ATX-304

The metabolic effects of acute exercise are significantly mediated by AMPK: during exercise, working muscle consumes ATP rapidly, raising AMP:ATP ratio, activating AMPK; active AMPK promotes: GLUT4 translocation (glucose uptake); fatty acid oxidation; mitochondrial biogenesis via PGC-1α; anti-inflammatory cytokine profiles. ATX-304 pharmacologically activates AMPK through a non-exercise mechanism, producing many of the same downstream signaling events. The animal evidence supports this: ATX-304 increased cardiac glucose uptake, improved stroke volume, increased peripheral blood flow, and enhanced exercise endurance in mice — an exercise-like cardiovascular and metabolic profile achieved without requiring exercise.

Exercise is a complex physiological intervention that AMPK activation alone does not fully replicate. Key exercise-associated effects that are partially or wholly independent of AMPK: Structural adaptation — exercise causes muscle protein synthesis (via mTOR, which is actually inhibited by AMPK), hypertrophy, and connective tissue strengthening; ATX-304's AMPK activation would be expected to suppress mTOR-driven muscle anabolism, not support it. Neurological adaptations — exercise improves coordination, proprioception, and neural drive; these are entirely outside ATX-304's peripherally restricted mechanism. Bone density — exercise (especially weight-bearing) increases bone mineral density through mechanical loading; pharmacological AMPK activation does not replicate this. Psychological and cognitive effects — exercise-induced endorphin/endocannabinoid release, neuroplasticity, BDNF upregulation; ATX-304 is peripherally restricted and would not produce these CNS effects. Hormonal cascade — acute exercise produces a complex hormonal response (catecholamines, GH, testosterone, cortisol) that ATX-304 does not replicate. The honest framing: ATX-304 mimics the metabolic signaling of exercise (AMPK activation) and adds mitochondrial uncoupling to increase energy expenditure — producing significant metabolic benefits. It does not replicate the physical, neurological, skeletal, or hormonal dimensions of exercise. It complements exercise rather than replacing it.

ATX-304 (formerly O-304) is a synthetic small molecule with two chlorine (halogen) substituents. The specific chemical structure has been characterized in published papers (including the Pan-AMPK activator paper in JCI/PMC). It is a peripherally restricted compound — its molecular properties limit CNS penetration, which was a deliberate design goal to avoid the CNS side effects seen with earlier AMPK activators. Molecular properties allow oral bioavailability — demonstrated in clinical trials with a 1,000 mg/day oral dose. The two chlorine substituents influence: lipophilicity (affecting tissue distribution); halogen bonding with target protein residues (potentially enhancing binding to AMPK complex); and metabolic stability.

Understanding ATX-304 requires understanding AMPK deeply. AMP-activated protein kinase (AMPK) is a heterotrimeric enzyme consisting of a catalytic α subunit (α1 or α2) and regulatory β (β1 or β2) and γ (γ1, γ2, or γ3) subunits — giving rise to 12 possible isoform combinations. The kinase is activated when the cellular energy charge falls: rising AMP and falling ATP concentrations promote AMP binding to the γ subunit, which prevents dephosphorylation of the critical threonine 172 (Thr172) residue on the α subunit. Phospho-Thr172 AMPK is the active form. AMPK activation is the biochemical signature of: exercise (energy-consuming muscle contraction increases AMP); caloric restriction (reduced substrate availability raises AMP:ATP ratio); ischemia/hypoxia (impaired ATP synthesis); metformin (complex I inhibition raises AMP:ATP). The downstream signaling of active AMPK: in the liver — phosphorylates and inactivates Acetyl-CoA Carboxylase (ACC) → reduces de novo lipogenesis and cholesterol synthesis → less hepatic fat accumulation; in skeletal muscle — promotes GLUT4 translocation → increased insulin-independent glucose uptake; activates PGC-1α → mitochondrial biogenesis; in the cardiovascular system — reduces cardiac inflammation; improves endothelial function; increases VEGF expression and blood flow. The beneficial metabolic effects of AMPK activation collectively explain why exercise and caloric restriction are protective against T2D, obesity, cardiovascular disease, and cancer — AMPK is a master switch that shifts the cell from growth/storage mode to maintenance/repair/oxidation mode.

