MOTS-c Peptide: The Mitochondrial Microprotein That Mimics Exercise in Mice, What the Studies Measured, and the Two Human Trials

MOTS-c Peptide: The Mitochondrial Microprotein That Mimics Exercise in Mice, What the Studies Measured, and the Two Human Trials

For decades human mitochondrial DNA was thought to encode thirteen proteins, two ribosomal RNAs, a set of transfer RNAs and nothing else. In 2015 a group at the University of Southern California reported a 51-base-pair reading frame hiding inside one of those ribosomal RNA genes, translated into a 16-amino-acid peptide that prevented obesity in mice on a high-fat diet through the same metabolite the doping compound AICAR supplies. Six years later the same laboratory showed that hard cycling raises the peptide nearly twelvefold in human muscle and that injecting it makes 22-month-old mice run twice as long. Those results, retold without the words 'mouse' and 'injected', are why MOTS-c is sold as an exercise mimetic, a fat-loss peptide and a longevity compound. This reference sets out what has been shown and in what species, how each study was designed, what the one completed human trial found with a modified analog, what the 2026 trial of the peptide itself is measuring, and where the anti-doping and compounding regulators have landed. It replaces an earlier version of this page that printed a human protocol; that table is gone, and the reason is explained.

Michael Phelps

Founder & Peptide Research Specialist, PrymaLab

Research reference · Last updated September 23, 2026 · ~26 min read

What MOTS-c peptide is: sixteen amino acids read off a ribosomal RNA gene

MOTS-c is a peptide of sixteen amino acids, Met-Arg-Trp-Gln-Glu-Met-Gly-Tyr-Ile-Phe-Tyr-Pro-Arg-Lys-Leu-Arg, with the formula C₁₀₁H₁₅₂N₂₈O₂₂S₂, a molar mass of 2,174.62 g/mol and the CAS number 1627580-64-6.^[1] What makes it unusual is not the sequence but the address. It is encoded by a 51-base-pair short open reading frame inside the mitochondrial 12S ribosomal RNA gene, MT-RNR1, a stretch of DNA that was assumed to make a structural RNA and nothing more. The name is an acronym for exactly that: mitochondrial open reading frame of the 12S rRNA type-c.^[2]

The discovery was made in silico first. Changhan Lee, in Pinchas Cohen's laboratory at the University of Southern California, searched the 12S rRNA for reading frames and found one of 51 base pairs 'with a strong Kozak sequence' that 'translates into a 16 amino acid peptide'. The group then showed the peptide was real: detectable 'in various tissues in mice and rats, as well as in circulation in human and rodent plasma' by an ELISA they built for the purpose. One control recurs in later arguments about the peptide's origin: nuclear DNA holds fossil copies of mitochondrial sequences, called NUMTs, so a skeptic could argue the peptide came from the nucleus. Rats have no NUMT for MOTS-c, which makes 'the mtDNA its exclusive source' in that species, and the peptide is there.^[2]

MOTS-c belongs to a small family. Humanin was reported in 2001 from the 16S rRNA gene; MOTS-c followed in 2015, and six small humanin-like peptides, SHLP1 to SHLP6, after that. A 2022 review from the Cohen group counts eight published MDPs and calls them 'microproteins'.^[3] The term matters because 'mitochondrial peptides', as the market uses the phrase, sweeps in compounds such as SS-31 that act on mitochondria but are not encoded by them; the comparison section below keeps the two ideas apart.

How MOTS-c works: the folate pathway, AICAR and AMPK, then the nucleus and CK2

The 2015 paper did something rare for a peptide discovery: it found the mechanism in the same study, by metabolomics. Cells engineered to express MOTS-c showed 'a concerted decrease in the levels of 5-methyl-tetrahydrofolate (5Me-THF), the most abundant form of activated folate, and methionine', a block in de novo purine synthesis, and 'an accumulation of endogenous AICAR (5-aminoimidazole-4-carboxamide ribonucleotide) to levels higher than 20-fold' compared with control cells. AICAR is the natural activator of AMP-activated protein kinase, and AMPK is the enzyme that shifts a cell toward glucose uptake and fat oxidation; the paper's summary is that MOTS-c's 'cellular actions inhibit the folate cycle and its tethered de novo purine biosynthesis, leading to AMPK activation', with skeletal muscle as 'its primary target organ'.^[2] Anyone who has read the AICAR reference will recognize the endpoint: the drug AICAR supplies the metabolite from outside; MOTS-c makes the cell accumulate its own.

