NAD+ — A Complete Research Reference

Nicotinamide adenine dinucleotide (NAD+) is an endogenous redox coenzyme present in every living cell. It is not a pharmacological agent in the classical sense — it is a substrate the organism manufactures and consumes continuously. The question driving the contemporary research interest is narrower: what happens when exogenous NAD+, or one of its biosynthetic precursors, is administered above the levels the body produces on its own? This reference summarizes what the peer-reviewed literature reports about NAD+ biology, the precursor strategies (NMN and NR), the two main administration routes used in research settings, and the unresolved gap between the rodent lifespan literature and the available human evidence.

Chemistry and Structure

NAD+ is a dinucleotide composed of two nucleotides joined through their phosphate groups: a nicotinamide-containing nucleotide and an adenine-containing nucleotide. The oxidized form (NAD+) carries a positive formal charge on the nicotinamide ring nitrogen; the reduced form (NADH) carries an additional hydride. NAD+ also exists in a phosphorylated variant (NADP+/NADPH) involved in reductive biosynthesis and antioxidant regeneration. Molecular weight of the free acid is approximately 663 Da.

In research settings, NAD+ is most commonly supplied as a lyophilized white-to-off-white powder, either as the free acid or as a disodium salt. The disodium form is more soluble in aqueous buffer. The two related precursors that appear repeatedly in the longevity literature are nicotinamide mononucleotide (NMN, ~334 Da) and nicotinamide riboside (NR, often supplied as the chloride salt, ~290 Da). All three molecules feed into the same pool downstream but enter the biosynthetic network at different steps.

Biosynthesis Pathways

Cellular NAD+ pools are maintained by three convergent pathways. The de novo pathway begins from tryptophan and proceeds through the kynurenine arm, ultimately yielding quinolinic acid and entering the NAD+ pool. This route is quantitatively minor in most tissues. The Preiss-Handler pathway begins from dietary nicotinic acid (niacin). The salvage pathway — the dominant route in most mammalian cells — recycles nicotinamide back into NAD+ via nicotinamide phosphoribosyltransferase (NAMPT), which catalyzes the rate-limiting step. NMN sits one enzymatic step downstream of NAMPT; NR enters the salvage pathway via nicotinamide riboside kinases (NRK1/NRK2) and is converted to NMN.

The salvage pathway matters because NAD+ is not stoichiometrically conserved during normal cell function. It is continuously consumed by three enzyme families: sirtuins (which transfer the ADP-ribose moiety during lysine deacylation), poly-ADP-ribose polymerases (PARPs, activated by DNA damage), and CD38 (a multifunctional ectoenzyme that cleaves NAD+ to generate cyclic ADP-ribose and other calcium-mobilizing metabolites). The balance between salvage synthesis and consumption by these three classes sets the steady-state NAD+ concentration in a given cell type.

Age-Related Decline

A consistent observation across rodent and human tissue studies is a decline in NAD+ levels with age. Reported declines are tissue-dependent and assay-dependent, but the directional signal is reproducible across laboratories. Proposed contributors include reduced NAMPT activity, increased CD38 expression in aging immune cells, and chronically elevated PARP activation from accumulated DNA damage. The Imai laboratory and Sinclair laboratory have published extensively on this decline and on the proposed link to age-associated metabolic and mitochondrial dysfunction. It is important to note that "NAD+ decline causes aging" is a hypothesis the field is actively testing, not an established causal relationship — reduced NAD+ may be a marker, a cause, a consequence, or all three depending on tissue context.

Precursor Strategies — NMN, NR, and Direct NAD+

The research literature distinguishes among three administration strategies, each with different pharmacokinetic implications.

• Nicotinamide riboside (NR). Orally bioavailable, well-tolerated in published human pharmacokinetic studies, and capable of raising blood NAD+ measured by LC-MS. NR is converted intracellularly to NMN and then NAD+. Most controlled human data on NAD+ precursors involve NR.
• Nicotinamide mononucleotide (NMN). A more recent focus, particularly in the Imai and Sinclair labs. Whether NMN crosses the plasma membrane intact or is dephosphorylated to NR extracellularly and re-phosphorylated inside the cell was a contested question; the Slc12a8 transporter described by Grozio and colleagues offers one proposed entry mechanism, though independent replication is ongoing. Oral NMN has been shown to raise blood markers in short-term human studies; outcome data are limited.
• Direct NAD+ administration. Because NAD+ itself is a charged dinucleotide, oral bioavailability is poor and the molecule is largely degraded in the gut and circulation. Research protocols therefore typically use intravenous infusion or subcutaneous injection. Pharmacokinetics differ markedly between routes: IV infusion produces high peak plasma exposure over the infusion window with rapid distribution, while SC administration shows a slower release profile with lower peak concentrations.

For comparative work, researchers should note that "raising NAD+ in cells" can mean very different things depending on which precursor was used and which tissue was sampled.

