AICAR
Acadesine
AICAR (Acadesine) is a metabolic & fat loss research compound. AMPK activator — mimics the low-energy signal of exercise, driving glucose uptake and fatty-acid oxidation.
AICAR quick facts
| Reported research dose | 5mg-25mg |
| Route | Subq |
| Frequency | 1x Daily AM · 5 On 2 Off or Daily |
| Half-life | ~30-45 min |
| Forms | Injectable |
| Evidence level | Animal + research |
Endurance/metabolic tool. Effective doses in studies are large — be realistic.
How AICAR works
AMPK activator — mimics the low-energy signal of exercise, driving glucose uptake and fatty-acid oxidation.
Proposed benefits
Exercise-mimetic AMPK activation for endurance and glucose handling.
Where to get AICAR
Buy AICAR at Ion Peptide →Bacteriostatic water is the diluent — sterile water with 0.9% benzyl alcohol, which is what lets a vial be drawn from more than once. It does not come with the vial, and unlike the compound it is bought again every time.
Need bacteriostatic water? Get it at AminoWell USA (my company) → Code CAMERON.
The evidence for AICAR
Graded by what exists behind each claim.
✅ Clinically validated
- Human studies exist and are metabolic, not performance. It has been given intravenously in clinical research settings to study AMPK activation and glucose uptake, and was investigated for cardiac protection during bypass surgery.
- No trial supports the endurance use it is sold for, and it is on the WADA prohibited list — which is the only reason most people have heard of it.
📊 Correlative data
- Known in sport as the 'exercise in a pill' compound after the 2008 Salk Institute mouse work. Practical use is limited by pharmacokinetics rather than legality: oral bioavailability is poor and the effective doses in the animal work were very large, which makes the real-world versions of this mostly wishful.
- Beyond that the record is self-reported: community dosing logs are real information about tolerability and almost none about efficacy.
🧪 Theoretical / extrapolated
- An adenosine analog converted intracellularly to ZMP, which mimics AMP and directly activates AMPK — the cell's low-energy sensor. Activating AMPK switches on glucose uptake, fat oxidation and mitochondrial biogenesis: the metabolic signature of exercise.
- AMPK activation is genuinely central and genuinely non-specific. It is downstream of exercise, metformin, and calorie restriction — and it also opposes mTOR, meaning sustained activation suppresses muscle protein synthesis. Anyone using it for performance is running two mechanisms against each other.
What community dosing logs are worth →
How to read these tiers: they say how much human evidence exists, not how well something works — and ✗ flags harm, never a disappointing trial. How the evidence tiers work →
What AICAR actually does
AICAR does nothing by itself. It is a prodrug for a molecule that looks like AMP. AICA riboside is a nucleoside; it enters cells on nucleoside transporters and is phosphorylated by adenosine kinase to ZMP — AICA ribotide, the monophosphate. ZMP is a structural mimic of 5′-AMP, and AMP is the signal that tells a cell it is short of energy.
What ZMP binds. AMP-activated protein kinase is a heterotrimer whose gamma subunit carries the nucleotide-binding sites. AMP occupancy there both allosterically activates the kinase and protects its activation loop from dephosphorylation. ZMP does the same thing, less well. That last clause is the entire pharmacological problem with this compound: ZMP is a weaker agonist at the gamma subunit than AMP is, so producing an effect requires driving intracellular ZMP to concentrations far above anything physiological — which in turn requires extracellular AICAR concentrations that are not achievable with a small injection.
What activated AMPK then does, in named steps. Merrill 1997 traced the whole chain in perfused rat muscle: AICAR activated AMPK, which phosphorylated and inactivated acetyl-CoA carboxylase, which lowered malonyl-CoA. Malonyl-CoA is the physiological inhibitor of carnitine palmitoyltransferase-1, the gatekeeper for long-chain fatty acids entering the mitochondrion. Remove the inhibitor and fat oxidation rises. In that experiment it rose 2.8-fold, alongside a significant increase in glucose uptake, with no difference in oxygen uptake between AICAR and control hindlimbs. That is a complete, named, four-enzyme mechanism, and it is why the compound is interesting at all.
