AICAR (Acadesine): How It Works, Human Studies, and Open Questions

AICAR attracted attention because it changed muscle metabolism and improved running performance in mice without exercise training. Human research tells a different, more complicated story: the compound has been studied in heart surgery, leukemia, and short metabolic experiments, but those findings do not establish a safe or effective way to improve fitness. AICAR is not an FDA-approved medicine. This article explains the research rather than providing instructions for using it.

The main findings

• Regulatory status: AICAR has no FDA-approved indication. Acadesine reached phase 3 cardiovascular trials, with development stretching from the 1990s through 2010, and phase I/II oncology research. Neither program produced an approved treatment. There is no FDA-approved prescription product for fitness or metabolic optimization.

• Stage of research: Human studies include cardiac protection, hematologic cancer, and small, short-term intravenous metabolic experiments. The research described here does not establish improved endurance, exercise performance, or lasting metabolic benefits in otherwise healthy people.

• Availability: Products advertised online as research chemicals are not approved medicines for enhancing performance. A sales listing does not establish pharmaceutical quality, reliable strength, sterility, or safety. Research access and routine medical prescribing are different things.

• Prescribing information: There is no FDA-approved prescribing label for AICAR. PubChem provides chemical information about acadesine, listed as CID 17513; a chemical database entry is not a treatment authorization. Source

• Mechanism: Cells convert AICAR into ZMP, an AMP analog that activates AMPK. This cellular energy sensor participates in glucose uptake, fat oxidation, and the formation of mitochondria. Activating it can influence these processes without reproducing every adaptation to exercise.

• What the animal research found: In the widely cited 2008 study by Narkar, Downes, and colleagues in Cell, previously untrained mice receiving AICAR ran about 44% farther. The study implicated the AMPK-PPARδ pathway. Comparable performance benefits have not been established in humans. Source

These findings make AICAR interesting as a research tool. They do not make it an exercise substitute. Animal and cell studies reveal effects on endurance-related signaling and metabolism; cardiac and cancer trials ask different clinical questions. Its prohibition in sport also cannot answer whether it improves performance safely in humans.

What AICAR is

AICAR is short for 5-aminoimidazole-4-carboxamide ribonucleoside. It is also known as acadesine or AICA-riboside. Chemically, it is a small nucleoside analog related to adenosine, not a peptide. The term “AICAR peptide,” sometimes used in searches and product listings, therefore puts it in the wrong chemical category. A peptide consists of an amino acid sequence; AICAR does not.

Winder’s 2000 review in Diabetes Technology and Therapeutics described AICAR as an adenosine analog that activates AMP-activated protein kinase, or AMPK, in skeletal muscle after conversion into an intracellular AMP mimic. This distinction matters because much of the research concerns a signaling pathway rather than an established drug effect in patients. Source

The original clinical purpose was cardiac protection during coronary artery bypass grafting, abbreviated CABG. Researchers thought that regulating adenosine could limit damage when the heart’s blood supply was interrupted and restored. International multicenter studies in the 1990s reported some encouraging findings, but the development program did not ultimately establish a treatment that received FDA approval. Later work on AMPK drew attention to AICAR’s possible effects beyond the heart, especially in skeletal muscle and metabolism.

How AICAR works

Conversion into ZMP: Mimicking an energy signal

After AICAR enters a cell, adenosine kinase phosphorylates it to form AICA-ribonucleotide, also called ZMP. ZMP resembles AMP, one of the molecules involved in cellular energy sensing. Beckers and colleagues’ 2006 study in Molecular Cancer Therapeutics examined how intracellular ZMP accumulation relates to the compound’s metabolic and antiproliferative effects.

AMPK responds to changes in energy availability, including a rising ratio of AMP to ATP. Sustained exercise or inadequate cellular energy can generate that signal. ZMP can imitate part of it without first requiring the ATP depletion that accompanies an actual energy shortage. That is different from saying AICAR never changes cellular energy status: it is a pharmacologic intervention with effects beyond a single switch.

Hardie, Ross, and Hawley’s 2012 review in Nature Reviews Molecular Cell Biology explains how AMPK coordinates responses to nutrient shortage, exercise, and metabolic stress. It changes the activity of downstream proteins involved in glucose uptake, fatty acid oxidation, and mitochondrial biogenesis. Hardie’s 2014 review describes interactions with other signaling networks, while Lin and Hardie’s 2018 article adds the importance of glucose sensing. Together, these accounts explain why the pathway attracts drug research, and why activating it does not predict one uniform response in every tissue. Source Source Source Source

AMPK and skeletal muscle metabolism

Some effects of AMPK activation overlap with adaptations to endurance exercise. In a 2002 study, Sakoda, Ogihara, and colleagues found that AMPK activation was essential for AICAR-stimulated glucose uptake in skeletal muscle. This involves the GLUT4 glucose transporter and differs from the usual insulin signaling route. The result identifies a mechanism; it does not establish that an AMPK activator can safely replace exercise or diabetes treatment.

Leick, Fentz, and colleagues used PGC-1α knockout mice to examine another part of the pathway in 2010. Repeated AICAR treatment increased GLUT4 and cytochrome c protein expression in white muscle from normal mice but not from mice lacking PGC-1α. Hexokinase II, however, responded similarly in both genotypes. PGC-1α was therefore necessary for some of the measured adaptations, rather than for every response to AICAR.