Three years later the same laboratory added a second arm. Under metabolic stress, glucose restriction in particular, MOTS-c 'translocates to the nucleus and regulates nuclear gene expression' in an AMPK-dependent manner, where it 'regulated a broad range of genes', including 'those with antioxidant response elements (ARE)', and 'interacted with ARE-regulating stress-responsive transcription factors, such as ... NRF2'.^[4] A peptide encoded by one genome entering the other genome's nucleus to change transcription was the paper's point, and it is the basis of the phrase 'mitonuclear communication' that appears in MOTS-c product copy.

In 2024 a direct binding partner was reported. Casein kinase 2 (CK2) 'is a direct and functional target of MOTS-c': the peptide 'directly binds to CK2 and activates it in cell-free systems', and in mice it 'prevented skeletal muscle atrophy' and raised muscle glucose uptake, effects that 'were blunted by suppressing CK2 activity'. The effect is tissue-specific, stimulating CK2 in muscle while suppressing it in fat, and the naturally occurring K14Q variant 'has reduced binding to CK2 and does not activate it or elicit its effects'.^[5] That last finding ties the mechanism to the human genetics below. What the three papers do not contain is a human: the folate-cycle data are from HEK293 cells, the nuclear data from cells under glucose restriction, and the CK2 data from cell-free systems and mice.

What MOTS-c did in mice: obesity, old muscle, and running

The 2015 findings in animals are the foundation of every claim made for the peptide. Outbred CD-1 mice on a 60% high-fat diet treated with MOTS-c by intraperitoneal injection were 'remarkably' prevented from becoming obese, while the same treatment 'had no effect on body weight in mice fed a normal diet'; three weeks of treatment raised the respiratory exchange ratio and heat production, consistent with more glucose burned. In a separate experiment, C57BL/6 mice on a high-fat diet treated for 7 days showed improved whole-body and skeletal-muscle insulin sensitivity under euglycemic clamps. And because muscle and circulating MOTS-c 'decline concomitantly with the development of insulin resistance during aging in mice', the group tested old animals: soleus muscle from 12-month-old mice was more insulin-resistant than from 3-month-olds, 'but 7 days of MOTS-c treatment restored sensitivity in the old mice to levels comparable to young animals'.^[2]

The 2021 Nature Communications paper extended the animal work to physical performance, and its numbers are the ones that travel. Young mice treated for two weeks improved on the rotarod 'but not grip strength' and showed no change in a memory test. On a high-fat diet, two doses were compared; only the higher dose improved treadmill running, and it did so after 10 days but not 7. In the sprint stage of the test, 100% of the high-dose mice reached the top speed against 16.6% of the low-dose and control groups. Middle-aged (12-month) and old (22-month) mice treated at the higher dose for two weeks ran longer with more power output; the old mice 'ran longer (2-fold) and farther (2.16-fold)', 17% reached the final stage where none of the untreated did, and treated old mice outperformed untreated middle-aged ones.^[6]

The most cited experiment is the late-life one. Mice started on MOTS-c at 23.5 months were treated intermittently for the rest of their lives, and past 30 months of age showed better grip strength, longer stride, a better 60-second walking test ('running was not possible at this age'), lower blood glucose, less fat mass and 'a modest reduction in age-dependent loss of lean mass'.^[6] Two smaller studies fill in the record. Ovariectomized mice treated for 12 weeks lost less bone on micro-CT, with osteoclast formation inhibited in an AMPK-dependent way; that paper is the basis of the 'osteoporosis' half of the 2026 compounding nomination.^[7] And in obese mice MOTS-c lowered circulating myostatin, the negative regulator of muscle mass, with the human observation that 'plasma MOTS-c levels are inversely correlated with myostatin levels'.^[8]

MOTS-c and exercise in humans: the 11.9-fold rise, and the measurement problem

The human part of the 2021 paper is a measurement study, and it is the strongest human evidence that MOTS-c does anything physiological. Sedentary healthy young men (24.5 ± 3.7 years, BMI 24.1 ± 2.1) rode ten 60-second cycling intervals at their individual peak power, with muscle biopsies before and after and plasma sampled during. Skeletal-muscle MOTS-c 'significantly increased after exercise (11.9-fold) and remained elevated after a 4-h rest', while circulating MOTS-c rose 1.6-fold during and 1.5-fold after exercise and returned to baseline at four hours.^[6] That result is the whole basis of 'exercise mimetic': the body makes more of the peptide when it exercises, and giving the peptide to mice reproduces some of what exercise does to them.

