Sage — A Metabolic Hypothesis of Grapefruit
Companion to Sage — Rapamycin (physician application only; see also Coach — Rapamycin). The Dual Pathway week this essay sits inside is explained on Sage — Strength (built on the Frugal Dual-Pathway Training Protocol). Pharmacokinetic figures appear only as citation context.
ENSURF DUAL-PATHWAY FRAMEWORK
A Metabolic Hypothesis of Grapefruit
Placement of a short AMPK accent on the repair side of a seven-day cycle. Not a rapamycin substitute.
Disclaimer
This is a metabolic hypothesis, not a protocol, a prescription, or a claim of benefit. It is not medical advice. Nothing described here should be done except under the supervision of a physician who knows the full medication list, the dose, and the labs.
Drug-interaction warning. Grapefruit irreversibly inactivates intestinal CYP3A4 and can inhibit intestinal P-glycoprotein. The enzyme must be resynthesized. The effect begins within hours and, on average, lasts about three days. Grapefruit juice must not be taken with Rapamune (sirolimus). In clinical work, about 8 ounces of grapefruit juice raised sirolimus exposure on the order of three- to five-fold, with wide individual spread. The same mechanism raises blood levels of many other drugs, including certain statins (simvastatin, lovastatin, and atorvastatin), several calcium-channel blockers, other immunosuppressants, and additional CYP3A4 or P-gp substrates. Do not combine grapefruit with rapamycin, or with any interacting drug, except under the directing physician. Grapefruit is not a substitute for rapamycin.
Hypothesis
Within a seven-day EnSurf cycle that separates a brief anabolic pulse from a longer repair phase, one grapefruit on the mornings of Days 3, 4, and 5 adds a short, upstream AMPK accent on top of residual mTORC1 inhibition. It does not reproduce rapamycin. It does not multiply the dose just taken. Timed this way, it should not multiply the next dose either.
The claim is narrow. Food-dose naringenin is a low-amplitude, few-hour signal through AMPK, not an FKBP12–mTOR ligand. The useful question is placement, not equivalence.
The cycle, numbered from the build day
Day 1 is the build day, whichever calendar day that is.
- Day 1 morning. Strength work and a leucine-rich meal. This is the mTOR pulse, aimed at muscle.
- Day 2 night. Low-dose rapamycin. Repair opens after the build, not before it.
- Days 3, 4, and 5, morning. One grapefruit, fasted, with morning light. No other juice.
- Days 6 and 7. Empty. No grapefruit.
- Next Day 1. The build pulse again.
The repair side of the cycle is the fasted morning, the low-leucine window, zone-2 work if it is done, and the declining rapamycin curve. The build side is Day 1 only. Grapefruit belongs on the repair side.
Two actions, only one of which is the hypothesis
Grapefruit is easy to misread because it does two unrelated things.
The interaction is furanocoumarins, mainly bergamottin and 6',7'-dihydroxybergamottin. They irreversibly inactivate intestinal CYP3A4. New enzyme has to be made. Recovery proceeds with a half-time of about a day. In the midazolam recovery study, exposure was on average within about 6 percent of baseline by 74 hours, with slower recovery in some people. A glass taken with oral rapamycin raises exposure on the order of three- to five-fold. The figure most often cited, a 350 percent increase, comes from daily 8-ounce juice in a cancer dose-finding study, not from a single glass with one low dose. That is a bioavailability effect. It is not mTOR inhibition by the fruit. Hepatic CYP3A4, which clears drug already in blood, is barely touched by one glass and begins to yield only when juice is heavy and repeated. This interaction is a boundary condition on timing. It is not the proposed benefit.
The signal in the hypothesis is naringin, converted in the gut to naringenin, with a smaller contribution from nootkatone in the rind. Naringenin phosphorylates AMPK and, in cell and animal models, leans on mTORC1 from upstream and favors autophagy. Nootkatone activates AMPK through LKB1 and CaMKK and raises the AMP/ATP ratio, but almost all of it is in the peel, and the doses that moved energy expenditure in mice are far above what juice supplies. Pure grapefruit juice averages about 31 mg naringin per 100 ml, with a reported range of roughly 7 to 66 mg. One fruit, yielding on the order of 150 to 250 ml, therefore supplies about 50 to 150 mg naringin, roughly 25 to 70 mg of aglycone equivalents. Elimination half-life of naringenin is about 1.3 to 2.2 hours after juice and about 2.5 to 3 hours after purified compound. Peak is at about 2 to 3 hours. It is gone by the next morning. A large juice load (8 ml/kg) has produced mean peaks around 6 µmol/L, with wide scatter. One fruit is usually lower. Cell papers that clearly move mTOR often use 10 to 50 µmol/L.