ATX-304 activates AMPK by a mechanism distinct from both metformin and AMP: it suppresses the dephosphorylation of AMPK Thr172. While AMP binding promotes phosphorylation by LKB1 and CaMKK2, ATX-304 acts downstream — keeping Thr172 in its phosphorylated (active) state by inhibiting the phosphatases that would normally terminate AMPK activity. This mechanism means ATX-304 does not require cellular energy deficit to activate AMPK — it maintains AMPK in its active state without elevating AMP or impeding ATP synthesis. This is mechanistically different from: Metformin (inhibits mitochondrial complex I → raises AMP → AMPK activation as a consequence of energy depletion); AICAR (a synthetic AMP analogue — directly mimics AMP to activate AMPK); Earlier compounds that targeted the AMP binding site or kinase domain directly. 'Pan-AMPK' means ATX-304 activates all AMPK isoforms (α1 and α2 containing complexes), not just selected subtypes.

THE DUAL MECHANISM — AMPK SIGNALING + MITOCHONDRIAL UNCOUPLING

ATX-304 has two documented mechanisms that are potentially independent: MECHANISM 1 — AMPK ACTIVATION: Suppresses AMPK Thr172 dephosphorylation → keeps AMPK active → downstream metabolic reprogramming as described above. This is the 'signaling' mechanism. MECHANISM 2 — MITOCHONDRIAL UNCOUPLING: ATX-304 increases basal oxygen consumption rate (OCR) in C2C12 myotubes (skeletal muscle cell model), which is the signature of mitochondrial uncoupling. Uncoupling means: protons pumped by the electron transport chain across the inner mitochondrial membrane (IMM) are allowed to 'leak' back across the IMM without passing through ATP synthase. This dissipates the proton gradient as heat instead of ATP. The consequence: more oxygen is consumed and more fuel (glucose, fat) is burned without additional ATP production — a 'futile cycle' that increases total energy expenditure. This is the mechanism by which mild mitochondrial uncouplers (e.g., DNP — dinitrophenol, which is dangerously toxic) cause weight loss: they make metabolism less efficient, requiring more fuel to maintain the same ATP output. ATX-304's uncoupling appears to be mild and occurs at therapeutic concentrations without the temperature-dysregulation that makes DNP dangerous. The metabolic demand created by uncoupling further activates AMPK (via increased AMP) — meaning the two mechanisms may be partially synergistic: uncoupling raises AMP → additional AMPK activation → further metabolic reprogramming. Recent data from JCI Insight (2025) suggests the uncoupling mechanism may contribute to the observed weight loss and hepatoprotective effects in MASLD mouse models. This dual mechanism is why Amplifier Therapeutics describes ATX-304 as a 'AMPK and mitochondrial activator.'

TELLUS study: Lundsgaard et al. and the Betagenon clinical team; randomized, double-blind, placebo-controlled, single-center, parallel-group, 28-day Phase 2a trial. Published in JCI/PMC (PMC6124394, 2018). Population: 65 T2D patients on stable metformin; HbA1c 6.5-9.0% at screening; exclusion for severe cardiovascular events. Treatment: O-304 (ATX-304) 1,000 mg/day vs placebo for 28 days. PRIMARY OUTCOME — FPG: In the clinically relevant FPG range (>7 to <13.3 mmol/L at Day 1; n=49 in this post-hoc subgroup): O-304 group mean FPG reduction = -0.60 mM; placebo = -0.10 mM; p = 0.0096. Absolute difference: -0.50 mM. This is a clinically meaningful glycemic effect. SECONDARY OUTCOME — MICROVASCULAR PERFUSION: MRI calf muscle perfusion showed improvement with O-304 vs placebo — reduced peripheral vascular resistance; improved blood flow to exercising muscle. This cardiovascular/vascular finding is consistent with the animal model cardiovascular data. SAFETY: well-tolerated; no significant adverse effects reported; pharmacokinetics as expected. LIMITATIONS: 28-day duration (no HbA1c change expected in this timeframe); the primary FPG analysis required a post-hoc protocol amendment (patients included based on HbA1c at screening, not FPG at Day 1); small sample size per subgroup. Grade B — DBRCT; positive results; post-hoc primary endpoint analysis is a meaningful limitation.

ATX-304 has been evaluated in multiple well-characterized animal models with consistent results across independent research groups. Diet-induced obesity (DIO) mice: ATX-304 reduced body fat mass, improved glucose homeostasis, reduced hyperlipidemia, protected against diet-induced metabolic deterioration. T2D mouse models: improved insulin sensitivity; stimulated insulin-independent glucose uptake in skeletal muscle; counteracted diabetic cardiomyopathy; reduced cardiac glycogen accumulation; improved left ventricular stroke volume. MASLD mouse model (JCI Insight 2025, Hörnblad et al.): choline-deficient high-fat diet mouse model of progressive MASLD/MASH; ATX-304 reduced body fat mass, lowered blood cholesterol, reduced hepatic steatosis and liver fibrosis; proposed mechanism: mitochondrial uncoupling driving increased hepatic fat oxidation. Cisplatin-induced acute kidney injury (AKI) model (ScienceDirect 2024): ATX-304 pre-treatment protected against cisplatin-mediated kidney injury in mice and primary kidney tubular epithelial cell cultures; mechanism: AMPK activation + metabolic uncoupling reducing energy stress. Cardiovascular function: improved cardiac glucose uptake; reduced diabetic cardiomyopathy; improved stroke volume and cardiac output; increased peripheral blood flow and exercise endurance.

