Why NAD+ Falls With Age: CD38, PARP and What Human Tissue Shows

LynxLab · 24 October 2026 · 8 min read

What the Camacho-Pereira 2016 and Massudi 2012 studies found about falling NAD+ with age, the enzymes that consume it, and what is measured in human tissue.

Photo: Bioscience Image Library by Fayette Reynolds, Unsplash

Almost every article about NAD+ opens with the same line: levels fall as we age. The claim is true, but the interesting part is why, and how anyone measured it in a living person. The answer involves three families of enzymes, a handful of small human studies and a lot of mouse work.

NAD+ is consumed, not just used

Nicotinamide adenine dinucleotide sits at the centre of cell metabolism. In its redox role it shuttles electrons between NAD+ and NADH and is not used up; the same molecule goes round again.

The problem is a second set of reactions where NAD+ is the substrate and is destroyed. Grant and colleagues (Frontiers in Aging Neuroscience, 2019) listed them: the PARP family of DNA repair enzymes, the seven sirtuins, the CD38 and CD157 ecto-enzymes, and SARM1 in neurons. Each of these breaks the molecule apart and releases nicotinamide.

Cells rebuild NAD+ from that nicotinamide through the salvage pathway. The age-related fall is best understood as the point where consumption outruns resynthesis, and researchers have spent a decade arguing over which consumer is to blame.

Massudi 2012: the first look at ageing human skin

For years the decline was only shown in rodents. Massudi, Grant and colleagues at the University of New South Wales changed that in a 2012 PLoS One paper that used human pelvic skin removed during unrelated surgery.

The sample was small: 49 people, from newborns to a 77-year-old. They measured DNA damage, lipid peroxidation, PARP activity, SIRT1 activity and total NAD+ in the tissue.

The pattern was clearer in males than females. PARP activity was inversely related to NAD+ in men (r = -0.639), which fitted the authors' hypothesis that oxidative DNA damage keeps PARP busy and PARP eats the cell's NAD+. In women the correlation with age was still present (r = -0.537) but the PARP link was weaker, and the authors suggested women might recycle nicotinamide more efficiently.

The paper's own conclusion is cautious. It reports a correlation between oxidative stress, PARP and declining NAD+ in human tissue and says the therapeutic value of restoring NAD+ "is yet to be established". That caution has not always survived into the marketing built on top of it, which is one reason to buy nad+ for research from a supplier that publishes a lab report rather than a claim.

Camacho-Pereira 2016: CD38 as the main culprit in mice

Four years later the Chini laboratory at the Mayo Clinic made a different enzyme the headline. Camacho-Pereira and colleagues published "CD38 dictates age-related NAD decline and mitochondrial dysfunction through an SIRT3-dependent mechanism" in Cell Metabolism in June 2016.

They showed that CD38 expression and NADase activity rise in multiple mouse tissues with age, while SIRT1 and PARP1 levels did not follow the same curve. Mice lacking CD38 kept their NAD+ as they aged and were protected from the mitochondrial decline and glucose intolerance seen in normal old mice.

In a commentary on the paper in the same issue, Schultz and Sinclair put the size of the effect plainly: at 32 months a wild-type mouse had about half the NAD+ of a young mouse, and the CD38 knockout showed no decrease. They titled the piece "Why NAD+ declines during aging: it's destroyed".

The same study found that CD38 is also the main enzyme degrading nicotinamide mononucleotide (NMN) in vivo. That matters for anyone reading NMN trials, because it means the enzyme that lowers NAD+ with age also lowers the precursor people take to raise it.

Where the extra CD38 comes from

If CD38 goes up with age, something must switch it on. Two papers in Nature Metabolism in November 2020 pointed at the immune system. Covarrubias and colleagues at the Buck Institute reported that senescent cells, through the inflammatory proteins they secrete, activate CD38 on macrophages that then accumulate in ageing tissue. Chini's group reported in parallel that the CD38 ecto-enzyme in immune cells is induced during ageing and regulates both NAD+ and NMN levels.

The picture that emerges is circular. Damaged, senescent cells inflame their surroundings; inflamed tissue fills with CD38-rich immune cells; those cells destroy NAD+ and NMN in the space around them. Covarrubias, Perrone, Grozio and Verdin reviewed this in Nature Reviews Molecular Cell Biology in 2021 and argued that inflammation, rather than a failure of synthesis, is the most plausible driver of the decline.