McConell, Ng, and colleagues studied L6 muscle cells in a separate 2010 experiment. Inhibiting nitric oxide synthase reduced AICAR-induced increases in cytochrome c oxidase proteins COX-1 and COX-4 without preventing AMPK phosphorylation. The nitric oxide donor SNAP reproduced the mitochondrial response. This points to another contributor to the process, rather than a simple linear chain in which AMPK alone explains everything.

In 2011, Viscomi, Bottani, Civiletto, and colleagues investigated three mouse models of cytochrome c oxidase deficiency: Surf1 knockout, Sco2 knockout/knockin, and muscle-restricted Cox15 knockout. Activating the AMPK-PGC-1α pathway with AICAR partly corrected COX deficiency. Motor performance in Sco2 KO/KI animals normalized during three weeks of treatment. These were preclinical findings in specific mitochondrial disease models, not evidence of a treatment benefit in patients. They help explain the scientific interest in muscle metabolism while leaving the question of human translation open. Source Source Source Source

The 2008 Narkar study: Exercise effects without training?

Narkar, Downes, Yu, and colleagues at the Salk Institute tested AICAR in untrained C57BL/6J mice. The experimental regimen was 500 mg/kg/day intraperitoneally for four weeks, with 15–20 animals in the endurance comparison. Treated mice ran approximately 23% longer and 44% farther than controls, with statistically significant differences (p < 0.05). These are separate measurements: the 44% figure describes distance, not running time.

The same paper also studied GW501516, called GW1516 in the paper, at 5 mg/kg/day. Combined with exercise training, it increased running time by 68% and distance by 70% relative to trained controls. Those were not endurance results from combining AICAR with GW501516. The two-drug experiments instead examined gene expression. This distinction is essential to interpreting the much-repeated numbers. The work implicated AMPK-PPARδ signaling, but did not test whether these effects occur in people. The animal doses describe the experiment and are not human dosing guidance. Source

Fat metabolism and insulin sensitivity

AICAR has also produced metabolic changes in animal models that do not depend on a treadmill outcome. In a 2002 Diabetes study, Buhl and colleagues treated obese Zucker rats, a model with features of insulin resistance, with subcutaneous AICAR at 0.5 mg/g body weight daily for seven weeks. There were six animals in the AICAR group, six in the control group, and eight lean controls.

Systolic blood pressure fell by 14.6 ± 4.3 mmHg (p < 0.05), and oral glucose tolerance normalized. Triglycerides and free fatty acids decreased compared with controls allowed unrestricted food (p < 0.01), while HDL cholesterol increased (p < 0.01). Skeletal muscle GLUT4 expression and insulin-stimulated glucose transport also improved. These small groups and the particular rat model limit extrapolation to humans, but the findings remain relevant to the biological hypothesis.

Sajan and colleagues’ 2010 experiments in L6 myotubes and muscle-specific atypical protein kinase C knockout mice compared AICAR, metformin, and exercise. AICAR and metformin both increased muscle glucose transport through an AMPK-, ERK-, and PDK1-dependent pathway involving atypical PKC. The same signaling requirement did not apply to treadmill exercise. The distinction was between the drug responses and the exercise response; the study did not establish wholly different downstream cascades for the two drugs.

Hall, Griss, and colleagues reported another difference between AMPK activators in EMBO Molecular Medicine in 2018. AICAR, but not metformin, protected against inflammation-associated muscle wasting in mouse models of C26 cancer cachexia and endotoxin-induced sepsis. The AMPK activator A-769662 reproduced the antiatrophy effect, and Compound C blocked it. Metformin, which can inhibit mitochondrial complex I and favor glycolytic metabolism, did not provide the same protection in these experiments.

These results show why a shared effect on AMPK does not make different compounds interchangeable. Steinberg and Hardie’s 2023 review places such findings within the wider effort to develop AMPK-targeted medicines. The pathway remains promising, while achieving the right effects in the right tissue is a substantial challenge. Source Source Source Source

What human studies have found

Cardiac protection during bypass surgery

The largest clinical program concerned heart surgery. It tested whether regulating adenosine during myocardial ischemia could protect the heart. Bolling and colleagues’ 1992 experiments supplied an early preclinical rationale by finding improved cardiac recovery after ischemia with acadesine.

Leung, Stanley, and colleagues then studied 116 CABG patients in a multicenter, randomized, double-blind trial published in 1994. Patients received placebo, low-dose acadesine, or high-dose acadesine during a seven-hour intravenous infusion, with the compound also added to the cardioplegia solution used during surgery. Before bypass, ischemia detected by transesophageal echocardiography occurred in 6% of the high-dose group, 15% of the low-dose group, and 19% of the placebo group. The difference was not statistically significant (p = 0.22). A mild high-dose increase in uric acid of 1.6 ± 0.2 mg/dL had no reported clinical consequences. These results encouraged further investigation but did not establish the proposed ischemia benefit.

