# Inside the cell, NAD+ has separate rooms

> NAD+ tissue biology resists a whole-body shortcut: CD38 in ageing mice, different clocks in fat and muscle, and an enzyme seen by crystallography.

**The tissue report**

The body is not one reservoir with one gauge. The fat-and-muscle clock experiment is the most awkward, and most useful, illustration of that problem.

## Carrying electrons and spending the molecule

NAD+ cycles through electron-carrying work, but some enzymes consume it as a substrate. Sirtuins, PARPs and CD38 draw on that supply for different jobs, described in the [review of NAD+ metabolism during ageing](https://pubmed.ncbi.nlm.nih.gov/33353981/). PARPs use NAD+ in DNA repair; sirtuins connect it to metabolic and stress responses. The cell has to make replacement material.

My question here concerns replacement: how much of a tissue’s response comes from making NAD+, and how much from spending it? A circulating measurement cannot separate those processes inside every organ.

## A missing enzyme changed old mice

In the [CD38 deletion experiment](https://pubmed.ncbi.nlm.nih.gov/27304511/), mice without that NAD+-consuming enzyme were protected against age-related NAD+ decline, with preserved SIRT3 activity and improved mitochondrial function. The intervention was genetic deletion in mice, with its own set of biological consequences.

The consumption side of this problem commands more of my curiosity than another bigger blood number: changing consumption changed the supply. The usual story centres on adding material; this experiment also asks where material is going. A report about losses tells me something a report about supply alone misses.

## Fat and muscle disagreed about the time

The [NAMPT deletion study in mice](https://pubmed.ncbi.nlm.nih.gov/36996103/) reduced NAD+ in different tissues. Brown fat lost 89% of circadian transcripts, and white fat lost 77%; skeletal muscle maintained normal clock function despite an equivalent NAD+ reduction. Circadian transcripts are gene messages whose levels normally change with the daily clock.

That is a wonderfully inconvenient result. I expected a shared reduction to yield a more uniform response, but the muscle clock did not follow the fat clocks. An “NAD+ level” without a tissue attached leaves out the very distinction this experiment exposes. It also gives no best time for a human pill or injection.

## The finishing enzyme was seen, not swallowed

The [X-ray structure of human NMN adenylyltransferase](https://pubmed.ncbi.nlm.nih.gov/11751893/) revealed the enzyme that catalyses the final NAD+ biosynthesis step involving NMN and ATP. The investigators examined a human enzyme in an in vitro structural study. The [biosynthesis comparison across 45 species](https://pubmed.ncbi.nlm.nih.gov/22404877/) places those routes in a much older biological context.

NAMPT, discussed in the [review of the salvage pathway](https://pubmed.ncbi.nlm.nih.gov/35903330/), controls an earlier recycling step. I enjoy this machinery for its own sake. Knowing how a molecule is made is a substantial finding, even when it cannot answer whether a person benefits from taking a precursor.

## Cancer cells complicate the supply question

The [review of NAD+ applications and unknowns](https://pubmed.ncbi.nlm.nih.gov/37364580/) discusses a theoretical cancer concern: NAD+ supports proliferating cells, so boosting it could have context-dependent consequences for an existing cancer. The research has not settled a general helpful-or-harmful answer.

For that reason, “More is better” is a slogan I won’t accept as a general account of this biology. The blood trials, the mouse tissue experiment and the cancer question have different destinations. None can erase the others just because the molecule's name is the same.

---

Documentation of the NAD+ record: the precursor trials logged green, the IV and injection gaps flagged amber, the species and route on every figure — no clinic behind the docs and nothing here dispensed, dosed, or sold.