Precursor · The oral route with the most data
Nicotinamide Riboside (NR): The Most-Studied Oral NAD+ Precursor
The precursor that refills the pool sirtuins and PARPs draw from — and the one with the cleanest dose-response data in healthy adults.
In plain English
Nicotinamide riboside (NR) is a building block — a precursor — that the body converts into NAD+ (the fuel-handling helper molecule cells use to make energy). It belongs to the vitamin-B3 family, and it is the precursor that has been tested most in people. The headline finding is clean: swallowing more NR raises the NAD+ level in blood, and it does so in proportion to the dose. NR is the rational oral choice precisely because plain "NAD+" capsules are poorly absorbed. This page summarizes the research on NR; it is not advice and gives no dosing instructions.
Why Most Oral NAD+ Products Are Precursors
NAD+ itself is a large, charged dinucleotide (molecular weight 663.43 Da), and it is poorly taken up by cells intact, so an oral "NAD+" capsule is an inefficient way to raise the coenzyme. The body's own logistics solve this with precursors: smaller molecules that are absorbed, enter cells, and are then assembled into NAD+ [5]. That is the whole reason the NAD+ supplement aisle is really a precursor aisle.
NR takes a distinctive route in. It is converted to NMN by NRK kinases and then to NAD+, entering the pathway downstream of NAMPT — the rate-limiting salvage enzyme — rather than competing for it [5]. Niacin and nicotinamide feed NAD+ by other routes, but NR is the precursor with the most controlled human pharmacodynamic data behind it.
The dose-response data
The defining NR study is a randomized, double-blind, placebo-controlled trial in healthy overweight adults. NR at 100, 300 and 1000 mg/day for 8 weeks raised whole-blood NAD+ by 22%, 51% and 142% respectively, and the elevation was maintained throughout the study [4]. Just as important for a daily supplement, NR produced no flushing, did not elevate LDL cholesterol, and did not disrupt one-carbon metabolism, with no significant adverse-event difference from placebo at any dose [4].
That clean, scalable, well-tolerated dose-response is why NR is the reference oral NAD+ booster in the literature. It is also why "does taking NAD orally work?" has a precise answer: plain NAD+ does little, but NR reliably moves the blood number.
NAD+ vs Its Precursors (NMN, NR)
The most common confusion in this topic — NAD+ vs NMN — is treating "NAD+," NMN and NR as interchangeable. They are not. NAD+ is the destination molecule; NMN and NR are the precursors the body converts into it. Oral NAD+ itself is poorly absorbed, so it is rarely what a swallowed product effectively delivers [5].
NMN sits one step from NAD+ and has its own randomized human data — 250 mg/day improved muscle insulin sensitivity in prediabetic women [1], and 300–900 mg/day raised blood NAD+ in a multicenter trial with 600 mg/day optimal [3]. NR sits one step further out but enters salvage by a different door [5], and carries the cleanest dose-response curve [4]. NMN's dietary-supplement status is contested by the FDA; NR's is not — a regulatory difference worth keeping straight, covered in full on the NAD injection and IV NAD+ page's status discussion and across the common questions about NAD+.
What NR has been studied for beyond blood levels
Beyond raising NAD+, NR's broader effects are mostly preclinical so far. In mice, NR given before noise exposure recovered cochlear hair-cell ribbon synapse counts and protected against noise-induced hearing loss through enhanced oxidation resistance [12]. NR has been safety-tested at high doses in clinical research — up to 3000 mg/day in the NR-SAFE study in Parkinson's disease — which speaks to tolerability, not to a proven disease benefit [research context].
The pattern mirrors the wider field: the blood-NAD+ pharmacodynamics are solid and human, while the downstream clinical outcomes lean on animal models and remain preliminary in people [13]. NR is the best-characterized precursor, and even for NR the honest summary keeps the measured and the unproven on separate cards.