MOTS-c: Mitochondrial Microprotein Structure and Signalling Mechanism

MOTS-c is a 16-residue microprotein translated from a 51-base-pair short open reading frame located inside the mitochondrial 12S ribosomal RNA gene, MT-RNR1, spanning mitochondrial DNA positions m.1343 to m.1393. Lee and colleagues reported the sequence in Cell Metabolism in 2015, naming it for the reading frame it occupies.

One production detail is often blurred. Although the reading frame sits in mitochondrial DNA, the peptide is understood to be translated in the cytoplasm using the universal genetic code, because tandem start and stop codons under the mitochondrial code would not yield the observed 16-residue product. The gene is mitochondrial; the ribosome is not. Review work describes the sequence as conserved across 14 species, the first 11 residues most strongly.

The physical chemistry follows from the sequence. The formula C101H152N28O22S2 carries two sulfur atoms, both from the methionine residues at positions 1 and 6, and no cysteine appears in the chain, so disulfide chemistry is not a consideration. PubChem computes an XLogP of minus 3.9 and a topological polar surface area of 890 square angstroms, describing a small, strongly polar peptide that does not cross lipid bilayers readily.

MOTS-c reaches AMP-activated protein kinase indirectly. Lee and colleagues reported in 2015 that the peptide inhibits the folate cycle and the de novo purine biosynthesis pathway tethered to it. Blocked purine synthesis raises intracellular AICAR, an AMP mimetic that binds and activates AMPK, with skeletal muscle identified as the principal target tissue.

The metabolomic signature reported in that work is specific and testable. Cells exposed to the peptide showed a concerted fall in 5-methyl-tetrahydrofolate, the most abundant activated folate, a fall in methionine, and a rise in homocysteine. Expression changes across folate-methionine cycle and purine synthesis enzymes were detectable four hours after exposure.

Downstream of AMPK the literature converges on the PGC-1alpha axis. Gudiksen and colleagues tested the dependency in Free Radical Biology and Medicine in 2026 using two transgenic mouse strains, and reported that augmented skeletal muscle mitochondrial bioenergetic capacity required both PGC-1alpha and AMPK. Reactive oxygen species emission and oxidation-related protein damage fell with no change in mitochondrial respiratory protein content, pointing to a change in the intrinsic quality of existing mitochondria rather than in mitochondrial volume.

Cardiac tissue supplies a numerical example. In high-resolution respirometry published by Pham and colleagues in 2025, diabetic rat mitochondria from animals not receiving the peptide showed oxygen flux in the complex I plus complex II oxidative phosphorylation state 13 percent below control, a difference absent in the exposed group. That same group showed elevated hydrogen peroxide production at the leak state, running counter to the redox findings in skeletal muscle.

MOTS-c translocates from the cytoplasm into the nucleus under metabolic stress. Kim and colleagues showed in Cell Metabolism in 2018 that the movement is AMPK-dependent, and that nuclear MOTS-c regulates a broad set of genes during glucose restriction, including genes carrying antioxidant response elements, while interacting with stress-responsive transcription factors such as NFE2L2/NRF2.

The significance of that paper sits in its framing. Nuclear factors had long been known to govern the mitochondrial genome; the 2018 work established traffic in the opposite direction, with a mitochondrially encoded peptide binding promoter regions of nuclear genes. Later review work adds activating transcription factors 1 and 7.

The NRF2 connection recurs across recent mechanistic work. Zhang and colleagues reported in Redox Biology in 2026 that an engineered analogue acted through increased nuclear NRF2 translocation, with the effect abolished under NRF2 inhibition and in NRF2-knockout animals. Li and colleagues reported in Life Sciences in 2026 that the peptide appears to interact with KEAP1, the sensor that holds NRF2 inactive, maintaining the KEAP1-PGAM5 interaction and blocking downstream AIFM1 nuclear translocation; KEAP1 overexpression abolished the effect.

MOTS-c has documented actions that do not require AMPK. Jia and colleagues reported in Theranostics in 2024 that the peptide binds the carboxy-terminus of TRIM72, a plasma-membrane repair protein, and accelerates its trafficking to damaged membrane. Knockdown of TRIM72 by siRNA blocked the effect, while an AMPK inhibitor did not.