Cross-sectional studies add correlations. Plasma MOTS-c was associated with insulin sensitivity 'in lean but not in obese individuals' in one cohort, and was lower in obese boys and adolescents in another, where it tracked insulin resistance.^[9][10] These are associations in the direction the mouse work predicts, and they are also the only kind of human evidence available for the native peptide until the 2026 trial reports.

There is a complication that product pages never mention. The 2019 anti-doping paper that built a mass-spectrometry test for MOTS-c in plasma, validated to WADA's laboratory standard with a detection limit of 100 pg/mL, found that the 'endogenous levels of a reference population of 20 healthy subjects studied by ELISA experiments (45.9-218.5 ng/mL) could not be confirmed by LC/MS'.^[11] An immunoassay reporting tens to hundreds of nanograms per milliliter, and a specific mass-spectrometric method that cannot see the peptide at a thousandth of that concentration, cannot both be right about the free peptide. Fold-changes measured by ELISA, including the 1.5-fold plasma rise, are relative and survive the disagreement; absolute 'normal levels' of MOTS-c, quoted anywhere, do not yet rest on a method that two techniques agree on.

The K14Q variant: what human genetics says about MOTS-c

Human mitochondrial DNA carries a common East Asian variant, m.1382A>C (rs111033358), inside the MOTS-c reading frame. It changes the fourteenth amino acid from lysine to glutamine, K14Q, and it has become the natural experiment for what MOTS-c does in people. The first paper, from Fuku and colleagues in 2015, noted that the variant 'is specific for the Northeast Asian population' and proposed it 'may be among the putative biological mechanisms explaining the high longevity of Japanese people'.^[12] Later work pointed the other way. A 2021 meta-analysis of three Japanese cohorts, 27,527 people in all, found that 'males but not females with the C-allele exhibit a higher prevalence of T2D', and in the largest cohort 'only males with the C-allele in the lowest tertile of physical activity increased their prevalence of T2D', which the authors call a kinesio-genomic interaction. High-fat-fed male mice given wild-type MOTS-c lost weight and improved glucose tolerance; mice given K14Q-MOTS-c did not.^[13]

The muscle-fiber paper complicates the 'exercise mimetic' story in an interesting way. Carriers of the C allele had more fast-twitch MHC-IIx fiber and, in men, higher peak leg torque; the allele was more common in sprint and power athletes (6.5%) than in controls (5.1%) and rarer in endurance athletes (2.9%).^[14] Less MOTS-c activity favors the sprinter's muscle; more favors the endurance phenotype the running studies measured. The genetics are consistent with a peptide that matters for metabolism and muscle in people. They say nothing about what happens when the peptide is supplied from outside, because no carrier has been given it.

The two human trials: an analog in twenty people, and a 2026 trial of the peptide itself

Until 2026, the only human trial connected to MOTS-c did not use MOTS-c. CohBar, a company founded around Cohen's mitochondrial peptides, developed CB4211, described in its own materials as 'a novel and improved analog of MOTS-c', and ran a Phase 1a/1b study registered as NCT03998514 from July 2018 to April 2021 with 88 participants.^[15][16][17] The Phase 1a enrolled 65 healthy adults in single and 7-day multiple ascending-dose cohorts against placebo. The Phase 1b was a randomized, double-blind, placebo-controlled four-week evaluation of CB4211 in 20 obese subjects with nonalcoholic fatty liver disease, eleven on drug and nine on placebo, all with at least 10% liver fat, kept inpatient on a standardized diet.^[15][16]

The topline, announced in August 2021 and presented at the AASLD Liver Meeting that November: at week four ALT fell 21% on CB4211 against a 4% rise on placebo, AST fell 28% against 11%, and glucose fell 6% against no change, each significant at p<0.05; body weight showed only a trend, about half a percent; and liver fat fell by almost the same amount in both groups (5.03% versus 4.88%), so the drug did not beat the inpatient diet on the endpoint the indication most needed. Safety was clean: 'no serious adverse events', and the only adverse events in more than 10% of treated subjects were 'transient and generally mild to moderate injection site reactions'.^[15][16] The program did not continue. In May 2023 CohBar agreed to merge with Morphogenesis, and the combined company, TuHURA Biosciences, is built around immuno-oncology assets; no further CB4211 development has been announced.^[18] Two cautions follow: CB4211 is a modified molecule whose structure CohBar did not publish, and its doses cannot be mapped onto the native peptide.

