The clinical and mechanistic record on NAD+ — sirtuins, PARP1, CD38, and twenty human RCTs

NAD+ is the electron carrier at the center of cellular metabolism. In glycolysis, it accepts a hydride ion at the GAPDH step, generating NADH. In the TCA cycle, three additional NADH molecules are produced per acetyl-CoA turn. NADH then donates electrons to Complex I of the mitochondrial electron transport chain, driving the proton gradient that powers ATP synthase.

Beyond energy metabolism, NAD+ is a consumed substrate for three enzyme superfamilies that deplete rather than cycle the pool.

Sirtuins (SIRT1–7). These NAD+-dependent protein deacylases remove acyl marks from histones and non-histone proteins, directly linking NAD+ availability to epigenetic regulation, mitochondrial biogenesis (via SIRT1/PGC-1α), and insulin signaling. Each catalytic cycle consumes one NAD+ molecule. SIRT1 also participates in circadian NAD+ recycling: CLOCK:BMAL1 drives rhythmic NAMPT expression, creating a 24-hour NAD+ oscillation, and SIRT1 is recruited to the NAMPT promoter as a feedback amplifier.^[14]

PARPs. PARP1 and PARP2 detect DNA strand breaks and consume NAD+ to poly-ADP-ribosylate target proteins, recruiting repair machinery. Severe DNA damage can deplete up to 90% of cellular NAD+ in minutes.^[13] Older cells carry more baseline DNA damage, sustaining higher chronic PARP demand.

CD38. This ectoenzyme cleaves NAD+ and NMN to produce ADP-ribose and cyclic ADP-ribose for calcium signaling. CD38 expression rises 2–3-fold with aging (Camacho-Pereira et al. 2016, Cell Metabolism).^[1] Inflammatory cytokines secreted by senescent cells (SASP) activate CD38-expressing macrophages in visceral fat and liver, creating a feedforward loop between cellular senescence and NAD+ depletion.^[2]

The NAD+/PARP1/SIRT1 axis is the unifying model: PARP1 and SIRT1 compete for the same NAD+ pool. When PARP is hyperactivated by DNA damage, it outcompetes SIRT1, reducing sirtuin-mediated mitochondrial maintenance and epigenetic regulation. Raising the NAD+ pool re-enables sirtuin activity (Mendelsohn and Larrick 2017).^[13]

NAD+ is the primary electron carrier in glycolysis and the TCA cycle. Without adequate NAD+, the GAPDH step in glycolysis stalls; the NAD+/NADH redox ratio shifts, impairing glucose oxidation and fatty acid metabolism; and mitochondrial ATP synthesis slows. The consequence is not merely reduced energy output — it is the simultaneous suppression of sirtuin-dependent gene regulation and PARP-dependent DNA repair, because all three functions draw from the same depleted pool.

A circadian dimension compounds this: the CLOCK:BMAL1 transcription complex drives rhythmic NAMPT expression, so NAD+ availability follows a 24-hour cycle synchronized with metabolic demand.^[14] Circadian disruption — shift work, irregular sleep — attenuates the amplitude of this cycle, reducing peak NAD+ availability.

Sirtuins (SIRT1–7) are NAD+-dependent deacylases that regulate mitochondrial biogenesis, inflammation response, and DNA repair coordination. NAD+ is consumed in each catalytic cycle; therefore, sirtuin activity is a direct readout of cellular NAD+ availability.

SIRT1 regulates PGC-1α (mitochondrial biogenesis), insulin receptor substrate-1 (IRS-1) phosphorylation, and the NAMPT promoter. SIRT3 governs mitochondrial protein deacetylation; mice lacking SIRT3 age-dependently lose mitochondrial function in a phenotype rescued by restoring NAD+.^[1] SIRT6 deacetylates histones at DNA break sites, linking it directly to the PARP1 competition for NAD+.

In the NAD+/PARP1/SIRT1 competition model, dietary supplementation with NAD+ precursors re-activates sirtuin signaling in mouse models of aging, improves mitochondrial function, and extends lifespan in some murine models (Mendelsohn and Larrick 2017).^[13] Human trials confirm NAD+ elevation; sirtuin activity as a downstream endpoint has not been directly measured in published clinical trials.

PARP1 is the primary sensor of DNA single- and double-strand breaks. When it detects a break, it consumes NAD+ to build poly-ADP-ribose chains on target proteins, recruiting the repair machinery. Under severe genotoxic stress, PARP1 activation can consume 90% of cellular NAD+ within minutes.^[13]

In human skin biopsies, NAD+ and NADH levels declined with age while PARP activity rose; the two correlated inversely (Massudi et al. 2012, PLoS One) — the first direct human tissue evidence for PARP-driven NAD+ depletion in aging.^[22]

In patients with inherited DNA-repair defects, where PARP is constitutively hyperactivated, NR supplementation has produced clinically meaningful results. A 52-week RCT in Werner syndrome patients (n=9) using 500–1000 mg/day NR produced a ~140% plasma NAD+ increase with improved arterial stiffness, HDL counts, and kidney function. In ataxia-telangiectasia trials, NR supplementation slowed neuromotor decline over 2-year follow-up (Bohr 2025, Aging Cell).^[21]

Both NMN (nicotinamide mononucleotide) and NR (nicotinamide riboside) raise blood NAD+ in human RCTs, but via mechanistically distinct routes.

