Mitochondria and energy: NAD+, MOTS-c, and why exercise is still the main lever

What happens to mitochondria with age

Mitochondria convert fuel and oxygen into ATP, and a muscle cell contains thousands of them. With age and inactivity three things happen: there are fewer of them per cell, the ones that remain are less efficient — they leak more electrons as reactive oxygen species and make less ATP per unit of oxygen — and the quality-control systems that recycle damaged ones (mitophagy) slow down. Mitochondrial DNA, which sits outside the nucleus with weaker repair, accumulates mutations. The felt result is lower VO2max, slower recovery, a harder time burning fat, and the afternoon fatigue that sends people looking for a supplement.

The important nuance is that most of this is disuse, not time. Muscle biopsies from lifelong endurance athletes in their seventies show mitochondrial density and function close to those of active people in their twenties. Ageing lowers the ceiling; sitting lowers the floor by far more.

Exercise: the lever with the evidence

The clearest human experiment is Robinson et al. 2017 in Cell Metabolism: 72 adults, young (18–30) and older (65–80), randomised to twelve weeks of high-intensity interval training, resistance training or a combination, with muscle biopsies before and after. In the older group, interval training raised mitochondrial respiratory capacity by 69% and reversed the age-related decline in the expression of hundreds of mitochondrial genes — the older adults' muscle moved towards a younger profile on the biopsy. Resistance training built more muscle; intervals built better mitochondria.

Behind that sits the mechanism: muscle contraction activates AMPK and PGC-1α, the master switch for making new mitochondria, in proportion to the energy demand. Zone-2 work increases the number and the fat-burning capacity; intervals increase the respiratory capacity per mitochondrion. Both, in the week described in the VO2max article, is the only intervention that has been shown to rebuild the organelle in a person.

NAD+: what the trials actually found

Nicotinamide adenine dinucleotide is the coenzyme every mitochondrion uses to shuttle electrons, and tissue NAD+ falls with age — by perhaps half in skin and muscle between twenty and eighty, though the figure is tissue-specific and the human data are thinner than the claims. The precursors nicotinamide riboside (NR) and nicotinamide mononucleotide (NMN) reliably raise blood NAD+, and that is where the certainty ends.

The honest summary: safe at the doses tested, reliably raises the blood number, and so far produces small or no changes in anything a person would notice. Nobody has shown a precursor rebuilds mitochondria or changes an outcome. NAD+ infusions add a further problem: the molecule is unlikely to cross cell membranes intact, the dose reaching tissue is unknown, there is not one controlled trial for "energy" or ageing, and the flushing and nausea people report during a drip is not evidence that something good is happening.

MOTS-c: a mitochondrial peptide

MOTS-c is a 16-amino-acid peptide encoded, unusually, by the mitochondrial genome itself, discovered by Changhan Lee and Pinchas Cohen in 2015. In mice it acts like an exercise signal: it activates AMPK, improves insulin sensitivity, prevents diet-induced obesity, and in a 2021 Nature Communications study (Reynolds et al.) treated old mice doubled their running capacity and treated young mice ran further than controls. Levels fall with age in humans and rise after exercise, and a rare variant is associated with longevity in Japanese populations.

That is a genuinely interesting biology, and it is also the whole story so far: no completed human trial of MOTS-c for fitness, metabolism or ageing has been published. A synthetic analogue (CB4211) completed a phase 1b in fatty liver disease. It belongs in the "early" column, studied in research settings for metabolic and exercise-capacity outcomes, and anyone presenting it as a proven mitochondrial treatment is ahead of the data.

Hype check

What to measure

• VO2max and pace at a fixed zone-2 heart rate — the functional read-out of mitochondrial capacity, and the only one that changes decisions.
• Ferritin, haemoglobin, B12, folate, thyroid: the commonest reasons for "no energy" are not mitochondrial, and a protocol that skips them to reach NAD+ has missed the point.
• Fasting insulin and HbA1c: insulin resistance is mitochondrial dysfunction seen from the glucose side.
• Sleep duration and apnoea screen, for the same reason.
• Lactate at a fixed workload if you have access to a lab — falling lactate at the same power is mitochondria improving.
• Not a blood NAD+ level. It rises with any precursor and tells you nothing about tissue.

Where peptides and therapies fit

The research library has pages on NAD+ and MOTS-c that describe what each is studied for, the labs and the mistakes, and the therapies section covers NAD+ infusions and injections with the same grading used here. All three say the same thing this article does: the compound layer sits behind training, sleep and a corrected iron or thyroid result, and the longevity goal page shows the sequence a protocol follows before anything mitochondrial is added. Where a compound does belong, it is as an addition to a plan that already has the interval session in it, not a substitute for it.

Common mistakes

• Taking NAD+ for fatigue before checking ferritin, thyroid and sleep. The expensive answer to a cheap question.
• Reading a higher blood NAD+ as a result. It is the one thing precursors reliably do, and it has not been linked to anything you would feel.
• Paying for an infusion without asking for the trial that supports it. There is not one.
• Expecting MOTS-c to replace the interval session. The mice still ran.
• Doing only zone 2 and wondering why the capacity per mitochondrion has stalled.
• A "mitochondrial stack" of six supplements with a combined evidence base of two small trials.