Why NAD+ Matters: The Central Role in Cellular Energy and Repair
Nicotinamide adenine dinucleotide (NAD+) is one of the most fundamental molecules in biochemistry — a coenzyme involved in over 500 enzymatic reactions. Its primary roles span two domains that are directly relevant to aging:
Energy metabolism: NAD+ is the electron acceptor in glycolysis and the TCA cycle. It accepts hydride ions (H⁻) to become NADH, which then donates electrons to Complex I of the mitochondrial electron transport chain to drive ATP synthesis. Without NAD+, neither glycolysis nor oxidative phosphorylation can proceed.
Signaling and repair: NAD+ is consumed (not just used as a cofactor) by three enzyme families: sirtuins (SIRT1–7, deacylases that regulate gene expression and mitochondrial function), PARPs (poly-ADP-ribose polymerases, which sense DNA damage and initiate repair), and CD38/CD157 (NADases involved in calcium signaling and immune function). These enzymes cleave NAD+ irreversibly — every activation event consumes one NAD+ molecule.
The collision between these two demands — energy metabolism requiring NAD+ cycling and repair/signaling requiring NAD+ consumption — creates a tension that becomes acute as NAD+ levels fall with aging.
The Age-Related Decline: 50% Loss in 40 Years
Multiple independent measurements across human tissues confirm that NAD+ declines substantially with age. Massudi et al. (2012) measured NAD+ in human skin biopsies across age groups and found a near-linear decline of approximately 1.0–1.3% per year from age 20 onward. Yoshino et al. (2021) measured NAD+ in skeletal muscle of older women and found levels approximately 50% lower than young women.
The consequences of this decline are well-characterized in animal models and increasingly supported in humans:
- Reduced SIRT1 and SIRT3 activity → impaired mitochondrial biogenesis (PGC-1α deacetylation) and reduced fatty acid oxidation
- Reduced PARP1 activity → slower DNA double-strand break repair → accumulation of somatic mutations
- Reduced SIRT6 activity → increased genomic instability and NF-κB-mediated inflammation
- Reduced NAMPT (the rate-limiting enzyme in the NAD salvage pathway) expression → NAD+ biosynthesis can't keep pace with consumption
CD38: The Overlooked NAD Drain
The dominant narrative in longevity circles attributes NAD+ decline to reduced biosynthesis — the standard "NAMPT declines with age" story. But Camacho-Pereira et al. (2016, Cell Metabolism) identified a different primary culprit: CD38, a plasma membrane glycohydrolase whose expression increases dramatically with age and inflammation.
CD38 is the dominant NADase in mammalian tissue — it is estimated to consume up to 100-fold more NAD+ per unit time than sirtuins and PARPs combined. In young tissue, CD38 expression is relatively low. With aging, CD38 expression rises — driven primarily by the accumulation of pro-inflammatory senescent cells (the "senescence-associated secretory phenotype" or SASP releases cytokines that upregulate CD38 in neighboring cells).
The Camacho-Pereira study: CD38 knockout mice maintained NAD+ levels into old age comparable to young animals, and showed dramatically better metabolic health, mitochondrial function, and physical fitness at 22 months (human equivalent ~70 years) than wild-type aged mice. Crucially, CD38 KO mice did not need NMN or NR supplementation — maintaining CD38 at baseline levels was sufficient to preserve NAD+.
The practical implication: senolytic interventions and anti-inflammatory strategies that reduce the senescent cell burden may address the CD38 problem upstream, potentially more efficiently than supplying more NAD+ precursor downstream. This is not a reason to dismiss NMN/NR — it's a reason to combine them with strategies that reduce CD38 induction (apigenin, quercetin, resveratrol, and senolytics like fisetin and dasatinib + quercetin all inhibit CD38 to varying degrees).
NMN vs NR: Biosynthetic Pathways and Transport
Both nicotinamide mononucleotide (NMN) and nicotinamide riboside (NR) are NAD+ precursors in the Preiss-Handler pathway. Their relationship:
NR → (NRK1/2 kinases, intracellular) → NMN → (NMNAT1/2/3) → NAD+
NMN bypasses the NRK step and enters the pathway one step further downstream. The critical question — debated most publicly between David Sinclair (pro-NMN) and Charles Brenner (pro-NR and NAD+ precursor agnostic) — is whether NMN can enter cells directly as NMN, or must first be dephosphorylated to NR extracellularly.
Grozio et al. (2019, Nature Metabolism) identified a specific NMN transporter, Slc12a8, that imports NMN directly into intestinal cells and potentially other tissues — supporting the idea that NMN has a direct uptake route. Brenner's group disputed the interpretation, arguing the evidence was insufficient to conclude systemic NMN transport in humans.
The practical resolution: both NMN and NR reliably raise blood NAD+ levels in humans — this is not disputed. The question of which raises tissue NAD+ more efficiently in specific organs (especially muscle and brain) remains open. Both compounds convert to NMN intracellularly before becoming NAD+; the kinetics differ but both work.
Yoshino 2021 (NEJM): The Pivotal Human Trial
The New England Journal of Medicine publication from Yoshino et al. (2021) represents the highest-quality human evidence available for NMN. The design: 25 postmenopausal, prediabetic women randomized to NMN 250mg/day vs. placebo for 10 weeks in a double-blind crossover design.
Primary endpoint: skeletal muscle insulin sensitivity (hyperinsulinemic-euglycemic clamp — the gold standard). Secondary endpoints: blood NAD+ metabolites, muscle biopsy gene expression, body composition.