Mangano’s 1997 JAMA meta-analysis pooled five international randomized trials from 81 medical centers. It included 4,043 CABG patients: 2,012 assigned to acadesine and 2,031 to placebo. Reported reductions were 27% for perioperative myocardial infarction (odds ratio 0.69; 95% confidence interval 0.51–0.95; p = 0.02), 50% for early cardiac death (OR 0.52; 95% CI 0.27–0.98; p = 0.04), and 26% for the combined outcome of myocardial infarction, stroke, or cardiac death (OR 0.73; 95% CI 0.57–0.93; p = 0.01). Stroke alone was not significantly reduced. The reported relative reductions and odds ratios are different statistics and should not be treated as interchangeable percentages.

A 2006 follow-up analysis by Mangano, Miao, Tudor, and Dietzel covered 2,698 patients from 54 institutions: 1,352 assigned to acadesine and 1,346 to placebo. The research protocol used 0.1 mg/kg/min intravenously for seven hours plus 5 micrograms/mL in cardioplegia. Among the 100 patients who had a perioperative myocardial infarction, two-year mortality was 27.8% with placebo (15 of 54) and 6.5% with acadesine (3 of 46; p = 0.006). This was a post hoc finding in the infarction subgroup, not the primary outcome for the entire trial.

Nawarskas’ 1999 review and the 2008 development report in Drugs in R&D document the rationale and earlier development history. Neither can report the results of a later trial. That later evidence matters: RED-CABG stopped for futility after enrolling 3,080 of a planned 7,500 patients. The 2012 report found the 28-day composite of death, nonfatal stroke, or mechanical support for severe left ventricular dysfunction in 5.1% of acadesine recipients and 5.0% of placebo recipients (OR 1.01; 95% CI 0.73–1.41; p = 0.94). Earlier positive signals therefore did not translate into benefit on this trial’s primary outcome. The program did not produce an FDA-approved treatment. Source Source Source Source Source Source

The later primary trial report is available at https://pubmed.ncbi.nlm.nih.gov/22782417/.

Research in chronic lymphocytic leukemia

Cancer research began from a different observation. Campàs and colleagues reported in Blood in 2003 that acadesine activated AMPK and induced programmed cell death in B-cell chronic lymphocytic leukemia cells, while sparing T lymphocytes in their experiments. Santidrián and colleagues subsequently found that the apoptotic response involved increased BIM and NOXA proteins and could occur independently of AMPK and p53. AICAR’s actions therefore extend beyond the pathway most often used to explain its metabolic effects.

Van Den Neste, Cazin, Janssens, and colleagues tested acadesine in 24 people with relapsed or refractory CLL in an open-label phase I/II study, published in 2013. Eighteen received single doses from 50 to 315 mg/kg; six received repeated doses of 210 mg/kg. The maximum tolerated dose was 210 mg/kg under the study’s monitoring conditions. These are oncology research doses, not an established regimen for any other use.

Some participants had lower circulating CLL cell counts or smaller enlarged lymph nodes. The findings varied, and the small study was not a definitive efficacy trial. Safety findings included uric acid elevation, gastrointestinal symptoms, transient anemia or thrombocytopenia, renal impairment, and infusion-related hypotension. Table 3 also records a hypoglycemia event; it should not be described as a common event across the program. Allopurinol prophylaxis was introduced after a severe hyperuricemia event. The paper supports an early clinical signal, not an approved leukemia treatment or a general safety claim.

Van Den Neste and colleagues’ 2010 review discusses the hematologic rationale and its limits. Human metabolic research also exists: Boon and colleagues studied a short intravenous infusion in ten men with type 2 diabetes and found changes in hepatic glucose output and lipolysis. Such acute physiological measurements do not establish improved fitness or lasting metabolic optimization. The available research therefore extends beyond heart surgery and cancer, but still leaves the most popular performance claims unanswered. Source Source Source Source Source

What the missing performance trials mean

There remains a gap between the 2008 animal results and evidence of a performance benefit in people. A sport prohibition can affect the practical appeal of developing a compound for athletes, but it does not establish why a clinical research program did or did not continue. Nor does it eliminate the need for properly designed human trials.

There are pharmacologic complications as well. Liu and colleagues’ 2014 PNAS study found that AICAR could arrest the cell cycle independently of AMPK by promoting proteasomal degradation of the G2/M phosphatase CDC25C. The direct AMPK activator A-769662 did not show the same antiproliferative effect. In other words, an effect observed after AICAR administration cannot automatically be attributed to AMPK activation.

Dolinar and colleagues added another limitation in 2018: naturally occurring nucleosides could block AICAR-stimulated AMPK activation in skeletal muscle and cancer cells. The experimental setting can therefore change how the compound behaves. These findings help refine the hypothesis rather than simply disproving it, but they also make it harder to predict a reliable clinical benefit from a mouse endurance result. Source Source

How AICAR compares with other AMPK activators

AICAR is one of several compounds capable of activating AMPK. Sharing a target does not mean sharing an overall mechanism, a clinical benefit, or a safety profile.