Researchers who want to test any part of this loop in cell culture need the molecule itself, which is why demand for research-grade NAD+ has grown alongside the papers.

Sirtuins: the reason the decline is thought to matter

Sirtuins are the third family. They are NAD+-dependent deacetylases, and SIRT1 and SIRT3 in particular have been tied to mitochondrial maintenance and the benefits of calorie restriction in animals.

The Camacho-Pereira data put SIRT3 in the chain of events: CD38 up, NAD+ down, SIRT3 activity down, mitochondrial function down. Massudi's human skin data showed SIRT1 activity falling with age in men but not women, again pointing to a sex difference nobody has fully explained.

The careful reading is that sirtuin loss is the proposed consequence of NAD+ decline, not a separately proven human finding. Most of the sirtuin evidence remains in yeast, worms and mice.

What is actually measured in people

Human tissue is hard to get. The 2012 skin study used surgical offcuts. Most later human work has measured blood instead, which raises the question of what blood NAD+ represents.

Clement, Wong, Poljak, Sachdev and Braidy (Rejuvenation Research, 2019) used liquid chromatography with tandem mass spectrometry on plasma from 30 healthy adults aged 20 to 87. Plasma NAD+, NADP+ and NAAD fell with age, the NAD+ to NADH ratio fell, and middle-aged subjects sat between the young and the old. The authors flagged their own limits: 30 people, a largely Caucasian US sample and a cross-sectional design.

Zhu and colleagues at the University of Minnesota took a different route in PNAS in 2015, using a high-field magnetic resonance scanner to measure NAD+ and NADH inside the intact human brain. They reported an age-dependent fall in NAD+ and total NAD and a rise in NADH, the first non-invasive measurement of its kind.

Two cautions apply to all of these numbers. First, NAD+ degrades quickly once a sample is taken, so handling and extraction methods change the result. Second, plasma NAD+ is measured in nanomoles per litre while intracellular NAD+ is in the micromolar range, so a plasma value is a signal of turnover rather than a readout of what cells contain.

What this means for a buyer in the UK

None of the studies above tested a product. They measured a molecule in tissue, plasma or brain and asked what changes with age. Anyone looking to buy NAD+ for a laboratory project should read them as a map of open questions: which consumer dominates in which tissue, why the sex difference appears, and whether raising NAD+ changes any downstream outcome in people.

LynxLab stocks NAD+ for research use in 500mg vials at £49.99, tested to 99% purity with the lab report on the product page. The batch lab reports are published for every product, and the companion piece on NAD+ precursors versus NAD+ itself covers the NMN and NR trials and the FDA's rulings.

The regulatory position is simple. NAD+ is not a licensed medicine in the UK, and NAD+ UK listings from any supplier are for laboratory use, not for human consumption. The human data on raising it are small, short and mostly about precursors, which the next post covers.

Frequently asked questions

At what age does NAD+ start to decline?

The human data do not give a single switch-on age. The 2012 skin study saw lower tissue NAD+ in adults aged 30 to 50 compared with newborns, and the 2019 plasma study found middle-aged values (41 to 60) already sat below those of young adults. Both describe a gradual slope rather than a cliff.

Does CD38 cause NAD+ decline in humans?

The CD38 evidence is strongest in mice. The 2016 Mayo Clinic study showed that CD38 knockout mice kept youthful NAD+ levels, and the 2020 Buck Institute study showed CD38-rich macrophages accumulate in ageing mouse tissue. In humans, CD38 on immune cells is documented, but a direct causal link to tissue NAD+ decline has not been shown in a controlled study.

How is NAD+ measured in human tissue?

The 2012 skin study used enzymatic assays on surgical tissue. Later work has used liquid chromatography with tandem mass spectrometry on plasma (Clement et al., 2019) and 31P magnetic resonance spectroscopy to read NAD+ and NADH in the living brain (Zhu et al., 2015). Each method measures something slightly different, and sample handling has a large effect on the result.

Is low NAD+ a cause of ageing or a marker of it?

Nobody has settled this. The correlations in human tissue are consistent, but the Massudi group themselves wrote that the therapeutic benefit of restoring NAD+ "is yet to be established". The mouse knockout data suggest a causal role for CD38-driven destruction, which still needs testing in people.

Sources

Research use only

NAD+ from LynxLab

500mg vials from £49.99. ≥99% purity, lab report on the page. Same-day dispatch, UK stock.

Buy NAD+