The trial of MOTS-c itself is new. NCT07505745, acronym MOTS-MET, is a Phase 2a, randomized, double-blind, placebo-controlled study of MOTS-c 'in Adults With Prediabetes and Overweight/Obesity', sponsored by Hudson Biotech and recruiting at Peking University Shenzhen Hospital. It plans 120 adults aged 18 to 65 with a BMI of 27 to 40 and documented prediabetes, randomized 1:1 to MOTS-c or placebo for 12 weeks. The primary outcomes are the change in insulin sensitivity by the OGTT-derived Matsuda index at 12 weeks and treatment-emergent adverse events through week 16; secondaries include HbA1c, fasting and 2-hour glucose, and anti-drug antibodies. It started on 2 February 2026, was first posted on 1 April 2026, and lists primary completion for February 2027.^[19] Until it reports, there is no efficacy or safety result for native MOTS-c in any human.

What the MOTS-c studies measured: models, durations and outcomes

The previous version of this page carried a table of human 'protocols'. It has been removed, because no such protocol exists in the literature: there is no published human study of native MOTS-c with results, and the only completed human trial used an analog. What the literature contains is a short list of study designs. Lee 2015 measured obesity, insulin sensitivity and glucose in high-fat-fed and normally fed mice over days to weeks. Kim 2019 measured the plasma metabolite profile in diet-induced obese mice. Ming 2016 measured bone loss on micro-CT in ovariectomized mice over 12 weeks. Reynolds 2021 measured rotarod, treadmill running, grip, gait and glucose in young, middle-aged and old mice, from two weeks to the end of life. In people, the CB4211 Phase 1a measured safety and pharmacokinetics in 65 healthy adults, the Phase 1b measured liver enzymes, glucose, body weight and liver fat in 20 adults with fatty liver over four weeks, and the MOTS-MET Phase 2a (NCT07505745) is measuring insulin sensitivity over 12 weeks.

Two things stand out. Every mouse study gave the peptide by intraperitoneal injection, and there is no oral data for the native peptide beyond the single negative test described below. And in the one study that compared two doses head to head, the lower one did nothing for running.^[6] Mouse results do not transfer to people, and the only completed human trial belongs to a different molecule.

Is there an oral MOTS-c?

Not one that has been shown to work. A 2023 colitis study is the only published test: intraperitoneal MOTS-c improved dextran-sulfate colitis in mice, but 'oral administration of MOTS-c did not result in any significant improvements', and the authors had to attach a cell-penetrating peptide to the C-terminus to make an orally active molecule.^[20] A 16-residue peptide with two arginines and a lysine is a meal for the gut's proteases. Tablets, capsules and 'oral MOTS-c' products sell a route that the one experiment on it found inactive.

For research material the practical questions are handled elsewhere: concentration arithmetic on the peptide calculator, technique in the reconstitution guide, and the diluent in the bacteriostatic water reference. PrymaLab's 10 mg and 40 mg vials and autoinjector are supplied for laboratory research use only, and nothing on this page or those pages is a protocol.

MOTS-c side effects: what has been reported, and by whom

The honest answer to 'what are the side effects of MOTS-c' is that for the native peptide in humans there is no safety database at all, and for the analog there is one small one. In the CB4211 Phase 1b, the only adverse events in more than 10% of treated subjects were 'transient and generally mild to moderate injection site reactions', and there were 'no serious adverse events' across the 20 subjects and four weeks; the Phase 1a in 65 healthy adults reported the drug as safe and well tolerated across its ascending doses.^[15][16] That is the entire human safety record connected to the peptide, and it belongs to a different molecule.