Metformin can activate AMPK indirectly through changes in cellular energy metabolism, including inhibition of mitochondrial complex I under studied conditions. AICAR instead supplies ZMP, an AMP mimic, so AMPK activation does not first require the same energy-depletion mechanism. The 2010 Sajan study nevertheless found that both compounds increased muscle glucose transport through the same AMPK-, ERK-, and PDK1-dependent atypical PKC pathway in the tested models. Their upstream actions differ, while some downstream effects overlap. Metformin has extensive human clinical evidence and FDA approval for type 2 diabetes; AICAR does not. Source

SLU-PP-332 is another experimental compound discussed in the search for drugs that reproduce selected exercise responses. It acts through estrogen-related receptors ERRα, ERRβ, and ERRγ, transcription factors involved in oxidative metabolism, rather than by supplying an AMP mimic. Its preclinical research asks related questions through a different molecular route. That is not evidence that it provides a proven alternative to AICAR in humans. Public information about research programs also cannot establish that no confidential investigational application exists.

Reviews by Weihrauch and Handschin in 2018 and Wall, Yu, Atkins, Downes, and Evans in 2016 describe both the appeal and the limits of exercise mimetics. Reproducing a particular molecular response is plausible. Reproducing the coordinated benefits of exercise in a person is a much larger claim. Each intervention needs its own evidence about tissue specificity, effectiveness, and long-term safety; those answers cannot be borrowed from another AMPK activator. Source Source

Side effects and safety questions

Most human safety experience comes from closely monitored cardiac surgery and cancer trials. These populations, routes of administration, treatment durations, and research doses do not establish safety for fitness use.

It helps to distinguish effects actually reported in trials from risks suggested by a mechanism:

• Elevated uric acid was observed clinically and is consistent with AICAR’s nucleoside metabolism. It is particularly relevant to people with hyperuricemia or gout. The CLL research also reported renal impairment and infusion-related hypotension, which should not be overlooked when describing tolerability.

• Low blood glucose is a plausible concern given AICAR’s effects on glucose handling, especially alongside glucose-lowering treatment. The CLL study recorded a hypoglycemia event, but the data do not establish its frequency in healthy users, during fasting, or with insulin or sulfonylureas.

• Gastrointestinal events, including nausea and diarrhea, occurred in the oncology study. Transient blood-count changes, including anemia and thrombocytopenia, were also reported. These observations come from patients with cancer, often with abnormalities already present before treatment.

• AMPK-independent effects on cell-cycle regulation have been observed in preclinical work. The 2014 PNAS paper describes AICAR-related degradation of CDC25C even when the effect is not driven by AMPK. This is a biological reason to avoid treating AMPK activation as a complete account of safety. Source

• Endogenous nucleosides can inhibit AICAR’s AMPK response in experimental systems. That makes a straightforward, predictable dose-response relationship less certain; it does not by itself tell us how often a clinical complication would occur.

Guerrieri, Moon, and van Praag’s 2017 review places AICAR among experimental exercise mimetics and discusses the broader implications of manipulating metabolic signaling. Effects in dividing cells warrant investigation, but mechanistic concern is not proof that AICAR causes cancer in people. Long-term human safety evidence for fitness-related exposure is lacking.

Hawley, Joyner, and Green’s 2021 review emphasizes another limitation: exercise coordinates mechanical, neural, cardiovascular, and metabolic signals. A drug that affects one part of this response cannot be assumed to reproduce the whole process. That is a reason to define the clinical outcome carefully, rather than to dismiss every possible therapeutic application of AMPK research. Source Source

Who should not use AICAR?

There is no approved AICAR treatment for performance or metabolic optimization and no established benefit-risk profile for those purposes. Its legitimate clinical investigation belongs in appropriately authorized research with ethics oversight and monitoring. Material sold as a research chemical adds uncertainty about identity, concentration, purity, and sterility; a label alone cannot resolve those uncertainties.

The following concerns illustrate why medical screening in a research protocol matters. They are not an approved prescribing checklist or a way to determine that self-use would be safe:

• A history of high uric acid or gout: uric acid elevations were reported in cardiac and oncology research, making this an especially relevant concern.

• Susceptibility to hypoglycemia, including treatment with insulin or sulfonylureas: AICAR can influence glucose uptake independently of insulin, while interaction risks outside research conditions remain insufficiently characterized.

• Active or suspected malignancy: the compound’s effects on BIM, NOXA, mTOR, and cell-cycle regulation have not been characterized as safe for unsupervised exposure. Experimental anticancer findings are not a reason to use it as a cancer treatment.

• Pregnancy or breastfeeding: the cited human research does not establish reproductive, developmental, or infant safety.

• Participation in regulated sport: WADA’s 2026 list prohibits AICAR at all times under section S4.4.1, regardless of whether the user believes it will improve performance.

AICAR and anti-doping rules

The 2026 WADA Prohibited List explicitly includes AICAR among AMPK activators in section S4.4.1, within hormone and metabolic modulators. It is prohibited both in and out of competition. The interest in its possible performance effects followed animal research such as the 2008 Narkar study, but that study itself is not a regulatory source and does not prove a human benefit. Source

The current classification can be checked directly in the official list: https://www.wada-ama.org/sites/default/files/2025-09/2026list_en_final_clean_september_2025.pdf.

