The Three-Way NAD+ Competition: Sirtuins, PARPs, and CD38
NAD+ is not merely a cofactor that gets recycled — it is consumed and destroyed in three distinct reactions. Understanding this consumption is essential to understanding why NAD+ declines with aging and why simply supplementing precursors may not fully offset the problem when consuming enzymes are upregulated.
Sirtuins: Gene Expression and Mitochondrial Biogenesis
Each sirtuin deacylation reaction consumes one NAD+ molecule: the acetyl group from the protein substrate is transferred to the ADP-ribose portion of NAD+, producing nicotinamide (Nam) and O-acetyl-ADP-ribose. The Nam product is a competitive inhibitor of sirtuin activity (feedback inhibition), which is why NAMPT (nicotinamide phosphoribosyltransferase), the enzyme that recycles Nam back into NAD+ via the salvage pathway, is so important — it both re-generates NAD+ and removes the inhibitory Nam product. SIRT1 activation promotes PGC-1α deacetylation → mitochondrial biogenesis. SIRT3 activates SOD2 (mitochondrial superoxide dismutase) by deacetylation at K122, increasing mitochondrial antioxidant capacity. SIRT6 maintains telomere structure by deacetylating H3K9 and H3K56 at telomeric chromatin.
PARPs: DNA Damage Sensing and Repair
PARP1 (poly-ADP-ribose polymerase 1) is the most active NAD+ consumer under conditions of DNA damage. Upon detecting a single-strand break, PARP1 immediately synthesizes long chains of poly-ADP-ribose (PAR) from NAD+ molecules — consuming up to 200 NAD+ molecules per activation event. This PAR modification of histones near the break site recruits DNA repair machinery. The problem in aging: as chronic oxidative stress increases single-strand breaks accumulate continuously, keeping PARP1 constitutively active and draining the NAD+ pool. This PARP1 hyperactivation is a major driver of the NAD+ decline seen in aging tissue — essentially the DNA repair system bankrupting the NAD+ budget and leaving insufficient substrate for sirtuins. Verdin 2015 (Science) described this as the "competition hypothesis": PARP and SIRT share the same NAD+ pool, and when DNA damage overwhelms PARP, sirtuins starve.
CD38: The Dominant Age-Related NAD+ Consumer
CD38 is the newest addition to the NAD+ consumption picture and the most quantitatively important in the aging context. CD38 is a multifunctional enzyme: it both hydrolyzes NAD+ to ADP-ribose + Nam, and synthesizes cyclic ADP-ribose (cADPR) from NAD+ for calcium signaling. While CD38 is expressed constitutively in immune cells, its expression increases dramatically in aging tissues — specifically because senescent cells secrete the SASP (senescence-associated secretory phenotype) cytokines (IL-6, IL-1β, TNF-α) that transcriptionally upregulate CD38 in nearby macrophages and endothelial cells. This creates a vicious cycle: aging → senescent cell accumulation → SASP → CD38 upregulation → NAD+ depletion → impaired SIRT3 → accelerated mitochondrial dysfunction → more senescent cells.
The Camacho-Pereira 2016 (Cell Metab) paper established this mechanism definitively: old CD38 knockout mice maintained NAD+ levels identical to young wild-type mice. CD38 expressed in 22-month-old mice consumed more NAD+ than PARP + sirtuins combined. Natural CD38 inhibitors identified in that paper: apigenin (a flavonoid in parsley, chamomile) at ~100μM in vitro, and quercetin at similar concentrations. Neither achieves these concentrations in vivo at standard dietary intakes.
| Precursor | Pathway | Key Human RCT | Blood NAD+ Change | Notable Effect | Key Caveat |
|---|---|---|---|---|---|
| NMN (nicotinamide mononucleotide) | Salvage (NMN → NR → NMN in tissue OR direct NMN transporter Slc12a8) | Yoshino 2021 (Science), n=25, 250mg/day × 10 weeks, women 65+ | +40% blood NAD+ | ↑ insulin sensitivity in skeletal muscle (hyperinsulinemic clamp); ↑ SIRT1/SIRT3 mRNA | No functional performance improvement in this cohort; Slc12a8 transporter expression is low in humans vs mice — most NMN likely converts to NR before uptake |
| NR (nicotinamide riboside) | Salvage (NR kinase 1/2 → NMN → NAD+) | Dollerup 2018 (Nat Commun), n=40, 1000mg/day × 12 weeks, obese men | +60% blood NAD+ | ↑ skeletal muscle NAD+; no metabolic benefit in this healthy obese cohort without exercise training | Metabolic benefits may require co-intervention (exercise + NR); Pirinen 2020 (Cell Metab) showed NR boosted muscle NAD+ but did not improve mitochondrial function in mitochondrial myopathy patients |
| Nicotinic acid (NA / niacin) | Preiss-Handler pathway (NA → NaMN → NaAD → NAD+); completely different from salvage | Pirinen 2020 (Cell Metab), 750–1000mg/day, mitochondrial myopathy patients | +2.3-fold blood NAD+; highest tissue NAD+ increase of any precursor | Significant improvement in muscle mitochondrial function and walking distance; uniquely replenishes both cytoplasmic and mitochondrial NAD+ pools via a different enzymatic pathway | Flushing reaction (prostaglandin D2-mediated skin vasodilation) at doses ≥100mg in most people — limits compliance; slow-release forms reduce flushing but may increase hepatotoxicity risk |
| Nicotinamide (NAM) | Salvage (NAMPT converts Nam → NMN) | Various small studies; no major longevity-focused RCTs | Modest NAD+ increase | Cheap, FDA GRAS, widely available | Feedback inhibits sirtuins at high concentrations; NNMT (nicotinamide N-methyltransferase) shunts excess Nam to methylnicotinamide, consuming SAM (methyl donor) — potential methylation depletion at high doses |
NAD+ Optimization: What the Evidence Currently Supports
- NMN vs NR — the mechanistic distinction is smaller than marketed: A key 2022 paper (Liu 2022, Nat Metab) demonstrated that orally administered NMN is converted to NR in the gut before cellular uptake in most tissues, then re-phosphorylated to NMN intracellularly by NRK enzymes. The Slc12a8 direct NMN transporter (Grozio 2019) is expressed in intestinal cells and may allow some direct NMN uptake, but its role in humans appears quantitatively minor based on isotope tracer studies. Both NMN and NR ultimately raise intracellular NAD+ via the same NRK → NMNAT enzymatic steps. The differences that matter: NR may have slightly better bioavailability per gram in some studies; NMN has better stability in acidic environments (gastric acid). At equivalent NAD+-raising doses, clinical outcomes appear similar in available head-to-head comparisons.