Animal Study Summary

The rodent NAD+ and NAD+-precursor literature is substantial. Findings most frequently cited include improvements in mitochondrial function in aged muscle, partial rescue of insulin sensitivity in diet-induced obesity models, improvements in measures of vascular function, and extension of healthspan markers (grip strength, gait, exercise tolerance) in aged mice. Bonkowski and Sinclair's 2016 review summarizes much of this work and frames the proposed sirtuin-dependent mechanism. Lifespan extension data in mice are mixed: some studies report modest extension with NMN or NR; others report no effect on maximum lifespan. The signal is more robust for healthspan endpoints than for survival itself.

A separate line of work has examined NAD+ precursors in disease models. Brakedal and colleagues reported in 2022 on a randomized trial of NR in Parkinson's disease patients (the NADPARK study), one of the few outcome-focused human trials in this space; the trial reported a modest signal on a clinical scale alongside measurable increases in cerebral NAD-related metabolites by 31P-MRS. Whether the clinical signal will replicate in a larger phase 3 design is unresolved.

Pharmacokinetics

Human pharmacokinetic data for NAD+ precursors come primarily from short-duration studies. Trammell and colleagues characterized oral NR pharmacokinetics in healthy adults, reporting dose-dependent increases in whole-blood NAD+ measured by LC-MS over several hours. Yoshino and colleagues published one of the earliest controlled NMN human pharmacokinetic studies in 2019, reporting that oral NMN was well-tolerated at the doses tested and that downstream NAD+ metabolites rose in plasma.

Direct NAD+ infusion pharmacokinetics in humans are sparsely characterized in peer-reviewed literature. Anecdotal IV protocols in clinical and research settings have not been accompanied by published, validated assays of intracellular NAD+ rise in target tissues. This is a meaningful gap: the assumption that infused NAD+ produces a sustained intracellular concentration change in, for example, muscle or brain has not been established with high-quality measurements. SC NAD+ pharmacokinetics are even less characterized.

The general principle from the available data is that exogenous administration — by any route — raises measurable NAD+ pool markers in blood, but tissue-level effects, dose-response curves, and the durability of those effects after dosing stops remain incompletely mapped.

Safety Signals

Across the published precursor literature, NR and NMN have been generally well-tolerated at doses tested in short-duration human studies. Reported adverse events have largely been mild and non-specific (gastrointestinal upset, fatigue), with no consistent signal of serious toxicity at studied doses. Long-term safety data — particularly for chronic, multi-year administration — are not yet available.

Several theoretical concerns recur in the review literature and warrant mention:

• Methylation balance. Nicotinamide is methylated by nicotinamide N-methyltransferase (NNMT) using S-adenosyl methionine (SAM) as the methyl donor. Sustained high-dose nicotinamide or NR administration could in principle shift methyl-group economy; this has been discussed but is not robustly demonstrated as a clinical concern.
• CD38 substrate availability. Raising NAD+ pools may also raise substrate availability for CD38 and PARP, with downstream effects on cyclic ADP-ribose signaling and calcium handling. The net effect of chronic precursor supplementation on these pathways in humans has not been mapped.
• Tumor metabolism. NAD+ is required for replicative cell metabolism. Several preclinical studies have examined whether NAD+ precursor administration could accelerate growth of established tumors via this dependency; results are mixed and tumor-type specific. This remains an open theoretical concern for high-dose chronic use.
• Direct IV NAD+. Adverse events anecdotally reported during IV NAD+ infusion (chest discomfort, flushing, gastrointestinal effects during infusion) are typically described as infusion-rate-dependent. Published controlled safety data for direct IV NAD+ are limited.

Open Research Questions

Several gaps in the NAD+ literature recur across review articles:

• Tissue-level pharmacodynamics. Whole-blood NAD+ measurements are convenient but may not reflect what happens in muscle, brain, or vascular endothelium. Better non-invasive tissue assays (e.g., 31P-MRS) need wider validation.
• Outcome studies vs biomarker studies. Most human NAD+ precursor trials measure biomarkers (NAD+ pool, methylation markers, performance proxies). Outcome studies on hard endpoints (cardiovascular events, cognitive decline, mortality) are largely absent.
• NMN vs NR vs direct NAD+. Head-to-head comparative human pharmacokinetic and pharmacodynamic data are sparse. Differences in cost, regulatory status, and oral vs parenteral delivery complicate clinical study design.
• Long-term safety. No multi-year controlled human safety data exist for any NAD+-raising strategy at the doses commonly used in research.
• Dose-response. Optimal dosing for any specific endpoint has not been established. The shape of dose-response curves — including possible inverted-U effects from over-shifting redox balance — is largely uncharacterized.
• The lifespan-to-healthspan extrapolation gap. The rodent data on NAD+ precursors are most consistent for healthspan endpoints; whether human aging trajectories respond similarly remains an empirical question that will require long-duration trials.

Researchers continuing to work with NAD+ and its precursors in preclinical contexts are encouraged to report tissue-specific pharmacodynamic measurements alongside circulating biomarker data, to disclose precursor identity and purity, and to design study durations long enough to capture both early biomarker shifts and any later off-target signals. The existing literature establishes that NAD+ pools can be raised exogenously; it has not yet established what raising them durably accomplishes in humans at the level of clinical outcome.