So the hypothesis does not say that grapefruit is a dietary rapalog. It says that a brief AMPK nudge, placed on the falling side of a weekly rapamycin pulse, is directionally aligned with repair and temporally isolated from the build day.
Why Days 3–5
The mean terminal half-life of sirolimus is 62 ± 16 hours after multiple dosing in stable renal transplant patients, with a commonly cited range of 46 to 78 hours. Tablet time to peak is about 1 to 6 hours, so a Day 2 night dose is absorbed well before the Day 3 morning. That dose is still early in its elimination curve on Day 3 morning and near one half-life out by Day 5 morning. A low weekly dose is closer to trough by Day 6 or 7, not at zero. Grapefruit on Days 3–5 therefore stacks on residual mTORC1 inhibition. That is accepted in the hypothesis. The point of those mornings is to deepen repair, not to open a drug-free AMPK window.
It does not potentiate the dose just taken. The exposure jump requires furanocoumarins on board before absorption. By Day 3 morning that dose is in. One fruit will not multiply it. Three mornings of one fruit may slightly slow clearance. Thrice-daily juice for three days has inhibited hepatic CYP3A4 and lengthened half-life; once-daily for three days is less exposure than that. The expected effect is a mild drag on the tail, not a second bioavailability spike.
It should not potentiate the next dose. Day 5 morning to the next Day 2 night is about 4.5 days, past the roughly 72-hour average recovery of intestinal CYP3A4. Recovery is slower in some people, which is why the gap is longer than three days and why a Day 6 grapefruit is avoided. Days 6 and 7 stay empty for that reason.
Day 1 stays clear of the flavonoid. The last grapefruit is two days behind the next strength and leucine pulse. Naringenin is gone. The build signal is not riding a few-hour AMPK accent. Residual drug from the prior Day 2 night is near trough, which is the reason the pulse is weekly.
What this is not
It is not a rapamycin-sparing strategy. Lowering the pill and adding juice is a different hypothesis, and a less predictable one, because the multiplier varies several-fold between people.
It is not an evening intervention. The half-life is hours. Taken at night it would sit closer to sleep and closer to the next build morning, for no gain.
It is not more effective at two or three fruits. More juice is what begins to reach hepatic CYP3A4 and lengthen the rapamycin tail. The hypothesis uses one fruit.
It is not a claim about lifespan, lipids, glucose, or immune function. Those belong to rapamycin, and they are monitored, not inferred from a citrus flavonoid.
Boundary conditions
Physician supervision is required. The medication list decides whether grapefruit is allowed at all. The labeled instruction for Rapamune is that grapefruit juice must not be taken with it. Statins, many calcium-channel blockers, other immunosuppressants, and other CYP3A4 or P-gp substrates are the same practical hazard.
No grapefruit on Day 2, and none within the day of the pill. Morning only. Days 3–5 only. Days 6–7 empty.
The build meal stays leucine-rich and grapefruit-free. The repair mornings stay fasted or low-protein until the flavonoid peak has passed.
White and Marsh fruit tend to run higher in naringin than some red cultivars. Fruit-to-fruit spread is large. Zest adds nootkatone and is optional; it is not required for the hypothesis.
People who start the cycle on a different weekday use the same numbering. Day 1 is the strength day. The calendar label does not matter.
What would count against it
- A rise in the next trough, or in a level drawn the morning after the next Day 2 dose, would mean Days 3–5 were not clean of the interaction, and the grapefruit would come out.
- Loss of the Day 1 training response, heavier soreness, or a fall in strength across cycles would mean the repair accent was bleeding into the build. The first correction is to drop Day 5, not to add juice.
- No change in any repair-side marker, and no subjective difference in the fasted mornings, would mean food-dose naringenin is below a meaningful AMPK signal in that person. The hypothesis allows that result. Grapefruit is then only a food, and the cycle does not need it.
Statement
Grapefruit is a timed AMPK accent inside an already-separated dual-pathway week. Days 3–5 morning, one fruit, fasted. It rides the falling rapamycin curve, stays off the build day, and stays off the next pill. Anything beyond that placement is a different hypothesis, and the interaction is the reason it stays under a physician.
A working hypothesis, open to correction and revised as new evidence arrives. Last revised October 17, 2026.
Related: Strength (Dual Pathway week) · Rapamycin (physician application) · Rapamycin Coach · Broccoli Sprouts · Coach — Grapefruit · Surfer — Kit
See the notice at the top of this page. Working hypothesis, not medical advice and not a dosing recommendation. The timing above is the essay’s own placement, as written, and it exists only inside a physician-directed rapamycin week. No rapamycin dose is given here; the weekly dose is physician-guided only. Grapefruit raises sirolimus exposure and blood levels of certain statins, several calcium-channel blockers, other immunosuppressants, and other CYP3A4 or P-gp substrates. The medication list decides whether grapefruit is allowed at all.