Amplifier Therapeutics presented preclinical data at ENDO 2025 (July 2025) investigating ATX-304 in combination with semaglutide in a diet-induced obesity mouse model. Key findings presented: ATX-304 monotherapy produced body weight reduction and body composition changes; ATX-304 + semaglutide combination showed greater effects than either alone; ATX-304 after semaglutide withdrawal reduced weight rebound compared to semaglutide withdrawal without ATX-304. This last finding is particularly clinically interesting: GLP-1 medications produce substantial weight loss during treatment, but weight regain after discontinuation is common. ATX-304's post-GLP-1 maintenance of weight loss — if it translates to humans — would address one of the most significant clinical challenges in GLP-1 therapy. Grade C — DIO mouse model; ENDO 2025 conference presentation (not yet published in full peer-reviewed form at time of this chapter's writing).

Application

Grade

Best Evidence

Community Relevance

FPG reduction in T2D on metformin (TELLUS)

B

DBRCT, n=65 (49 in analysis), 28 days, -0.60 vs -0.10 mM, p=0.0096; microvascular perfusion improvement by MRI

The strongest and only human efficacy data; T2D population

AMPK activation (cellular biomarker)

A

ACC phosphorylation confirmed in multiple cell and animal models; Thr172 dephosphorylation suppression mechanism characterized

Mechanistic foundation; applies to all downstream effects

Metabolic/fat mass reduction (animal)

B

DIO mouse models; MASLD models (JCI Insight 2025); consistent results across multiple research groups

Basis for community fat loss claims; animal only

MASLD/liver protection (animal)

B

JCI Insight 2025 (PMC11981618); reduced steatosis and fibrosis in CD-HFD mice

Most recent and rigorous preclinical data; human MASLD not studied yet

Cardiovascular function (animal)

B

Improved stroke volume, cardiac output, peripheral blood flow; counteracted diabetic cardiomyopathy in mice; supports TELLUS MRI findings

Consistent with TELLUS microvascular finding; human cardiovascular outcomes not established

AKI protection (animal)

B

ScienceDirect 2024; cisplatin AKI model in mice and cell cultures

Emerging application; no human kidney data

GLP-1 combination + rebound prevention

C

ENDO 2025 preclinical DIO mouse presentation; conference data not yet published in full

Potentially clinically important; human data required

Healthy adult fat loss / longevity

E

No human controlled trial in this population

The primary community application; no controlled evidence

Phase

Year

Sponsor

Outcome

Key Finding

Preclinical (animal)

2010-2015

Betagenon AB

Positive across multiple DIO, T2D, cardiovascular models

Exercise mimetic profile established; selection for clinical development

Phase 1 (healthy + T2D)

2015-2016

Betagenon AB

Completed; positive

Safe and well-tolerated in young healthy, overweight middle-aged, and T2D on metformin subjects; pharmacokinetics characterized

Phase 2a TELLUS

2016-2017

Betagenon AB / Baltic Bio

Completed; positive (post-hoc analysis)

Significant FPG reduction in T2D subgroup (-0.60 vs -0.10 mM, p=0.0096); microvascular perfusion improvement; well-tolerated

Acquisition by Cambrian/Amplifier

2023

Cambrian Bio / Amplifier Therapeutics

$26M acquisition; $33.25M Series A

RA Capital and Future Ventures backing; compound renamed ATX-304; dedicated obesity/metabolic disease focus

Phase 1b

2023-2024 (ongoing)

Amplifier Therapeutics

First patient dosed; ongoing

Safety and pharmacology in updated dosing regimen; results pending

ENDO 2025 preclinical

July 2025

Amplifier Therapeutics

Conference presentation

DIO mouse data: ATX-304 alone, + semaglutide, and after semaglutide withdrawal presented; post-GLP-1 weight rebound reduction data

