this is an area to watch rather than a settled recommendation.
MTHFR
MTHFR variants reduce the efficiency of folate conversion, with downstream effects on methylation, homocysteine levels, neurotransmitter synthesis, and mitochondrial CoQ10 production. Roughly 40% of the population has some degree of reduced MTHFR function, and the picture is further complicated by other methylation-related variants (MTR, MTRR, COMT) that interact with it. Homocysteine in Tier 2 cuts through that complexity as a functional readout; genetic testing confirms the variant and its severity. MTHFR-positive individuals may benefit from higher doses of methylated B vitamins and closer attention to status over time.
Limits and Open Questions
What this framework does not account for. Genetic risk beyond APOE4 and MTHFR is real but diffuse — dozens of variants contribute small effects, and the lifestyle interventions in Tier 1 dominate over most of them. Environmental toxin load (heavy metals, persistent organic pollutants) creates mitochondrial stress through pathways outside this framework. Psychosocial inputs — chronic psychological stress, social connection, cognitive engagement — are genuine signal domains that affect inflammation, mitochondrial function, and neurodegeneration risk; they fit the framework but belong in a separate treatment. Serious psychiatric conditions, autoimmune disease, and established neurodegenerative disease require clinical management that this framework complements but does not replace.
The fat-soluble accumulation caveat. Vitamins A, D, E, and K are fat-soluble and accumulate in tissue. Vitamin D and K2 at the doses described here are appropriate for healthy adults; very high doses sustained over time warrant monitoring. Vitamin A as preformed retinol (from liver or high-dose supplements) is genuinely toxic at excess and teratogenic in pregnancy — plant-derived beta-carotene is self-limiting and safe. Vitamin E as isolated alpha-tocopherol at high supplemental doses has shown adverse signals in some populations; mixed tocopherols from food are fine.
Kidney disease changes the equations. Magnesium clearance, protein metabolism, and several supplement recommendations above are predicated on normal renal function. Anyone with known kidney disease should work through these decisions with a physician.
What we genuinely do not know. The optimal protein intake by age, sex, and activity level remains contested. The long-term effects of sustained mitochondrial stimulation are unknown — the research establishing mitochondrial dysfunction as causal in neurodegeneration is recent and primarily in animal models. The degree to which longevity effects observed in specific populations are driven by diet versus movement versus social connection versus chronic stress reduction versus genetics has not been cleanly separated. What can be restored versus merely slowed — the reversibility question — is genuinely open. The magnitude of morbidity compression achievable through this framework in humans remains to be established.
The signal-ecology thesis as hypothesis. The central claim — that a meaningful fraction of late-life cognitive and metabolic decline is a demand-signal deficiency disease rather than inevitable degeneration — is supported by converging evidence across multiple research areas but not yet proven in prospective human trials at the scale required to be definitive. The interventions described here have strong individual evidence bases. The unified framework is a synthesis, and syntheses can be wrong in ways their components are not. Nothing here is a panacea, and individual response will vary. But the safety profile of these interventions is high, the mechanistic rationale is coherent, and the cost of not implementing them — given what the evidence already shows — seems higher than the cost of acting on an incomplete picture.
cafebedouin.org 