Could GLP-1 drugs affect mitochondria?

A paper published on 24 September 2026 in the journal Obesity proposes a hypothesis, not a finding: that incretin-based therapies such as tirzepatide may loosen the coupling between mitochondrial electron transport and ATP production, a process researchers call oxidative phosphorylation, or OXPHOS, uncoupling^[1]. The authors, Bryn Falahee, Caroline Apovian and Nawfal Istfan, are careful to label it a testable hypothesis rather than an established mechanism. Nobody has yet run the experiment that would confirm or reject it in humans.

What the hypothesis proposes

The paper is a narrative literature review: the authors read and synthesised existing research instead of running a new trial^[1]. Their argument runs like this: GLP-1 receptor agonism appears to counteract insulin signalling during periods of nutrient excess by switching on anti-inflammatory pathways that resemble a fasting state. Separately, preclinical work in animals has shown that incretin-based therapies can activate thermogenic genes and uncoupling proteins, the molecular machinery that lets mitochondria release stored energy as heat instead of converting all of it to ATP.

Put those two observations together and the authors arrive at a proposal: that drugs in this class, which include tirzepatide and semaglutide, might nudge mitochondria toward a less efficient, more heat-generating mode of operation. If that is true, it could help explain why incretin-based weight loss also tends to come with reduced markers of oxidative stress. The paper frames this as a relationship worth testing directly, not as something the current evidence already proves.

The reasoning draws on decades of separate incretin research, not a single new dataset. GLP-1 receptor agonism was already known to shift how cells respond to a surplus of nutrients, and rodent studies have linked incretin exposure to increased expression of uncoupling protein 1 in fat tissue. What the new paper does is connect that older, scattered evidence to a specific clinical question: whether the same uncoupling effect extends to skeletal muscle mitochondria in humans at the doses used for diabetes and weight management. That extension is the untested part.

The case that started the inquiry

The authors say their interest was prompted by a single clinical observation: a patient with an existing mitochondrial myopathy who developed acute muscle weakness after exposure to tirzepatide. Patients with mitochondrial myopathies already have impaired oxidative phosphorylation in their muscle cells. If incretin-based therapies do uncouple OXPHOS at the cellular level, a patient whose mitochondria are already working with a smaller safety margin could be more likely to notice the effect as muscle weakness. That is the chain of reasoning behind the hypothesis, and it is exactly why the authors are asking for it to be checked, not treating it as confirmed.

What this hypothesis is not

It is not a change to the safety label. The current US prescribing information for Mounjaro (tirzepatide), the branded product cleared by the FDA, lists gastrointestinal effects such as nausea, diarrhoea and vomiting as the common adverse reactions and carries a boxed warning about thyroid C-cell tumours seen in rats^[2]. It does not describe a muscle weakness or mitochondrial warning of any kind. Nor is this the same question as the well-documented lean-mass loss that shows up in every large GLP-1 weight-loss trial. A body-composition analysis of the SURMOUNT-1 trial found that people on tirzepatide lost a mean of 10.9% of lean mass alongside 33.9% of fat mass at 72 weeks, compared with 2.6% lean mass loss on placebo^[3]. That is a separate, already-established phenomenon tied to large-scale weight reduction generally. The new hypothesis is narrower and more specific: a proposed direct effect on how mitochondria handle energy, distinct from the ordinary muscle loss that comes with losing a large amount of body weight quickly.

What we still do not know

The paper does not report a clinical trial, a biopsy series, or even a case series. It is one case observation plus a synthesis of preclinical and mechanistic literature, and the authors say as much. Whether OXPHOS uncoupling actually occurs in human skeletal muscle at doses used for weight management, how common it might be, and whether it matters clinically outside patients who already have a mitochondrial disorder, are all open questions the paper does not answer. The European Medicines Agency's authorisation for tirzepatide, granted in September 2022, describes its mechanism as activating the GLP-1 and GIP receptors and makes no reference to mitochondrial uncoupling as part of the drug's approved profile^[4].

Why this matters

For most people taking an incretin-based therapy, this paper changes nothing about known risks or the current safety profile. Its value is as a research direction: if the mechanism holds up under real testing, it could help explain part of how these drugs improve metabolic health beyond calorie reduction alone, and it flags a small, specific population, people with a diagnosed mitochondrial disorder, who may warrant extra caution and monitoring before starting therapy. Anyone with a known mitochondrial disease considering tirzepatide or another incretin-based drug should raise that history directly with their prescriber rather than relying on this hypothesis either way. For the general picture of how tirzepatide is regulated and what its trial programme has and has not shown, see the tirzepatide regulation overview.

The next useful step, per the authors, would be a study that measures OXPHOS activity directly in muscle tissue from patients on incretin-based therapy, ideally with a dose-response design and both healthy volunteers and people with a known mitochondrial disorder. Until that work exists, the honest summary is a plausible mechanism worth testing, built from a single case and a review of older, separate research, not a confirmed effect of tirzepatide or any other incretin-based drug.