That paper also identified a direct lipid interaction. Dot blot analysis showed binding to phosphatidylinositol 4,5-bisphosphate, and in TRIM72-knockout mice the peptide still increased fusion of vesicles with the membrane, indicating a second route independent of the protein partner. On its own it had negligible effect on membrane repair, making it a facilitator of TRIM72 traffic rather than a repair agent itself.

Cell entry remains open. Published work describes intracellular targets rather than a characterized cell-surface receptor, and the high polarity of the molecule is described as a limiting factor for membrane permeability. Zhang and colleagues addressed this in 2026 by substituting alanine for the arginine at position 13, raising computed hydrophobicity from minus 0.938 to minus 0.544, and showed that the R13A analogue enters cells through LAT1-mediated transport.

Context also changes the result. Xing and colleagues reported in Inflammation and Regeneration in 2026 that human mesenchymal stromal cells from donors at or above 30 kg per square metre body mass index expressed less of the peptide than cells from donors below that threshold, and that supplying it activated AMPK signalling while reducing proliferation and raising p16, p21 and TNF-alpha expression. Metabolic signalling and reparative function came apart in that system.

The m.1382A>C polymorphism in mitochondrial DNA replaces the lysine at position 14 of MOTS-c with glutamine, producing the variant designated K14Q. Fuku and colleagues described the variation in Aging Cell in 2015 as specific to Northeast Asian populations and carried on the ancestral haplogroup D4b2, which had been reported in association with exceptional lifespan in Japanese cohorts.

Zempo and colleagues tested the functional consequences in Aging in 2021. Meta-analysis of the J-MICC, MEC and TMM cohorts, totalling 27,527 individuals, associated the C allele with higher prevalence of type 2 diabetes in males but not females, and described the substitution as altering peptide structure in a way that reduces insulin-sensitizing activity. An expanded analysis in 736 individuals showed no effect on lifespan. For laboratory work the variant is a naturally occurring single-residue structure-activity probe, and a genotype that varies by ancestry, which can confound human sample sets.

MOTS-c is quantified in plasma either by immunoassay or by liquid chromatography coupled to mass spectrometry. Knoop, Thomas and Thevis validated an LC/MS method against World Anti-Doping Agency laboratory standards in Rapid Communications in Mass Spectrometry in 2019, reporting a lower limit of detection of 100 pg/mL for the human peptide, alongside four metabolites and two oxidation products.

That paper contains a discrepancy anyone reading this literature should know about. Endogenous reference concentrations obtained by ELISA sat between 45.9 and 218.5 ng/mL, while the mass spectrometric method did not return values consistent with that range. The authors flagged the mismatch as warranting further investigation. Immunoassay-derived concentrations are method-specific numbers, and results from different platforms should not be pooled.

Relative change is measured more consistently than absolute level. Reynolds and colleagues reported in Nature Communications in 2021 that in 10 sedentary males performing stationary cycling, skeletal muscle expression rose 11.9-fold immediately after the session, while the circulating pool rose 1.6-fold during and 1.5-fold after, returning to baseline within four hours. Gudiksen and colleagues added a sourcing observation in 2026: interstitial concentrations rose during exercise, but no arterio-venous difference across the limb was detectable.

MOTS-c is supplied as a lyophilized solid. Peer-reviewed handling data is thin, and most storage and solubility figures in circulation come from supplier specifications rather than published measurement, so they belong in a method section labelled as such. The peer-reviewed material that exists concerns analytical stability: Knoop and colleagues evaluated long-term stability and in vitro metabolism while validating their assay.

The chemistry points to one specific liability. Methionine occupies positions 1 and 6 and no cysteine appears in the chain, so methionine sulfoxide formation is the oxidation route to anticipate and disulfide scrambling is not. Purity and stability assessments should resolve the oxidized species explicitly rather than reporting a single main-peak percentage.

Solution behaviour follows from the computed descriptors. An XLogP of minus 3.9 and a topological polar surface area of 890 square angstroms describe a strongly hydrophilic molecule, and the 2026 Redox Biology work states that membrane permeability is poor because of that polarity, which is why an alanine substitution at position 13 was engineered. Groups studying uptake should expect the unmodified sequence to enter cells inefficiently and should include an uptake control.