N=1 After 40

Current experiment · September 2026

High-Dose Melatonin: 300 mg

At sleep doses, melatonin is a timing signal. At hundreds of milligrams, research interest shifts to its antioxidant and oxidative-stress effects. I am testing whether 300 mg a night changes a measurable marker of oxidative damage in me. Personal experiment record, not medical advice, prescribing, sourcing guidance, or an instruction to copy.

Primary question: After nine nights at 300 mg, is my urinary F2-isoprostane level, a marker of lipid oxidation, lower than before the first dose, with liver markers still in range?

The Experiment

Current record

Added to the active experiment ledger on September 30, 2026. The run has two parts: a one-night tolerance test, completed on September 29, and a nine-night block with bloodwork before and after, planned for late October. The 300 mg doses come from PureBulk bulk melatonin powder, filled into capsules.

Why 300 mg

Sleep and jet-lag use typically involves a few milligrams, which is already enough to reach or exceed normal night-time blood levels. The interest in hundreds of milligrams is different: very high tissue concentrations for the effects melatonin shows beyond its receptors in laboratory work, including direct free-radical scavenging and antioxidant breakdown products. This experiment is about that question, not about sleep.

300 mg is the dose with the closest human oxidative-stress precedent. I chose it over 400 mg because nothing shows that the extra 100 mg adds antioxidant benefit.

Night 1: Tolerance Test

September 29, 2026

I took a single 300 mg dose. The only question was how I tolerate it; sleep quality and antioxidant effects were not being tested yet.

I woke at my usual time with no sleep hangover, grogginess, or headache, and nothing felt different getting out of bed. I did not notice deeper sleep, extra sedation, or unusual dreams. The one noticeable effect: I felt unusually cold for the first couple of hours after falling asleep.

Measure Baseline (median, 7 September mornings) Morning after 300 mg
Sleep duration 6:00 6:00
Sleep efficiency 82% 82%
Resting heart rate 61 bpm 62 bpm
HRV 41 ms 45 ms
Morning reaction time (PVT mean) 348 ms 375 ms

Context

Sleep and recovery values are from WHOOP. Night 1 fell during a separate 10-day Cortexin course, between the Day 8 and Day 9 injections, and during an Anavar course. Reaction time had already been running slower on the Cortexin mornings (371 ms the morning before), so the 375 ms reading belongs to that context rather than to melatonin alone.

Result

Tolerated. Sleep, resting heart rate, and HRV matched my baseline, with no next-day impairment. The cold feeling is the signal I will track more carefully during the block.

Starting Point

DNA: melatonin receptor (MTNR1B)

My FitnessGenes DNA report flags a variant in MTNR1B, the gene for a melatonin receptor that is also active in the insulin-secreting cells of the pancreas. The report says the variant makes my tissues more sensitive to melatonin and lowers insulin further at night, with a slightly increased risk of higher blood sugar while asleep. It recommends avoiding late-night meals and says melatonin supplements may push insulin lower still. The report does not name the exact variant.

That shaped the dose timing. The 7:30 pm dose comes about ten hours after my daily large meal. The concern in the research is food arriving while melatonin is high, so keeping the dose far from meals is the practical response.

September bloodwork

These are the values from my September 16, 2026 fasting panel that relate to this experiment. The draw was during the Anavar course and after heavy training the day before.

Marker September 16, 2026 Reference range
AST 64 U/L (high) 10–40 U/L
ALT 32 U/L 9–46 U/L
ALP 59 U/L 36–130 U/L
Total bilirubin 0.5 mg/dL 0.2–1.2 mg/dL
GGT 7 U/L 3–95 U/L
CK (total) 1,496 U/L (high) 26–366 U/L
Fasting glucose 81 mg/dL 65–99 mg/dL
Insulin 2.1 uIU/mL ≤18.4 uIU/mL
HbA1c 5.0% <5.7%

What these show

ALT, ALP, bilirubin, and GGT were in range. AST was high alongside a CK of 1,496, which the lab verified by repeat analysis. AST rising with CK while ALT stays normal points to a muscle contribution from training, which is why CK is measured at both visits. Because this draw was during Anavar, the October 15 panel and Visit 1 are the reference points for the block.

Fasting glucose, insulin, and HbA1c were all in range, so the MTNR1B flag has not shown up in my daytime glucose markers. It concerns night-time glucose, which is why the dose timing matters more than these numbers.

The Nine-Night Block

Timing and go/no-go check

The block starts after the Cortexin course and its follow-up days, and after the Anavar course ends on October 17. Bloodwork on October 15 falls near the end of Anavar and serves as the go/no-go check: the block starts only if ALT, ALP, GGT, and bilirubin are in range, with AST read alongside CK. My routine stays unchanged from October 18 through the final visit: no additions, removals, or dose changes.

Schedule

Visit 1: the first Zone 2 morning on or after October 21. The first 300 mg dose is that evening.