Now the part that undermines the branding, and it is measured in humans. Babraj 2009 infused AICAR into people and measured muscle glucose uptake alongside the signaling. Glucose uptake rose — and total AMPK-alpha activity and the phosphorylation of AMPK, acetyl-CoA carboxylase and AS160 were all unchanged. What did change was ERK1/2 phosphorylation, and the fold-changes in ERK1/2 and in glucose uptake correlated closely, R² = 0.55, P = 0.003. In human muscle, at the doses that have been given, the glucose effect of AICAR traveled with a MAP-kinase signal and not with the kinase the compound is named after.
And ZMP is not a clean tool. AMP regulates several enzymes besides AMPK — fructose-1,6-bisphosphatase and glycogen phosphorylase among them — and an AMP mimic accumulating to millimolar concentrations engages them too. Bosselaar 2011 measured AICAR and ZMP in plasma and in erythrocytes during a human infusion and found AICAR was rapidly taken up by erythrocytes and phosphorylated to ZMP. Red cells have no mitochondria and no reason to be the target; they are a large, avid sink that converts the prodrug where it can do nothing useful.
Cell, rodent, human — and where it stops
Step one, perfused rat hindlimb, and note the concentration. Merrill 1997 perfused rat hindlimbs with Krebs-Henseleit buffer containing 4% albumin, washed bovine red cells, 200 µU/mL insulin and 10 mM glucose, with or without AICAR at 0.5 to 2.0 mM. At 2 mM for 45 minutes: AMPK activated, acetyl-CoA carboxylase inactivated, malonyl-CoA down, fat oxidation up 2.8-fold. 2 millimolar is the number to carry forward. It is the concentration bathing the tissue, held constant by a pump.
Step two, mice, and this is the paper that created the market. Narkar 2008 tested whether an orally active AMPK agonist could substitute for training. In sedentary mice, 4 weeks of AICAR treatment alone induced metabolic genes and enhanced running endurance by 44%. The title — “AMPK and PPAR-delta agonists are exercise mimetics” — is the source of the phrase every vendor uses. The paper is real and the result is real. It is a mouse, dosed daily for a month.
Step three, human muscle in a dish. Koistinen 2003 showed AICAR increased glucose transport and cell-surface GLUT4 content in skeletal muscle taken from people with type 2 diabetes — genuine human tissue, and a mechanism-level result in isolated muscle rather than in a person.
Step four, whole humans, by intravenous infusion, and the two trials disagree in an instructive way. Babraj 2009 gave 10 or 20 mg/kg/hour intravenously from hour 3 to hour 6 of a 6-hour 2-deoxyglucose infusion, with quadriceps biopsies at 0, 3 and 6 hours, in six young men, eight older men and eight men with type 2 diabetes. At 10 mg/kg/h glucose uptake rose 2.9 ± 0.7-fold in young men (P < 0.001), 1.8 ± 0.2-fold in older men (P < 0.01) and 1.6 ± 0.1-fold in men with type 2 diabetes; at 20 mg/kg/h, 2.5-fold and 2.2-fold in the two older groups. The authors' conclusion is the interesting one: the blunting tracked age rather than diabetic status.
Step five, the trial that found nothing metabolic. Bosselaar 2011 gave ten healthy participants two euglycemic hyperinsulinemic clamps, with AICAR 67 mg/kg or saline infused over the last 60 minutes. Forearm glucose uptake and whole-body glucose infusion rate did not differ between AICAR and placebo. What AICAR did do was raise heart rate — from 58 ± 3 to 70 ± 3 bpm, against 60 ± 4 to 63 ± 4 on placebo, P < .05 between treatments — and lower blood pressure significantly. Their conclusion: AICAR “does not seem to have a direct effect on systemic or local glucose uptake in humans” and its hemodynamic effects are most likely systemic vasodilation. Rantzau 1985 sits alongside this, contrasting exercise, AICAR and increased fatty acid supply on in vivo and skeletal muscle glucose metabolism and finding the three are not equivalent.
