NR

Nicotinamide Riboside (Niagen)

NR (Nicotinamide Riboside (Niagen)) is a longevity & bioregulators research compound. NAD+ precursor (a vitamin B3 form) — converts to NMN then NAD+ to support mitochondrial energy and sirtuin activity.

How NR works

NAD+ precursor (a vitamin B3 form) — converts to NMN then NAD+ to support mitochondrial energy and sirtuin activity.

Proposed benefits

Researched for mitochondrial and cellular-energy support, tissue-specific bioregulation and healthy-aging pathways.

Where to get NR

The evidence for NR

Graded by what exists behind each claim.

✅ Clinically validated

• Multiple randomized human trials, and the results are consistent in one direction and disappointing in another. NAD+ levels rise reliably and dose-dependently — that part is settled across trials.
• What has not shown up is clinical benefit. Trials in healthy older adults, obese adults and heart failure have generally failed to demonstrate meaningful improvements in insulin sensitivity, mitochondrial function, muscle performance or exercise capacity. A few small studies report reductions in inflammatory markers.
• This is the most important honest read on the whole NAD+ category: the biomarker moves, the outcome largely has not. That is a real finding, not a reason to dismiss the biology — it may mean the endpoint, population or duration was wrong. But it is where the evidence currently stands.

📊 Correlative data

• Very wide use with an excellent safety record across trials at doses up to 1,000 mg daily. Reported experience is subtle — most users cannot tell, which is consistent with the trial results.

🧪 Theoretical / extrapolated

• Nicotinamide riboside is an NAD+ precursor. NAD+ declines with age and is the required cofactor for the sirtuins and PARPs — the enzymes doing DNA repair and metabolic regulation — so restoring it is a coherent target.
• NR enters the salvage pathway via NRK1/NRK2 kinases, a route NMN and nicotinamide do not use, which is the mechanistic argument for it over the alternatives.
• The gap between raised NAD+ and absent benefit is the interesting question. Possibilities include tissue distribution (blood NAD+ may not reflect muscle or brain), that NAD+ was not the limiting factor in healthy people, or that the trials were too short. All are testable; none are resolved.

How to read these tiers: they say how much human evidence exists, not how well something works — and ✗ flags harm, never a disappointing trial. How the evidence tiers work →

What NR actually does

Nicotinamide riboside is the one NAD+ precursor with a dedicated enzyme, and that enzyme is the entire argument for it. Most routes into the NAD+ pool converge on NAMPT, the rate-limiting salvage enzyme that converts nicotinamide to nicotinamide mononucleotide. NAMPT is feedback-inhibited by nicotinamide itself, which is why swallowing more nicotinamide is a poor way to raise NAD+. NR does not use that door. It is phosphorylated directly to NMN by the nicotinamide riboside kinases NRK1 and NRK2, and NMNAT finishes the job. One step, bypassing the regulated one.

NRK2 is the detail worth knowing, because it is where the tissue specificity comes from: NRK2 is concentrated in skeletal and cardiac muscle and is upregulated by muscle stress and damage. A precursor whose activating kinase is enriched in muscle and induced by injury is a plausible muscle-directed agent in a way that nicotinamide is not, and it is the mechanistic reason the muscle trials were the ones worth running.

The evidence that this route is real in a person is direct. Trammell 2016 is the first-in-human pharmacokinetic work, and its title states the finding: NR is uniquely and orally bioavailable in mice and humans. Oral NR raises the blood NAD+ metabolome. That is the claim, it was measured, and it holds.

What that claim does not include is the second half. Bioavailability is a statement about the molecule arriving. Whether arriving changes anything is a separate question with a separate answer, and this page is largely about the gap between the two. Cantó 2022 makes exactly that point about the whole precursor class — its title calls their redundancy questionable — and the reason it can is that all of them terminate in the same pool, and that pool is drained by the same ectoenzyme, CD38, whose expression rises with age Zeidler 2022. Filling a leaking tank faster is a real intervention. It is not the same intervention as fixing the leak.

Cell, rodent, human — and where it stops

Step one, pharmacokinetics in humans, and it is clean. Trammell 2016: oral NR raises blood NAD+ in people, with a characterized metabolite pattern. This is a compound whose target engagement is not in question, which puts it ahead of most of this Vault.

Step two, chronic dosing in healthy adults. Martens 2018 gave NR for 6 weeks to healthy middle-aged and older adults: well tolerated, and whole-blood NAD+ went up. The biochemical endpoint was met. The clinical endpoints in that literature have been far more modest than the biochemistry, and saying so is not a criticism of the trial — it is the trial's own result.

Step three, in disease, where the endpoints get harder. The NADPARK study randomized NR in Parkinson's disease and reported evidence of cerebral target engagement Brakedal 2022; the follow-up NR-SAFE trial tested high-dose NR double-blind and its purpose was safety rather than efficacy Berven 2023. Note what that sequence tells you: the field moved to dose escalation because the standard dose had not produced a clinical effect worth the trial.

Step four, the lifespan question, answered and rarely quoted. The Interventions Testing Program is the most rigorous lifespan platform in rodent biology — genetically heterogeneous UM-HET3 mice, three sites, both sexes. Harrison 2021 reports that 17-alpha-estradiol extended lifespan in males and that nicotinamide riboside and three other drugs did not affect lifespan in either sex. The longevity claim attached to this compound has been tested in the best available animal model and it failed. That is the single most important sentence about NR and it appears on almost no page selling it.

