NAD+ Research Guide: Sirtuins, Mitochondria, and What the Data Actually Shows

NAD+ has become the most marketed molecule in the longevity-research market. Most of what you read about it conflates three different things — direct NAD+, NMN, and NR — and skips the one detail that matters: which of these has actually been tested in a human RCT. Here's the literature, sorted properly.

1. What NAD+ is — and why decline matters

NAD+ (nicotinamide adenine dinucleotide) is a coenzyme present in every living cell. It does two distinct jobs:

• Energy production. NAD+ is the central electron carrier of the mitochondrial electron transport chain. Glucose, fatty acid, and amino acid oxidation all converge on NAD+ → NADH electron transfer.
• Signalling substrate. NAD+ is the obligate substrate for the sirtuin (SIRT1–SIRT7) and PARP enzyme families. These enzymes consume NAD+ when they act — meaning sirtuin activity is rate-limited by NAD+ availability.

NAD+ levels decline measurably with age in human tissues. Massudi et al. (2012) documented ~50% reduction in human skin NAD+ between age 30 and age 60. Similar patterns appear in liver, brain, and skeletal muscle. The Sinclair/Imai aging hypothesis follows: declining NAD+ reduces sirtuin activity, which contributes to age-related physiological decline. Restoring NAD+ should partially reverse this.

2. The Imai/Guarente sirtuin connection

The reason NAD+ became interesting to longevity researchers — rather than just biochemists — was the 2000 paper from Shin-ichiro Imai and Leonard Guarente (then at MIT) showing that the yeast longevity gene Sir2 was a NAD+-dependent deacetylase. Without NAD+, Sir2 can't function. This was the first molecular link between an aging-regulator protein and a cellular metabolite that changes with age.

Mammals have seven sirtuins (SIRT1–SIRT7) doing related but distinct work — mitochondrial biogenesis, DNA repair, inflammation control, glucose homeostasis. All of them are NAD+-dependent. Reduce NAD+, you reduce sirtuin signalling across the board.

3. Direct NAD+ vs the precursors NMN and NR

Most "NAD+ research" you'll see cited is actually research on the precursors. The distinction matters because they're different molecules with different bioavailability and different research records.

Translation: when someone says "NAD+ extends lifespan in mice" they almost always mean "NMN given to mice extends physiological function." The molecule actually tested was usually a precursor.

4. The Mills 2016 mouse study — the foundational longevity paper

Mills KF et al., Long-Term Administration of Nicotinamide Mononucleotide Mitigates Age-Associated Physiological Decline in Mice, Cell Metabolism 2016. The Imai-lab paper that put NMN on the longevity-research map:

• Design: NMN administered in drinking water to mice for 12 months, starting at 5 months of age (rough equivalent to middle-aged human).
• Outcomes: Reduced age-associated decline in insulin sensitivity, energy expenditure, eye function, bone density, and immune function compared to controls.
• Mechanism: Restored NAD+ tissue levels, increased SIRT1 activity, improved mitochondrial function.
• Lifespan: Not the primary endpoint, but trends positive.

This is the cleanest long-term mouse dataset in the field. Replication by independent labs has been consistent in direction if not magnitude.

5. The Yoshino 2021 human RCT — the headline trial

Yoshino M et al., Nicotinamide mononucleotide increases muscle insulin sensitivity in prediabetic women, Science 2021. The first published RCT of NMN supplementation in humans:

• Design: Randomised, double-blind, placebo-controlled. 25 prediabetic, postmenopausal women.
• Intervention: 250 mg oral NMN daily for 10 weeks.
• Primary outcome: Muscle insulin sensitivity (hyperinsulinemic-euglycemic clamp).
• Result: Significant improvement in muscle insulin signalling and glucose disposal vs placebo.
• What it didn't show: Effects on body weight, body composition, or aging biomarkers — these were secondary endpoints and didn't reach significance.

Yoshino 2021 is the strongest human evidence for an NAD+ precursor to date. It also bounds the claim — the effect was real but specific to insulin signalling, not a generalised anti-aging outcome.

6. NR — the parallel-track precursor

Nicotinamide riboside (NR) has been in humans longer than NMN. Trammell SA et al. (Nature Communications 2016) established human bioavailability. Subsequent Phase 1/2 studies (Martens 2018, Conze 2019) confirmed that NR raises blood NAD+ levels in healthy adults. Effects on functional outcomes (blood pressure, arterial stiffness) have been smaller and less consistent than the bioavailability signal.

NR vs NMN: both work as precursors. NR is one enzymatic step further upstream. Direct head-to-head trials are sparse — practical choice usually comes down to cost and supplier quality.

7. The honest gaps

8. The summary

• NAD+ is the obligate substrate for sirtuin and PARP enzymes. Declines with age in human tissues.
• Direct NAD+ administration has limited human RCT data.
• NMN — Mills 2016 mouse aging paper; Yoshino 2021 human muscle-insulin-sensitivity RCT.
• NR — Trammell 2016 bioavailability; several follow-up Phase 1/2 trials.
• "NAD+ research" in the marketing is usually NMN or NR research. Distinction matters.
• Strongest single human result to date: Yoshino 2021. Bounded but real.