Results:
- Blood NAD+ metabolomics: NMN significantly increased blood NMN, NR, and NAD+ levels vs. placebo
- Muscle insulin sensitivity: +25% improvement (Rd, glucose disposal rate) in NMN group vs. placebo — statistically significant
- Muscle gene expression: Upregulation of genes involved in muscle remodeling, oxidative phosphorylation, and extracellular matrix organization
- Body composition, blood pressure, lipids: No significant changes in 10 weeks
- Safety: No adverse effects at 250mg/day
The effect on insulin sensitivity is clinically meaningful — a 25% improvement in glucose disposal in a prediabetic population is comparable to what's achieved with moderate aerobic exercise programs. The absence of body weight changes in 10 weeks is expected; metabolic improvements precede body composition changes.
| Study | Population | Intervention | Key Result |
|---|---|---|---|
| Yoshino et al. 2021 (NEJM) | N=25, postmenopausal women, prediabetic, RCT | NMN 250mg/day × 10 weeks | +25% muscle insulin sensitivity; blood NAD+ metabolites increased; upregulated mitochondrial gene expression |
| Camacho-Pereira et al. 2016 (Cell Metab) | CD38 KO mice vs. wild-type aged | Genetic KO of CD38 | CD38 KO maintained NAD+ into old age; dramatically better metabolic health and fitness at 22 months vs WT |
| Gomes et al. 2013 (Cell) | Aged mice + NMN gavage | NMN 500mg/kg/day × 1 week | Restored pseudohypoxic state in muscle; reversed age-related mitochondrial deterioration; SIRT1 activation confirmed |
| Trammell et al. 2016 (Nat Commun) | N=12 healthy adults, RCT | NR 1,000mg single dose | Blood NAD+ increased 2.7-fold at 6 hours; SIRT1 activity in PBMCs increased; PBMC NAD+ metabolomics confirmed |
| Dollerup et al. 2018 (Nat Commun) | N=40, obese men, RCT | NR 2,000mg/day × 12 weeks | Blood NAD+ increased significantly; no improvement in insulin sensitivity, body composition, or lipids vs placebo — questions systemic efficacy |
| Martens et al. 2018 (Nat Commun) | N=24, healthy older adults, RCT | NR 500mg/day × 6 weeks | Blood NAD+ up ~60%; SIRT1 activity trend; aortic stiffness reduced −6% (trend, p=0.07); no CV events |
The Sinclair vs. Brenner Debate: What It's Actually About
David Sinclair (Harvard, author of Lifespan) advocates NMN as the superior precursor and publicly takes 1,000mg/day himself. Charles Brenner (University of Iowa, discoverer of NR's NAD+ precursor role) argues that NMN and NR are effectively equivalent at the cellular level and that the data base is insufficient to prefer either.
The disagreement has several layers:
- The transport question: Sinclair favors Grozio's Slc12a8 data showing direct NMN transport; Brenner finds the evidence for systemic NMN transport in humans unconvincing
- Commercial conflict: Both researchers have affiliations with companies selling their preferred precursor — this should be factored into weighting their public statements
- Dose: Sinclair uses 1,000mg/day NMN; most human trials use 250–500mg NR or NMN — whether higher doses provide proportionally more benefit is unknown
- End-organ delivery: The real question is not blood NAD+ but tissue NAD+ — particularly muscle, liver, and brain. These compartments have not been directly measured in humans with biopsy studies at comparable doses
The data-driven answer: Both work to raise blood NAD+. NMN has the Yoshino 2021 human efficacy trial on insulin sensitivity. NR has the Trammell 2016 pharmacokinetics data and multiple smaller human trials. For most purposes, the form you choose is less important than dose consistency and the supporting stack (CD38 inhibitors, NAD+ biosynthesis support via niacin/nicotinamide).
Evidence-Based NAD+ Protocol
- NMN or NR dose: 250–500mg/day is the human-trial validated range. Yoshino 2021 used 250mg NMN. Higher doses (500–1,000mg) extrapolate from animal data — not unreasonable but not directly human-validated for efficacy endpoints beyond blood NAD+.
- Timing: Morning with food. Some data suggests NAD+ metabolism follows circadian patterns; morning dosing aligns with natural NAD+ peak. Avoid taking late — SIRT1 activation affects wakefulness circuits.
- CD38 inhibition (synergistic): Apigenin (50mg/day, from parsley or supplement), quercetin (500mg), or luteolin reduce CD38 expression and slow NAD+ degradation. Address the drain alongside increasing supply.
- NAMPT cofactors: NAMPT (the rate-limiting biosynthesis enzyme) requires phosphoribosyl pyrophosphate — ensure adequate niacin/B3 intake, as niacin is the upstream precursor in the de novo pathway.
- Resveratrol (debated): Sinclair combines NMN with resveratrol (SIRT1 activator). The evidence that resveratrol activates SIRT1 directly remains contested (Pfizer/Sirtris controversy). As a CD38 inhibitor and mild AMPK activator, it may still add value in this stack.
- Exercise amplifies NAD+ effects: Aerobic exercise independently increases NAMPT expression and NAD+ biosynthesis. The Yoshino 2021 trial showed NMN specifically improved muscle NMN transport gene expression — combining NMN with resistance training may be synergistic.
Recommended Products (Amazon)
Look for third-party tested NMN with purity certification. Sublingual or enteric-coated formulations may improve bioavailability vs. standard capsules. 250–500mg daily is the human-trial validated range.
NR (e.g., Tru Niagen / ChromaDex) has the most clinical trial data of any NAD+ precursor. 300–500mg/day is the standard dose range. Equivalent to NMN for most users at comparable doses.