Testing presents a scientific challenge because AICAR-related compounds also occur naturally in purine metabolism. Detecting a signal is not the same as establishing that someone took an external substance. Piper and colleagues’ 2014 study investigated carbon isotope ratios to distinguish sources of urinary AICAR. Sobolevsky and Ahrens’ 2019 study characterized AICAR and mannitol concentrations in an athlete population to inform reference ranges. A 2022 study by Sobolevsky and colleagues examined the AICAr-to-SAICAr ratio as an additional marker of use.

Thevis and Schänzer’s 2016 review places these developments in the wider challenge of testing compounds that affect muscle function and mitochondrial biogenesis. Wong and colleagues’ 2017 study investigated AICAR in horse urine and plasma after administration, illustrating that this research also extends to equine sport. These papers address analytical detection, not a demonstration of safe or effective athletic use. Source Source Source Source Source

For athletes subject to WADA rules, use or a properly established finding of exogenous AICAR can constitute an anti-doping rule violation. Strict liability means that a lack of intent does not automatically prevent a violation. Individual findings still require interpretation under the applicable testing and results-management rules; the mere presence of an endogenous molecule is not a sufficient description of that process.

Approval status and unanswered research questions

AICAR has no FDA-approved indication. Its largest clinical development effort, cardiac protection during CABG, reached a phase 3 trial that stopped in 2010 after enrolling 3,080 patients. The published RED-CABG results did not show benefit on the primary outcome. The smaller CLL program reached phase I/II research but has not established an approved treatment. Source

An investigational new drug application, or IND, is not a marketing approval, and its existence is not necessarily public. FDA rules generally protect the existence of an IND from disclosure unless it has already been publicly acknowledged. Absence from a public registry therefore cannot prove that no active IND exists. The clinically relevant point is that there is no approved AICAR fitness treatment and that the cited research does not establish a safe, effective performance application.

Whether AICAR itself has a future therapeutic role remains uncertain. Sport restrictions would need consideration in any trial involving regulated athletes; they do not make research in all physically active people impossible. More general challenges include effects outside AMPK, interference by endogenous nucleosides, tissue-specific responses, and defining an acceptable exposure range. The animal experiments do not establish a human therapeutic window.

AMPK remains an active target in drug development. More selective activators may help researchers separate desirable metabolic effects from unwanted actions elsewhere. AICAR continues to be useful for investigating these questions, provided its limitations as a selective AMPK tool are recognized. Its scientific value does not depend on presenting it as an available exercise medicine.

IMPORTANT SAFETY CONTEXT

AICAR is not FDA-approved. There is no established safety profile for self-directed fitness, endurance, or metabolic-optimization use. The experimental regimens described above are included to explain the studies, not to guide use outside research.

Uric acid elevation, kidney impairment, blood-pressure changes during infusion, gastrointestinal events, and blood-count changes were observed in clinical research. Hypoglycemia was also reported, but its frequency cannot be generalized from a small cancer study. Prior gout, susceptibility to low blood glucose, possible hypersensitivity to related compounds, pregnancy, and breastfeeding all raise questions that cannot be resolved by a product description. Athletes also face the all-times prohibition in WADA section S4.4.1.

Long-term effects remain uncertain. Cell experiments identify actions on the cell cycle beyond AMPK, and endogenous nucleosides can alter the expected AMPK response. Unverified research-chemical products add separate quality risks. Neither uncertainty about long-term harm nor a promising mechanism amounts to evidence that unsupervised use is safe.

The available trials do not establish clinical efficacy for fitness. A public trial search also cannot establish the complete IND history, because investigational applications may be confidential. There is no FDA-approved prescribing information for AICAR; PubChem’s chemical record serves a different purpose. Source

Frequently asked questions

What is AICAR used for?

Acadesine was investigated for protecting the heart during coronary bypass surgery and later for treating chronic lymphocytic leukemia. Short human metabolic studies also examined its physiological effects. None of these uses produced an FDA-approved AICAR medicine. In laboratories, it remains a tool for studying AMPK and related pathways. That research role does not establish a prescription treatment for fitness, endurance, or metabolic optimization.

Is AICAR a peptide?

No. It is a small nucleoside analog structurally related to adenosine. A peptide is made from amino acids joined in a chain; AICAR has a different structure. Its appearance alongside peptides in performance or aging discussions does not change its chemistry.

Why is AICAR prohibited in sport?

AICAR is listed as an AMPK activator in WADA section S4.4.1 and is prohibited at all times. Animal research raised interest in its possible effects on endurance, but the ban is not proof that it improves human performance. Researchers have developed isotope-ratio and AICAr-to-SAICAr methods to help identify exogenous use. The official 2026 list is the source for its current status; the papers below address animal findings and detection. Source Source Source

Can you get AICAR on prescription?

There is no FDA-approved AICAR prescription product for fitness, metabolism, or endurance. Access to an investigational compound is distinct from ordinary prescribing and requires the applicable research or regulatory framework. Claims that no IND exists cannot be verified simply by searching a public registry. Online research-chemical sales do not establish medicine-quality manufacturing or a safe clinical use.

What does AICAR do in the body?

Cells convert it into ZMP, which can mimic AMP and activate AMPK. Experimental effects include changes in muscle glucose uptake, fatty acid oxidation, mitochondrial proteins, and energy-consuming pathways. Some overlap with exercise-related signaling, but AICAR also acts outside AMPK. The observed response depends on the tissue and experimental conditions, and does not predict a complete exercise-like benefit in a person.