The obstacle, and it is arithmetic rather than biology. Do the sums, because they are the page. AICAR's molecular weight is 258.2 g/mol. The perfusate concentration that produced the textbook effect was 2 mM, which is 516 mg per liter. This site's card lists a dose of 5 to 25 mg. Assume the most generous case imaginable — the entire dose absorbed instantly, distributed into plasma alone (about 3 liters), nothing cleared, nothing taken up by red cells. 25 mg is 0.097 mmol; in 3 liters that is 32 micromolar. Five milligrams gives 6.5 micromolar. The concentration that worked in the perfused hindlimb is therefore 62 to 310 times higher than the best case a retail dose can reach — and the real case is far worse, because Bosselaar 2011 showed erythrocytes strip AICAR out of plasma and trap it as ZMP.
The human doses point the same way. Babraj 2009 used 10–20 mg/kg/hour: for an 80 kg adult that is 800 to 1,600 mg every hour, for three hours — 2.4 to 4.8 grams total, intravenously. Bosselaar 2011 used 67 mg/kg, which is 5.4 grams in one hour. The retail dose is 0.005 to 0.025 grams. The smallest dose that has ever been given to a human in a published study is roughly 100 to 1,000 times the dose in circulation — and at 5.4 grams intravenously, the measured metabolic effect was nil.
One more mismatch worth naming. This site's card lists the route as subcutaneous. Every human study of AICAR has used intravenous infusion Babraj 2009 Bosselaar 2011, and the rodent and cell work used perfusate or intraperitoneal dosing. There is no published subcutaneous pharmacokinetic study of AICAR in any species.
AICAR pharmacokinetics — how much of it actually gets in
There is no published human half-life for AICAR given subcutaneously, and this site's card gives 30–45 minutes. The published human work infused it continuously precisely because a nucleoside of this kind does not persist Babraj 2009 Bosselaar 2011. A three-hour infusion is what you design when a bolus would not hold a concentration.
Where it goes, measured. Bosselaar 2011 sampled AICAR and ZMP in both plasma and erythrocytes during infusion and found plasma AICAR rising significantly and AICAR being rapidly taken up by erythrocytes and phosphorylated to ZMP. That is the single most useful pharmacokinetic observation about this compound and it is almost never quoted. Phosphorylation traps the molecule: ZMP carries a charge and cannot diffuse back out. So the red cell is not a transient compartment, it is a one-way sink, and a large one — erythrocytes are roughly 40% of blood volume and every one of them expresses adenosine kinase.
What activates it and what disposes of it. Activation is adenosine kinase, the same enzyme that salvages adenosine. Disposal of ZMP proceeds through the purine de novo pathway — ZMP is a natural intermediate of purine synthesis, which is why the cell has machinery for it — and unphosphorylated AICAR is cleared renally as a small polar nucleoside. The rate-limiting step for anything useful happening is therefore adenosine kinase capacity in the target tissue, and skeletal muscle has far less of it per gram than liver or red cells do.
The arithmetic that bounds a subcutaneous dose. A 25 mg subcutaneous bolus of a small, water-soluble, renally cleared nucleoside absorbs over tens of minutes and is cleared over tens of minutes, so the peak plasma concentration is a fraction of the instantaneous-distribution figure calculated above — call the 32 µM an unreachable ceiling rather than an estimate. Against a 2 mM working concentration Merrill 1997, and against 5.4 grams intravenously producing no measurable glucose effect Bosselaar 2011, the honest conclusion is that a retail-dose subcutaneous injection of AICAR cannot reach a concentration at which any published effect of AICAR has ever been observed. That is not a claim that the molecule does nothing. It is a claim about milligrams, liters and molecular weight, and anyone can check it.