The regulator's view of the gap is on the public record. FDA's compounding page states that 'compounded drugs containing MOTs-C may pose significant risk for immunogenicity for certain routes of administration and may have complexities with regard to peptide-related impurities and API characterization', that 'FDA has not identified any human exposure data on drug products containing MOTs-C administered via any route of administration', and that the agency 'lacks important information regarding any safety issues raised by MOTs-C, including whether it would cause harm if administered to humans'.^[21] Immunogenicity is the specific concern: a foreign-looking peptide injected repeatedly can raise antibodies, which is why the 2026 Phase 2a lists 'anti-drug antibodies' among its secondary outcomes.^[19]

The mouse literature reports effects rather than side effects, but two belong here. In the 2015 study, young mice on a normal diet treated for four days showed 'modest reductions in body weight, food intake, and blood glucose levels', so the peptide is not inert in a healthy animal, and that finding is the source of the appetite claims on forums.^[2] And the K14Q genetics show that MOTS-c activity shapes muscle fiber type and diabetes risk in a sex-dependent way, with male carriers affected and females not, or oppositely.^[13][14] A peptide whose loss has sex-specific consequences may have sex-specific effects when supplied, and no study has looked.

SS-31 vs MOTS-c, humanin and AICAR: a pairing that is not a family

MOTS-c is most often sold beside SS-31, and the pairing rests on a category error that the SS-31 reference spells out. SS-31, elamipretide, is a synthetic four-residue peptide that binds cardiolipin in the inner mitochondrial membrane; it is not encoded by mitochondrial DNA, is not a mitochondrial-derived peptide, and has a human record MOTS-c cannot match in either direction, an FDA accelerated approval for Barth syndrome in 2025 and five randomized trials that missed their primary endpoints. MOTS-c is a natural product of the mitochondrial genome acting on the folate cycle, AMPK, the nucleus and CK2, with a mouse record and a human trial that has not reported. The two share the word 'mitochondria' and nothing mechanistic.

Searches for SS-31 and MOTS-c together have a one-line answer: no study has given the two together, in any species. Humanin, the other genuine mitochondrial-derived peptide, has biomarker associations in humans and no trial. AICAR is the closest mechanistic relative, since MOTS-c raises the cell's own AICAR, and the AICAR reference covers its three decades of intravenous trials and its 3,080-patient futility result; it is also MOTS-c's neighbor in the WADA clause.

WADA, FDA compounding and the research-use line

Anti-doping caught MOTS-c early. The Cologne laboratory published its plasma test in 2019 on the reasoning that 'it appears advisable to monitor the potential use of synthetic MOTS-c in sports'; the method includes four metabolites and two oxidation products to lengthen the detection window.^[11] WADA named the peptide explicitly from the 2024 List, and the 2026 Prohibited List places it in S4.4.1, 'Activators of the AMP-activated protein kinase (AMPK)', by its full name, 'mitochondrial open reading frame of the 12S rRNA-c (MOTS-c)', beside AICAR and the uncoupler BAM15, in the same subsection as GW1516, SR9009 and SR9011. Section S4 is prohibited at all times, and S4.4 substances are non-specified.^[1][22]

The compounding story is where 2026 has been eventful. FDA had placed MOTS-c among bulk drug substances that 'may present significant safety risks' for 503A compounding, with the immunogenicity and characterization concerns quoted above; in April 2026 the agency moved it and eleven other peptides out of that category pending review by its Pharmacy Compounding Advisory Committee, and its page now lists 'MOTs-C' among substances 'nominated but withdrawn' with the concern text intact.^[21] For the committee's meeting of 23 and 24 July 2026, FDA's briefing document states that 'FDA is proposing that MOTS-c (free base) NOT be included on the 503A Bulks List', and the same for the acetate.^[23] The committee voted the other way, 7 yes, 5 no, 2 abstentions, to recommend MOTS-c for 'obesity and osteoporosis', one of six peptides recommended that week alongside BPC-157, KPV, TB-500, Epitalon and Semax, with emideltide rejected 6 to 7; some members 'conditioned their recommendations for addition on various safeguard implementations, including requirements for authorized active pharmaceutical ingredient sourcing, adverse-event reporting, and formulations'.^[24]

The vote will be retold as an approval, and it is not one. 'PCAC's votes are recommendations and are not binding on the FDA'; the agency has to run notice-and-comment rulemaking, a process that 'could occur in 2027 or extend into a multi-year process', and 'pharmacies will not be permitted to compound them until the final rulemaking is in place'.^[24][25] Even at the end of that road the result would be permission for licensed pharmacies to compound MOTS-c on prescription; it would not make the peptide an approved drug, would not review any product sold online, and would not change its WADA status. For research-grade material the line is unchanged: legal to sell and hold for laboratory research under the framework the legal status guide describes, not for human use, and prohibited for athletes at all times.