How does it compare with metformin?

The two compounds can activate AMPK through different upstream mechanisms. Metformin can alter mitochondrial energy metabolism; AICAR supplies an AMP mimic through ZMP. In Sajan and colleagues’ experiments, both increased muscle glucose transport through an AMPK-, ERK-, and PDK1-dependent atypical PKC pathway, unlike exercise in the same models. Metformin has an established evidence base and FDA approval for type 2 diabetes. AICAR has neither an equivalent approval nor interchangeable clinical effects. Source

Is AICAR safe for humans?

Human trials provide some safety information for specific, monitored research settings, particularly cardiac surgery and relapsed or refractory CLL. They do not establish safety for long-term performance use. Reported adverse events and experimental effects outside AMPK make it important to distinguish tolerability within a protocol from general safety. A promising biological mechanism cannot fill the gaps in long-term clinical evidence.

The research remains worth discussing: it has helped clarify how muscle metabolism can change and why translating those changes into a treatment is difficult. The unanswered questions concern particular outcomes, populations, and exposures, rather than a simple judgment that the compound either “works” or “does not work.”

Sources and references

PubChem (National Library of Medicine). (n.d.). Acadesine. https://pubchem.ncbi.nlm.nih.gov/compound/17513 Publication

Narkar VA, Downes M, Yu RT, Embler E, Wang YX, Banayo E, Mihaylova MM, Nelson MC, Zou Y, Juguilon H, Kang H, Shaw RJ, Evans RM (2008). AMPK and PPARdelta agonists are exercise mimetics. Cell, 134(3), 405-15. https://doi.org/10.1016/j.cell.2008.06.051 Publication

Winder WW (2000). AMP-activated protein kinase: possible target for treatment of type 2 diabetes. Diabetes technology & therapeutics, 2(3), 441-8. https://doi.org/10.1089/15209150050194305 Publication

Beckers A, Organe S, Timmermans L, Vanderhoydonc F, Deboel L, Derua R, Waelkens E, Brusselmans K, Verhoeven G, Swinnen JV (2006). Methotrexate enhances the antianabolic and antiproliferative effects of 5-aminoimidazole-4-carboxamide riboside. Molecular cancer therapeutics, 5(9), 2211-7. https://doi.org/10.1158/1535-7163.MCT-06-0001 Publication

Hardie DG, Ross FA, Hawley SA (2012). AMPK: a nutrient and energy sensor that maintains energy homeostasis. Nature reviews. Molecular cell biology, 13(4), 251-62. https://doi.org/10.1038/nrm3311 Publication

Lin SC, Hardie DG (2018). AMPK: Sensing Glucose as well as Cellular Energy Status. Cell metabolism, 27(2), 299-313. https://doi.org/10.1016/j.cmet.2017.10.009 Publication

Hardie DG (2014). AMPK–sensing energy while talking to other signaling pathways. Cell metabolism, 20(6), 939-52. https://doi.org/10.1016/j.cmet.2014.09.013 Publication

Sakoda H, Ogihara T, Anai M, Fujishiro M, Ono H, Onishi Y, Katagiri H, Abe M, Fukushima Y, Shojima N, Inukai K, Kikuchi M, Oka Y, Asano T (2002). Activation of AMPK is essential for AICAR-induced glucose uptake by skeletal muscle but not adipocytes. American journal of physiology. Endocrinology and metabolism, 282(6), E1239-44. https://doi.org/10.1152/ajpendo.00455.2001 Publication

Leick L, Fentz J, Biensø RS, Knudsen JG, Jeppesen J, Kiens B, Wojtaszewski JF, Pilegaard H (2010). PGC-1{alpha} is required for AICAR-induced expression of GLUT4 and mitochondrial proteins in mouse skeletal muscle. American journal of physiology. Endocrinology and metabolism, 299(3), E456-65. https://doi.org/10.1152/ajpendo.00648.2009 Publication

McConell GK, Ng GP, Phillips M, Ruan Z, Macaulay SL, Wadley GD (2010). Central role of nitric oxide synthase in AICAR and caffeine-induced mitochondrial biogenesis in L6 myocytes. Journal of applied physiology (Bethesda, Md. : 1985), 108(3), 589-95. https://doi.org/10.1152/japplphysiol.00377.2009 Publication

Viscomi C, Bottani E, Civiletto G, Cerutti R, Moggio M, Fagiolari G, Schon EA, Lamperti C, Zeviani M (2011). In vivo correction of COX deficiency by activation of the AMPK/PGC-1α axis. Cell metabolism, 14(1), 80-90. https://doi.org/10.1016/j.cmet.2011.04.011 Publication

Buhl ES, Jessen N, Pold R, Ledet T, Flyvbjerg A, Pedersen SB, Pedersen O, Schmitz O, Lund S (2002). Long-term AICAR administration reduces metabolic disturbances and lowers blood pressure in rats displaying features of the insulin resistance syndrome. Diabetes, 51(7), 2199-206. https://doi.org/10.2337/diabetes.51.7.2199 Publication