The obstacles, named one at a time. (1) Raising blood NAD+ is not the same as raising NAD+ in a neuron or a myocyte; blood NAD+ sits overwhelmingly in red cells. (2) The drain side is untouched: supplying precursor does nothing about CD38 Zeidler 2022. (3) The healthy-population trials are short — weeks — against a phenotype that develops over decades. (4) The one long, well-powered, hard-endpoint experiment that has been done was negative Harrison 2021.

What would have to be true, and how you would know it was not

Three predictions. The first is the one that would settle whether this compound does anything for the person taking it, and the second is the one that argues against the dose creep the market has drifted into.

1. If NR is doing metabolic work, the metabolic markers should move — and in the healthy-adult trials they largely have not. Draw fasting insulin and HbA1c at baseline and at 12 weeks, alongside a lipid panel. The precursor arrives Trammell 2016 and blood NAD+ rises Martens 2018; if insulin sensitivity and glycemia are unchanged after three months, the honest conclusion is that the pool being fuller is not, by itself, doing anything you can measure. That is the outcome the published record predicts, and printing the prediction that way is the difference between a research page and an advertisement.

2. Higher is not obviously better, and hs-CRP is one place a wrong dose could show. The field escalated to high dose because the ordinary dose did little Berven 2023, and gram-level nicotinamide exposure is not biologically neutral — the downstream methylation load and the immunomodulatory effects of nicotinamide are both real. A hs-CRP and a CMP before and after a high-dose run are the cheap instruments, and a person running 1 g or more daily has more reason to draw them than a person on 300 mg.

3. The falsification test for the muscle hypothesis is performance, not chemistry. The NRK2 argument predicts a muscle-selective effect. So test muscle: grip strength and a repeatable submaximal VO2 or time-trial measure at baseline and at 12 weeks, with training volume held constant — because training will move both of those far harder than this will, and an uncontrolled training block is how a supplement gets credit for a mesocycle.

What nobody has tested yet

Four experiments that have not been run and are within reach of an ordinary research budget.

Nobody has run NR against a CD38 inhibitor head to head in humans. The mechanistic argument on this page is that the drain matters at least as much as the tap Zeidler 2022. A three-arm study — NR, a dietary CD38 inhibitor, both — with whole-blood NAD+ as the endpoint would test that directly, and nothing about it is expensive.

Nobody has stratified NR responders by NRK2 expression. NRK2 is the muscle kinase that activates NR and it is inducible. A muscle biopsy substudy in an exercise trial would answer whether the people who respond are the people whose muscle can phosphorylate the compound — which is the most obvious personalization question this molecule has and has never been asked.

Nobody has repeated the lifespan test with an NAD+ readout. Harrison 2021 reported no lifespan effect, but the mice were not assayed for whether tissue NAD+ actually rose. A null result without a target-engagement measurement leaves two explanations alive — the pathway does not extend lifespan, or the dose never reached the tissue — and distinguishing them is one added assay.

Nobody has tested timing. NAD+ is under strong circadian control through NAMPT, which oscillates across the day. Whether a precursor given at the trough behaves differently from the same dose at the peak has never been tested in a person, and it is a crossover study with a single blood metabolite panel.

NR — its own safety story, not its class's

NR is among the better-tolerated compounds in this Vault, and the specific things worth saying are not the generic ones.

The safety data are real and they are recent. Berven 2023 is a randomized, double-blind, high-dose safety trial — the design that exists specifically to find harms, run in a patient population rather than in healthy volunteers. Most supplements on this site have nothing comparable. Citing it is the strongest safety statement this page can honestly make, and it is stronger than "generally recognized as safe".

What it does not cover is years. Trials run weeks to months. People take this for a decade on a longevity rationale that the longest hard-endpoint experiment did not support Harrison 2021. The mismatch between the duration studied and the duration used is the real open question, and it is not resolvable by any trial that has been run.

The methyl-group question belongs here and is usually overstated in one direction and ignored in the other. All NAD+ precursors ultimately generate nicotinamide, and disposing of nicotinamide consumes methyl groups. At 250–300 mg this is unlikely to matter in a person with adequate folate and B12. At gram-level chronic dosing it is a real chemistry with a cheap measurement attached, and the reasonable position is to measure rather than to argue.

The interaction nobody flags. NR raises NAD+ availability, and PARP enzymes consume NAD+ during DNA repair. In someone taking a PARP inhibitor for cancer, the theoretical direction of that interaction is unknown and unstudied — flagged here as extrapolation from enzymology, not as an observed event, because it is exactly the kind of question that should be asked of an oncologist rather than discovered.

Sources read for this page

• Trammell SA, et al. Nicotinamide riboside is uniquely and orally bioavailable in mice and humans. Nature Communications 2016 · PMID 27721479
• Berven H, et al. NR-SAFE: a randomized, double-blind safety trial of high dose nicotinamide riboside in Parkinson's disease. Nature Communications 2023 · PMID 38016950
• Harrison DE, et al. 17-a-estradiol late in life extends lifespan in aging UM-HET3 male mice; nicotinamide riboside and three other drugs do not affect lifespan in either sex. Aging Cell 2021 · PMID 33788371

NR — safety, predicted from mechanism

NR — interference & stacking

Bloodwork to run alongside NR

Baseline first, then again at 8–12 weeks.

The Longevity Baseline panel covers these in one order — 13 markers, $219.10 with the discount applied.

Check results you already have → · All 103 markers A–Z

NR — frequently asked questions

NR inside a finished plan

One arm of 1 Protocol Blueprint, free to read in full.

What NR is used for

NR appears under 2 goals in the goal router.

Related Longevity & Bioregulators compounds

← Explore the full Protocol Vault

↑ Back to on this page