NMN is absorbed in the intestine via the Slc12a8 sodium-dependent transporter, entering cells directly and bypassing one phosphorylation step to NAD+. Slc12a8 expression increases in aged mouse ileum, potentially compensating for declining NMN bioavailability with age (Grozio et al. 2019, Nature Metabolism).^[15]

NR enters cells through nucleoside transporters and is phosphorylated to NMN intracellularly before conversion to NAD+. At 100–1000 mg/day over 8 weeks, NR raised whole-blood NAD+ by 22%, 51%, and 142% respectively in a dose-dependent manner in healthy overweight adults, with no flushing and no serious adverse events at any dose.^[8]

A 2026 Nature Metabolism four-arm RCT (Christen et al.) directly compared NR, NMN, and nicotinamide (NAM) vs placebo over 14 days: NR and NMN each approximately doubled blood NAD+ vs placebo (differences ~49 µM and ~43 µM respectively; p<0.001), while NAM had no significant effect. The study also identified gut microbiome conversion of NR and NMN to nicotinic acid as a contributor to the observed NAD+ boost — a source of inter-individual response variability.^[16]

NR is a form of vitamin B3 that serves as a precursor to NMN and then NAD+. It enters cells through nucleoside transporters and requires one additional phosphorylation step relative to NMN before reaching NAD+. Critically, NR is a legal dietary supplement in the US (unlike NMN, which FDA classified as a drug ingredient in January 2023).

The safety profile of NR is the most extensively documented of any NAD+ precursor in humans. The NR-SAFE trial (Berven et al. 2023, Nature Communications) administered 3000 mg/day (the highest published dose for any NAD+ precursor) to Parkinson disease patients for 30 days — no moderate or severe adverse events were reported, and the blood and urine NAD+ metabolome was substantially elevated.^[18] Importantly, NR does not cause the vasodilatory flushing associated with niacin.

In older adults with mild cognitive impairment, 1 g/day NR for 10 weeks raised blood NAD+ 2.6-fold but did not significantly improve cognitive performance on the Montreal Cognitive Assessment or related metrics (Orr et al. 2024, GeroScience).^[20] This trial is the clearest current evidence that NAD+ elevation alone is insufficient for cognitive benefit in MCI.

Oral NAD+ itself is largely hydrolyzed by intestinal NADases (CD38, CD73) before reaching systemic circulation. The intact molecule cannot efficiently cross the gut wall; the primary absorption products are nicotinamide (NAM) and ADP-ribose, which are then resynthesized to NAD+ intracellularly.

This is why NMN and NR — which are absorbed as intact nucleosides via dedicated transporters before conversion to NAD+ — are the effective oral delivery vehicles for NAD+ repletion. New 2026 evidence (Christen et al.) adds another layer: gut microbiome bacteria convert a portion of orally administered NR and NMN to nicotinic acid, which enters via the Preiss-Handler pathway, contributing to but also introducing variability in the observed NAD+ boost.^[16] Individual microbiome composition is now recognized as a significant variable in oral precursor response.

No. Niacin (nicotinic acid, vitamin B3) is one biosynthetic precursor to NAD+ via the Preiss-Handler pathway, but NAD+ itself is structurally and pharmacologically distinct. NAD+ is a dinucleotide coenzyme with a molecular formula of C₂₁H₂₇N₇O₁₄P₂ and a molecular weight of 663.4 Da — more than four times the molecular weight of niacin (123.1 Da). The therapeutic properties associated with NAD+ research derive from sirtuin activation, PARP modulation, and mitochondrial electron transport function — mechanisms that nicotinic acid does not directly recapitulate. NMN and NR are more direct precursors, with higher bioavailability and no flush reaction.

NAD+ vs NADH: Understanding the Redox Pair

NAD+ and NADH are the oxidized and reduced forms of the same molecule. NAD+ accepts electrons (is reduced to NADH); NADH donates electrons to Complex I of the mitochondrial electron transport chain (is re-oxidized to NAD+). The therapeutic literature focuses on raising the total NAD+ pool rather than directly shifting the ratio: sirtuin activity depends on NAD+ availability, not on the redox ratio per se.

Does NMN Raise NAD+ in Human Tissue?

Yes, unambiguously in blood. Multiple RCTs confirm that oral NMN raises whole-blood NAD+ in humans: Okabe et al. (2022, Frontiers in Nutrition) showed 250 mg/day for 12 weeks produced significant and persistent blood NAD+ elevation that reversed to baseline within 4 weeks of stopping.^[4] Yi et al. (2022, GeroScience) identified 600 mg/day as the optimal dose in a multicenter double-blind RCT, with blood NAD+ elevated significantly at days 30 and 60.^[6]

Skeletal muscle is the only tissue beyond blood with published human data: the Yoshino 2021 Science RCT showed improved muscle insulin sensitivity,^[5] and Igarashi et al. (2022) showed improved gait speed and grip strength,^[7] consistent with muscle-level benefit. Direct tissue measurement in liver, brain, or adipose has not yet been published in human trials.