Sajan MP, Bandyopadhyay G, Miura A, Standaert ML, Nimal S, Longnus SL, Van Obberghen E, Hainault I, Foufelle F, Kahn R, Braun U, Leitges M, Farese RV (2010). AICAR and metformin, but not exercise, increase muscle glucose transport through AMPK-, ERK-, and PDK1-dependent activation of atypical PKC. American journal of physiology. Endocrinology and metabolism, 298(2), E179-92. https://doi.org/10.1152/ajpendo.00392.2009 Publication

Hall DT, Griss T, Ma JF, Sanchez BJ, Sadek J, Tremblay AMK, Mubaid S, Omer A, Ford RJ, Bedard N, Pause A, Wing SS, Di Marco S, Steinberg GR, Jones RG, Gallouzi IE (2018). The AMPK agonist 5-aminoimidazole-4-carboxamide ribonucleotide (AICAR), but not metformin, prevents inflammation-associated cachectic muscle wasting. EMBO molecular medicine, 10(7). https://doi.org/10.15252/emmm.201708307 Publication

Steinberg GR, Hardie DG (2023). New insights into activation and function of the AMPK. Nature reviews. Molecular cell biology, 24(4), 255-272. https://doi.org/10.1038/s41580-022-00547-x Publication

Bolling SF, Groh MA, Mattson AM, Grinage RA, Gallagher KP (1992). Acadesine (AICA-riboside) improves postischemic cardiac recovery. The Annals of thoracic surgery, 54(1), 93-8. https://doi.org/10.1016/0003-4975(92)91148-3 Publication

Leung JM, Stanley T, Mathew J, Curling P, Barash P, Salmenpera M, Reves JG, Hollenberg M, Mangano DT (1994). An initial multicenter, randomized controlled trial on the safety and efficacy of acadesine in patients undergoing coronary artery bypass graft surgery. SPI Research Group. Anesthesia and analgesia, 78(3), 420-34. https://doi.org/10.1213/00000539-199403000-00002 Publication

Mangano DT (1997). Effects of acadesine on myocardial infarction, stroke, and death following surgery. A meta-analysis of the 5 international randomized trials. The Multicenter Study of Perioperative Ischemia (McSPI) Research Group. JAMA, 277(4), 325-32. https://doi.org/10.1001/jama.277.4.325 Publication

Mangano DT, Miao Y, Tudor IC, Dietzel C, Investigators of the Multicenter Study of Perioperative Ischemia (McSPI) Research Group, Ischemia Research and Education Foundation (IREF) (2006). Post-reperfusion myocardial infarction: long-term survival improvement using adenosine regulation with acadesine. Journal of the American College of Cardiology, 48(1), 206-14. https://doi.org/10.1016/j.jacc.2006.04.044 Publication

Nawarskas JJ (1999). Acadesine: a unique cardioprotective agent for myocardial ischemia. Heart disease (Hagerstown, Md.), 1(4), 255-60. https://pubmed.ncbi.nlm.nih.gov/11720632/ Publication

(2008). Acadesine: AICA riboside, ARA 100, arasine, GP 1 110. Drugs in R&D, 9(3), 169-75. https://doi.org/10.2165/00126839-200809030-00004 Publication

Campàs C, Lopez JM, Santidrián AF, Barragán M, Bellosillo B, Colomer D, Gil J (2003). Acadesine activates AMPK and induces apoptosis in B-cell chronic lymphocytic leukemia cells but not in T lymphocytes. Blood, 101(9), 3674-80. https://doi.org/10.1182/blood-2002-07-2339 Publication

Santidrián AF, González-Gironès DM, Iglesias-Serret D, Coll-Mulet L, Cosialls AM, de Frias M, Campàs C, González-Barca E, Alonso E, Labi V, Viollet B, Benito A, Pons G, Villunger A, Gil J (2010). AICAR induces apoptosis independently of AMPK and p53 through up-regulation of the BH3-only proteins BIM and NOXA in chronic lymphocytic leukemia cells. Blood, 116(16), 3023-32. https://doi.org/10.1182/blood-2010-05-283960 Publication

Van Den Neste E, Cazin B, Janssens A, González-Barca E, Terol MJ, Levy V, Pérez de Oteyza J, Zachee P, Saunders A, de Frias M, Campàs C (2013). Acadesine for patients with relapsed/refractory chronic lymphocytic leukemia (CLL): a multicenter phase I/II study. Cancer chemotherapy and pharmacology, 71(3), 581-91. https://doi.org/10.1007/s00280-012-2033-5 Publication

Van Den Neste E, Van den Berghe G, Bontemps F (2010). AICA-riboside (acadesine), an activator of AMP-activated protein kinase with potential for application in hematologic malignancies. Expert opinion on investigational drugs, 19(4), 571-8. https://doi.org/10.1517/13543781003703694 Publication

Liu X, Chhipa RR, Pooya S, Wortman M, Yachyshin S, Chow LM, Kumar A, Zhou X, Sun Y, Quinn B, McPherson C, Warnick RE, Kendler A, Giri S, Poels J, Norga K, Viollet B, Grabowski GA, Dasgupta B (2014). Discrete mechanisms of mTOR and cell cycle regulation by AMPK agonists independent of AMPK. Proceedings of the National Academy of Sciences of the United States of America, 111(4), E435-44. https://doi.org/10.1073/pnas.1311121111 Publication