Research-grade MOTS-c: what to verify

MOTS-c is an easy peptide to verify and an easy one to sell degraded. Identity is by mass spectrometry against the 2,174.62 Da parent; purity is by HPLC by lot; and the specific failure to look for is oxidation. The sequence has two methionines, at positions 1 and 6, and each can pick up an oxygen to become methionine sulfoxide, adding 16 Da; a certificate that shows a +16 or +32 Da species beside the parent is showing a peptide that has started to change, and the anti-doping laboratory that built the detection assay included 'two oxidation products' in it for exactly that reason.^[11] The purity testing guide covers the general method, and the storage and stability guide covers methionine oxidation as one of the four chemistries that degrade a peptide.

PrymaLab's 10 mg and 40 mg vials and preloaded autoinjector are documented on that basis; the pen versus vial guide covers what a preloaded format changes for an oxidation-prone sequence. The version of the story the record supports: a genuine product of the mitochondrial genome with a mechanism worked out in cells; a mouse record in which injected doses prevented obesity, restored old muscle and doubled old mice's running; a human peptide that rises with exercise and whose loss-of-function variant changes diabetes risk and muscle type; one small trial of an analog, abandoned in a merger; a trial of the peptide itself with no results; an oral route that failed its only test; a WADA listing; and a compounding vote that FDA opposed and has not acted on.

What this article does not settle

Any human effect of native MOTS-c. No efficacy or safety result exists; the 2026 Phase 2a is the first controlled trial of the peptide itself and its dose is undisclosed.

What CB4211 is. CohBar described it as a modified analog and did not publish its structure, so its results cannot be transferred to the native peptide.

Endogenous concentrations. ELISA and LC/MS disagree by more than a thousandfold; I could not find a reconciliation in the literature, so no 'normal level' is quoted here.

Effect sizes in the 2015 obesity study. The full text was read and the direction is reported; the article does not print body-weight percentages because the figures are graphical and were not extracted.

Sex differences with supplied peptide. The K14Q genetics are sex-specific; no study has compared males and females.

What is well established: the sequence, mass and genomic origin; the folate-AICAR-AMPK mechanism, nuclear translocation and CK2 binding in cells and mice; the mouse outcomes; the 11.9-fold exercise response in human muscle; the K14Q associations across 27,527 people; the CB4211 Phase 1b design, results and safety; the 2026 Phase 2a design; the negative oral result; the WADA listing; and the July 2026 advisory vote against FDA's proposal.

MOTS-c, MDP, sORF, K14Q, CB4211 and MOTS-MET: what the names mean

MOTS-c stands for mitochondrial open reading frame of the 12S rRNA type-c, the name Lee and colleagues gave the 51-base-pair reading frame and its peptide in 2015; the 'c' distinguishes it from other candidate frames in the same gene.^[2] MDP is mitochondrial-derived peptide, the class that also contains humanin and the six SHLPs (small humanin-like peptides); the Cohen group now also uses 'microprotein'.^[3] sORF is short open reading frame. MT-RNR1 is the gene symbol for the mitochondrial 12S ribosomal RNA. K14Q names the amino-acid change, lysine (K) to glutamine (Q) at position 14, produced by the DNA variant m.1382A>C (rs111033358).^[13] AICAR is 5-aminoimidazole-4-carboxamide ribonucleotide, the AMPK-activating metabolite that MOTS-c causes to accumulate. CB4211 was CohBar's modified analog of MOTS-c; MOTS-MET is the acronym of the 2026 Phase 2a trial of the native peptide (NCT07505745).^[16][19] None of these names refers to SS-31, which is not a mitochondrial-derived peptide.

Frequently asked questions

Related reading

• FOXO4-DRI: senolytic evidence, mechanism and human trials
• 5-Amino-1MQ: what it is, benefits, dosage and research
• SLU-PP-332: exercise mimetic, dosage and research

References