Phase 2 obesity

Planned/ongoing 2025-2026

Amplifier Therapeutics

Not yet completed

Primary Phase 2 obesity human trial; key data expected in next 12-18 months

THE DOSE GAP — CRITICAL FOR COMMUNITY USERS

The TELLUS Phase 2a trial used 1,000 mg/day of ATX-304 (O-304). Community research chemical vendors typically offer 100-300 mg capsules, with community protocols citing 100-200 mg/day. The pharmaceutical trial dose is 3-10x the most common community dose. WHAT THIS MEANS: (1) The human clinical data (TELLUS FPG reduction, microvascular improvement) was generated at 1,000 mg/day. The extent to which 100-200 mg/day produces comparable pharmacological effects in humans is unknown — the dose-response relationship at sub-pharmaceutical doses is not characterized. (2) AMPK activation in human tissues at 100-200 mg/day has not been measured. Animal studies used doses scaled to animal body weight; the translation to human pharmacology at community doses is uncertain. (3) The compound is orally bioavailable and safe at 1,000 mg/day in humans — so 100-300 mg/day is likely pharmacologically active (sub-therapeutic rather than inactive), but whether it produces the metabolic effects observed in trials is not established. Community users should understand they are using a pharmacologically active compound at a dose range not directly studied in controlled human trials.

ATX-304 has been administered to humans in two clinical trial programs: Phase 1 (young healthy volunteers, overweight middle-aged volunteers, T2D on metformin) and Phase 2a TELLUS (T2D on metformin, 28 days, 1,000 mg/day). Results across both programs: no significant adverse effects attributed to ATX-304; well-tolerated at 1,000 mg/day oral doses; pharmacokinetics were as expected; no organ toxicity signals. The Phase 1b study is currently ongoing, which means additional safety data will emerge over the next 1-2 years. Long-term safety (beyond 28 days in humans) is not yet characterized from clinical trial data.

Mitochondrial uncoupling has a complex safety history. 2,4-Dinitrophenol (DNP), a potent mitochondrial uncoupler used for weight loss in the 1930s, caused multiple deaths from hyperthermia — because DNP's uncoupling was powerful enough to raise core body temperature to lethal levels. The safety concern from mild mitochondrial uncoupling: dose-dependent temperature dysregulation. ATX-304 appears to produce mild uncoupling at therapeutic doses, sufficient to increase energy expenditure modestly without causing problematic temperature elevation. In DIO mice, ATX-304-treated animals consumed more food (compensating for the energy expenditure from uncoupling) but maintained their low fat mass. In clinical trials, no hyperthermia adverse effects were reported. The safety profile of mild vs potent uncoupling differs substantially. The ongoing Phase 1b clinical trial will provide more definitive safety data at the doses being advanced toward obesity indications.

ATX-304 is not currently listed on the WADA Prohibited List as of 2025-2026. However, WADA's prohibited list includes substances and methods that are relevant to the AMPK pathway — most directly: GW501516 (Cardarine/GSK-516), another AMPK-related metabolic activator, was added to the WADA prohibited list and is banned across most sports. AMPK activators as a class have been subjects of WADA surveillance discussion. ATX-304 may be subject to future WADA categorization as a metabolic modulator as its pharmacological profile becomes better characterized. For competitive athletes: the absence from the current WADA list should not be interpreted as a permanent clear — WADA regularly updates the list based on doping risk assessments. Performance effects of AMPK activation (enhanced endurance via fatty acid oxidation and glucose uptake, improved cardiovascular efficiency) are consistent with doping concerns.

A specific naming confusion exists in community research chemical vendor space: some vendors sell 'ATX-304 (OS-01)' or note that ATX-304 is 'sometimes referred to as ATX-304 (OS-01) online.' OS-01 is a completely different compound — a topical peptide researched for preventing accumulation of senescent cells in skin (related to the OneSkin company's research program). ATX-304 (O-304, pan-AMPK activator) and OS-01 (topical senolytic peptide) share nothing except appearing in the community chemical catalog space. Vendors who list them as the same compound or use the combined designation are creating confusion. Always confirm the compound identity (CAS number, mechanism of action, AMPK activation context) when purchasing research chemicals.

Partially valid for the metabolic signaling dimension of exercise; significantly overstated as a complete characterization. ATX-304 activates AMPK (which exercise also activates) and produces mitochondrial uncoupling (increasing energy expenditure) — mimicking two important metabolic mechanisms of exercise. It does not replicate the structural, neurological, hormonal, or skeletal adaptations of physical exercise. It does not build muscle mass (AMPK actually inhibits mTOR-driven muscle anabolism). It should be understood as a metabolic signaling activator and energy expenditure enhancer — not as a replacement for physical exercise.