Dolinar K, Jan V, Pavlin M, Chibalin AV, Pirkmajer S (2018). Nucleosides block AICAR-stimulated activation of AMPK in skeletal muscle and cancer cells. American journal of physiology. Cell physiology, 315(6), C803-C817. https://doi.org/10.1152/ajpcell.00311.2017 Publication

Weihrauch M, Handschin C (2018). Pharmacological targeting of exercise adaptations in skeletal muscle: Benefits and pitfalls. Biochemical pharmacology, 147, 211-220. https://doi.org/10.1016/j.bcp.2017.10.006 Publication

Wall CE, Yu RT, Atkins AR, Downes M, Evans RM (2016). Nuclear receptors and AMPK: can exercise mimetics cure diabetes?. Journal of molecular endocrinology, 57(1), R49-58. https://doi.org/10.1530/JME-16-0073 Publication

Guerrieri D, Moon HY, van Praag H (2017). Exercise in a Pill: The Latest on Exercise-Mimetics. Brain plasticity (Amsterdam, Netherlands), 2(2), 153-169. https://doi.org/10.3233/BPL-160043 Publication

Hawley JA, Joyner MJ, Green DJ (2021). Mimicking exercise: what matters most and where to next?. The Journal of physiology, 599(3), 791-802. https://doi.org/10.1113/JP278761 Publication

Piper T, Thomas A, Baume N, Sobolevsky T, Saugy M, Rodchenkov G, Schänzer W, Thevis M (2014). Determination of ¹³C/¹² C ratios of endogenous urinary 5-amino-imidazole-4-carboxamide 1β-D-ribofuranoside (AICAR). Rapid communications in mass spectrometry : RCM, 28(11), 1194-202. https://doi.org/10.1002/rcm.6891 Publication

Sobolevsky T, Ahrens B (2019). Urinary concentrations of AICAR and mannitol in athlete population. Drug testing and analysis, 11(3), 530-535. https://doi.org/10.1002/dta.2557 Publication

Sobolevsky T, Piper T, Ahrens B, Thevis M (2022). AICAr to SAICAr ratio can serve as additional marker of AICAr use. Drug testing and analysis, 14(11-12), 2017-2025. https://doi.org/10.1002/dta.3399 Publication

Thevis M, Schänzer W (2016). Emerging drugs affecting skeletal muscle function and mitochondrial biogenesis - Potential implications for sports drug testing programs. Rapid communications in mass spectrometry : RCM, 30(5), 635-51. https://doi.org/10.1002/rcm.7470 Publication

Wong JKY, Kwok WH, Chan GHM, Choi TLS, Ho ENM, Jaubert M, Bailly-Chouriberry L, Bonnaire Y, Cawley A, Ming Williams H, Keledjian J, Brooks L, Chambers A, Lin Y, Wan TSM (2017). Doping control study of AICAR in post-race urine and plasma samples from horses. Drug testing and analysis, 9(9), 1363-1371. https://doi.org/10.1002/dta.2205 Publication

Boon H, Bosselaar M, Praet SF, Blaak EE, Saris WH, Wagenmakers AJ, McGee SL, Tack CJ, Smits P, Hargreaves M, van Loon LJ (2008). Intravenous AICAR administration reduces hepatic glucose output and inhibits whole body lipolysis in type 2 diabetic patients. Diabetologia, 51(10), 1893-900. https://doi.org/10.1007/s00125-008-1108-7 Publication

ClinicalTrials.gov. (n.d.). The effect of acadesine on clinically significant adverse cardiovascular and cerebrovascular events in high-risk subjects undergoing coronary artery bypass graft (CABG) surgery using cardiopulmonary bypass: RED-CABG trial (NCT00872001). https://clinicaltrials.gov/study/NCT00872001 Publication

Newman MF et al. (2012). Effect of adenosine-regulating agent acadesine on morbidity and mortality associated with coronary artery bypass grafting: the RED-CABG randomized controlled trial. JAMA, 308(2), 157–164. https://pubmed.ncbi.nlm.nih.gov/22782417/ https://doi.org/10.1001/jama.2012.7633

World Anti-Doping Agency. (2025). The 2026 Prohibited List. Section S4.4.1. https://www.wada-ama.org/sites/default/files/2025-09/2026list_en_final_clean_september_2025.pdf

US Food and Drug Administration. New Drug, Antibiotic, and Biologic Drug Product Regulations, Part 4. Section 312.130, availability for public disclosure of data and information in an IND. https://www.fda.gov/science-research/clinical-trials-and-human-subject-protection/new-drug-antibiotic-and-biologic-drug-product-regulations-part-4

Original-source citation correction: PubChem CID 378611 identifies Cu-GHK, not acadesine. The original URL is retained here to document the corrected mismatch; it does not support the AICAR claims. https://pubchem.ncbi.nlm.nih.gov/compound/378611

US Food and Drug Administration. Search Orphan Drug Designations and Approvals: acadesine, chronic lymphocytic leukemia. Designated March 3, 2011; not FDA approved for orphan indication. https://www.accessdata.fda.gov/scripts/opdlisting/oopd/detailedIndex.cfm?cfgridkey=203105