Nights 1–9: 300 mg at 7:30 pm, about ten hours after the day's large meal and 30 minutes before bed.

Visit 2: the next Zone 2 morning, nine days later and the morning after night 9. Dosing stops.

Both visits fall on the morning after a Pull day in my 9-day training rotation, so the previous day's training is the same for each sample.

Test Visit 1 (before) Visit 2 (after night 9)
F2-isoprostane/creatinine ratio, urine (primary) Yes Yes
Hepatic function panel: ALT, AST, ALP, total and direct bilirubin, albumin, total protein Yes Yes
GGT Yes Yes
CK (total) Yes Yes
Fasting lipid panel and ApoB No Yes

Collection conditions

Both visits are at Quest, in the first morning slot, before training. Water only beforehand, the same wake-up routine both mornings, caffeine and pre-workout after the visit, no alcohol for 48 hours, and normal hydration. CK is there because hard training also raises AST. No acetaminophen during the block.

Daily log during the block

WHOOP sleep duration, sleep efficiency, resting heart rate, HRV, respiratory rate, and skin temperature; a 0–10 cold rating on waking; the morning reaction-time test; and any headache, dizziness, or next-day grogginess.

Stop rules

During the block, symptoms are the stop signal: dark urine, itching, yellowing skin or eyes, or pain under the right ribs mean I stop and get checked. If Visit 2 shows ALT above range and more than double the Visit 1 value, or bilirubin, ALP, or GGT above range, I repeat the panel until it normalizes and do not extend or repeat the block.

How the result will be read

The primary readout is the change in urinary F2-isoprostane/creatinine from Visit 1 to Visit 2. With one sample on each side, a small change will be recorded as no clear signal; a drop well beyond ordinary day-to-day movement counts as a signal. The block starts after Anavar ends, so that context sits beside the result. The liver markers answer the tolerability question for nine nights. The Visit 2 lipid panel and ApoB also give a post-Anavar lipid reading.

Research Summary

300 mg in people

In a study of 31 people with ALS, 300 mg/day of melatonin given rectally was well tolerated for up to two years. Serum protein carbonyls, a marker of oxidative protein damage, were elevated in the ALS patients and returned to control values during treatment. This is the closest human precedent for the dose and the oxidative-stress question. Weishaupt et al., 2006.

Antioxidant markers in trials

A meta-analysis of 12 randomized placebo-controlled trials in patient populations found that melatonin increased total antioxidant capacity, glutathione, and the antioxidant enzymes SOD, glutathione peroxidase, and glutathione reductase, and reduced malondialdehyde, a lipid-oxidation product. Morvaridzadeh et al., 2020.

Why F2-isoprostanes

F2-isoprostanes form when free radicals oxidize arachidonic acid in cell membranes. In the NIH-led Biomarkers of Oxidative Stress Study, a multi-laboratory validation effort, isoprostanes in plasma and urine were among the few candidates that reliably rose with oxidative injury, alongside malondialdehyde and urinary 8-OHdG. Measuring them in urine against creatinine corrects for how dilute the sample is. Kadiiska et al., 2005.

Why the liver panels

A systematic review of 79 randomized trials using at least 10 mg (3,861 participants) found no detectable increase in serious adverse events in the four low-bias trials, but more drowsiness, headache, and dizziness. Menczel Schrire et al., 2022.

The MELATOMS-1 trial gave 300 mg/day orally to people with primary progressive multiple sclerosis alongside ocrelizumab. It was stopped early after liver enzymes rose in 3 of 4 melatonin-treated participants and none of 4 on placebo, first detected at the 1- and 3-month checks. The rise returned when one participant restarted melatonin. The authors point to competition with other medications in shared liver drug-processing pathways, and the trial restarted at 100 mg/day with closer monitoring. That is why liver markers are measured at both visits. MELATOMS-1, 2026.

Melatonin, food, and MTNR1B

In a randomized crossover trial of 845 people, an evening glucose test taken one hour before habitual bedtime, when melatonin was 3.5 times higher, produced lower insulin and higher glucose than the same test four hours before bedtime. The effect was stronger in carriers of the MTNR1B risk variant. Garaulet et al., 2022. In a smaller study of 17 women, melatonin before a morning glucose test worsened glucose tolerance about six times as much in risk-variant carriers as in non-carriers; the evening difference was not significant. Garaulet et al., 2015. Both studies gave food or glucose while melatonin was high, the combination my dose timing avoids.

The cold feeling

Melatonin increases heat loss through the skin and slightly lowers core body temperature, mostly in the first hours after a dose, which matches when I felt it. The size of the drop levels off quickly: a meta-analysis of 30 datasets found about 0.2 °C at 5 mg and little additional drop above that. In a separate dose study, 1 mg and 5 mg lowered temperature by a similar amount, but only 5 mg kept it down for the whole session. Those studies used daytime doses far below 300 mg. Marrin et al., 2013; Dawson et al., 1996.

Sources