TELLUS proved ATX-304 produces significant fasting glucose reduction in T2D patients on metformin over 28 days. Body composition and fat loss were not the primary endpoints of TELLUS; the study was 28 days in duration (too short for significant fat mass changes); and the population was T2D patients on metformin, not healthy adults seeking fat loss. The animal evidence for fat mass reduction is strong. Human fat loss evidence is from DIO mouse models. No human controlled trial has established fat loss from ATX-304 in any population.

There is no basis for assuming equivalence across a 5-10x dose difference. The pharmacological dose-response of ATX-304 at sub-trial doses in humans is not characterized. At lower doses, AMPK activation and uncoupling may occur to a lesser degree, requiring longer duration or producing more modest effects. The community is not wrong to use lower doses (which may be both safer and more accessible), but should not claim equivalence with the clinical trial evidence at those doses.

ATX-304 is not a peptide. It is a synthetic small molecule organic compound — a halogenated substance with two chlorine substituents. It is not an amino acid sequence, not a protein fragment, not a hormone analog. It is included in this peptide book because of its widespread community use alongside peptides in the metabolic and longevity protocol space. Users who think of ATX-304 as a peptide may have incorrect expectations about storage, administration (oral capsule only, not injectable), and mechanism.

Both activate AMPK-related metabolic pathways and are classified as 'exercise mimetics' in community discussions. They are mechanistically different: Cardarine (GW501516) is a PPARδ agonist — activates peroxisome proliferator-activated receptor delta, which drives fatty acid oxidation gene programs in muscle. ATX-304 directly activates AMPK and produces mitochondrial uncoupling. The downstream metabolic effects overlap significantly but the mechanisms differ. Cardarine has been abandoned by its pharmaceutical developer (GSK) due to cancer concerns in animal studies (see the Cardarine chapter). ATX-304 does not share Cardarine's cancer signal and has active pharmaceutical development. They are not pharmacological equivalents.

Liivak N, Ingman T, Bie P, Ekblom B, Edlund T, et al. (Pan-AMPK activator O304 improves glucose homeostasis and microvascular perfusion in mice and type 2 diabetes patients). JCI/PMC. PMC6124394. [The TELLUS Phase 2a trial; n=65 T2D; 28 days; significant FPG reduction in post-hoc subgroup; microvascular perfusion improvement by MRI; safety confirmed. The definitive human evidence paper.]

Hörnblad A, Holm E, Vermeulen I et al. (2025). AMPK activator ATX-304 reduces oxidative stress and improves MASLD via metabolic switching. JCI Insight. 10(7):e179990. PMC11981618. [Choline-deficient HFD mouse model of MASLD/MASH; ATX-304 reduced fat mass, cholesterol, hepatic steatosis and fibrosis; mitochondrial uncoupling mechanism proposed for liver benefits.]

AMPK activator ATX-304 alters cellular metabolism to protect against cisplatin-induced acute kidney injury. ScienceDirect. 2024. [Cisplatin AKI in mice and primary kidney tubular cell cultures; AMPK activation and uncoupling reduce energy stress and protect against injury.]

Amplifier Therapeutics / Cambrian Bio. Press releases 2023-2025. Phase 1b dosing initiation; $33.25M Series A; ENDO 2025 presentation of DIO mouse combination data with semaglutide. [Current development status documentation.]

Betagenon AB / Baltic Bio. (November 2, 2017). Positive results from 28-day Phase IIa trial of first-in-class AMPK activator O304 in type 2 diabetics. PR Newswire. [The TELLUS Phase 2a top-line results press release; -0.60 mM vs -0.10 mM FPG, p=0.0096.]

• T2D on metformin: the TELLUS data supports ATX-304 as an adjunct for glucose management — the most evidence-supported human application. Not a substitute for established T2D pharmacotherapy; discuss with physician.
• Metabolic syndrome, obesity: animal evidence is strong; Phase 2 human trial is ongoing and will clarify this. Community use before Phase 2 results represents using an active pharmaceutical development compound ahead of its efficacy proof in the target indication.
• GLP-1 combination or post-GLP-1 weight maintenance: ENDO 2025 mouse data is compelling; no human trial data yet; watch for Phase 2 results.
• Healthy adult longevity/metabolic optimization: Grade E — mechanistically plausible (AMPK activation and longevity are linked in multiple model organisms); no controlled human evidence in this population.
• As substitute for exercise: a useful metabolic complement, not a replacement; ATX-304 does not build muscle, improve bone density, produce neurological or hormonal adaptations, or replicate the full physiology of physical training.
• WADA/competitive sports: not currently prohibited; AMPK activators as a class are under surveillance; future listing is possible; monitor WADA updates.

— End of ATX-304 —

THE PEPTIDE BIBLE | ATX-304 | For Research & Educational Purposes Only