- Nicotinic acid (niacin) is the most potent NAD+ precursor per unit dose and uniquely accesses the Preiss-Handler pathway: While NMN and NR both feed into the salvage pathway (and are rate-limited by NAMPT activity), nicotinic acid enters via NAPRT (nicotinic acid phosphoribosyltransferase) into a completely separate enzymatic route that produces NaMN, then NaAD, then NAD+ via NADS. This pathway can raise tissue NAD+ more dramatically than the salvage pathway — Pirinen 2020 showed 2.3-fold blood NAD+ increase at 750mg/day NA vs ~1.4–1.6-fold for NR at 1000mg/day. The barrier is the flushing reaction: 50–100mg niacin triggers skin flushing (red, warm, itching skin) via COX-1 → PGD2 → DP1 receptor in skin mast cells. Aspirin 325mg taken 30 minutes before niacin blocks this reaction. Inositol hexanicotinate ("flush-free niacin") is largely ineffective at raising NAD+ because it has very poor hydrolysis to free NA in vivo.
- NAMPT is the rate-limiting bottleneck in the salvage pathway: NAMPT catalyzes the first and slowest step in NAD+ salvage: Nam + PRPP → NMN. This is why flooding the salvage pathway with NMN or NR does not indefinitely raise NAD+ — once NAMPT is saturated, additional precursor is excreted. NAMPT activity is induced by exercise (acute resistance and endurance exercise both upregulate NAMPT mRNA in muscle within hours), by caloric restriction, and by the circadian clock (NAMPT is a CLOCK:BMAL1 target gene — it oscillates ~2-fold over 24 hours). Conversely, NAMPT is inhibited by inflammatory cytokines (TNF-α, IL-6) — another SASP connection linking senescent cell accumulation to NAD+ decline. This means that individuals with high systemic inflammation may have both increased CD38 NAD+ consumption AND decreased NAMPT-mediated NAD+ production simultaneously.
- Timing: NMN/NR timing with meals and circadian biology: Since NAMPT is a CLOCK:BMAL1 target gene with a circadian peak in early morning (roughly 6–10am in most chronotypes), several researchers have proposed that NAD+ precursor supplementation may be most effective in the morning when NAMPT activity is highest. Limited human data on this exists, but mouse studies show morning NMN administration (relative to active phase) produces larger NAD+ peaks than evening dosing. The practical recommendation from most NAD+ researchers: take NMN or NR in the morning with or without food. The acid-stability of NMN (better than NR at low pH) suggests NMN may be better suited to taking with coffee or other acidic morning drinks, while NR is better protected in enteric-coated capsules.
- Synergy with resveratrol and pterostilbene — the SIRT1 activator debate: Resveratrol was initially reported (Howitz 2003, Nature) to be a direct SIRT1 activator — but subsequent work showed it only activates SIRT1 when a fluorophore is attached to the substrate (an artifact of the assay). The current understanding: resveratrol and pterostilbene are indirect SIRT1 activators via AMPK activation and mTOR inhibition — they increase cellular NAD+ availability by reducing NAD+ consumption (via AMPK inhibition of anabolic pathways that consume NAD+), making existing NAD+ more available for sirtuins. Combined NAD+ precursor + pterostilbene strategies are being studied but have no strong human RCT data as of 2025.
When comparing NAD+ precursors: look for third-party tested products (NSF, USP, or Informed Sport certified) — the NMN and NR market has significant purity variation. Liposomal delivery forms (NMN or NR in phospholipid vesicles) claim improved bioavailability but lack head-to-head RCT data vs standard oral forms. Sublingual NMN dissolves under the tongue for potential direct mucosal uptake bypassing the gut-to-NR conversion step — plausible but unproven in humans. Dose: 250–500mg/day NMN or 300–1000mg/day NR are the human-trial-tested ranges. Higher doses do not proportionally raise NAD+ further due to NAMPT rate-limiting.
20+ compounds — including the NAD+ precursors above — ranked by human evidence, plus the complete VO2 max, sleep, and hormone optimization protocols for performing after 40.