What would have to be true, and how you would know it was not
Four predictions with a marker, a direction and a window. The first is the only human effect anyone has reproduced, and it is the one nobody wants.
1. Resting heart rate should rise and blood pressure should fall — if the dose is large enough to do anything at all. Bosselaar 2011 measured heart rate going from 58 to 70 bpm during AICAR infusion against 60 to 63 on placebo, P < .05, with a significant fall in blood pressure, attributed to systemic vasodilation. AICAR is a nucleoside relative of adenosine and vasodilation is what adenosine-like molecules do. Measure resting heart rate and blood pressure seated, before and at 30 and 60 minutes after a dose. This is the cheapest and most direct test available of whether the material is reaching the circulation in a pharmacologically meaningful amount.
2. The prediction that cuts against it: nothing metabolic should change. At 67 mg/kg intravenously — over two hundred times the retail dose — forearm glucose uptake and whole-body glucose infusion rate did not differ from placebo Bosselaar 2011. Draw Fasting Glucose, Fasting Insulin and Hemoglobin A1c (HbA1c) at baseline and 12 weeks. The prediction is no change, and it is made from a human trial at a vastly larger dose, not from scepticism.
3. Age should predict responsiveness more than metabolic status does. Babraj 2009 found the glucose-uptake response fell from 2.9-fold in young men to 1.8-fold in older men, with men with type 2 diabetes indistinguishable from age-matched controls, and concluded the blunting was dependent on age rather than diabetic status. That is a genuinely unusual and specific prediction: if two people respond differently to this compound, the published literature says the discriminator is age.
4. Blood pressure and heart rate should respond before anything else does, and if they do not, stop. This is the falsification that matters. In the only human experiments performed, the hemodynamic effect appeared at doses where the metabolic effect did not Bosselaar 2011. So hemodynamics is the more sensitive read-out. If a dose produces no heart-rate or blood-pressure change at all, the concentration is below the threshold for the effect that appears first, which makes the effects that appear later impossible.
What nobody has tested yet
Nobody has published a subcutaneous pharmacokinetic study of AICAR in any species. The route the entire market uses has never been characterized. One study — eight people, a subcutaneous dose, plasma AICAR and erythrocyte ZMP over four hours by the assay Bosselaar 2011 already used — would convert every argument on this page from arithmetic into measurement.
Nobody has resolved whether the human effect is AMPK at all. Babraj 2009 found glucose uptake rising with no detectable AMPK activation and with ERK1/2 phosphorylation tracking it at R² = 0.55. If the human mechanism is MAP kinase rather than AMPK, then the entire ‘exercise mimetic’ framing — which is built on the AMPK-PPAR-delta axis Narkar 2008 — is being applied to the wrong pathway. A biopsy study designed to separate the two has never been done.
Nobody has tested whether erythrocyte trapping can be saturated. Bosselaar 2011 showed red cells sequester AICAR as ZMP. If that sink is saturable, a slow infusion or a repeated small dose might fill it and let plasma concentrations rise for a subsequent dose — and if it is not, no schedule of subcutaneous injections will ever reach muscle. That is a straightforward in vitro experiment with whole blood and it would settle the feasibility of the entire retail category.
Nobody has run the mouse endurance experiment at a human-scaled dose. Narkar 2008 produced a 44% endurance increase in mice after four weeks. No human trial has ever measured an endurance endpoint on AICAR at any dose, by any route — the human studies measured glucose uptake and hemodynamics over hours. The single result that sells this compound has never been tested for in the species buying it.
AICAR — its own safety story, not its class's
The measured human effect of AICAR is cardiovascular, and it is the opposite of harmless-sounding. Bosselaar 2011 is explicit in its own title: intravenous AICAR “induces systemic hemodynamic changes but has no local metabolic effect”. Heart rate rose from 58 to 70 bpm and blood pressure fell significantly, in healthy young participants, during a controlled infusion with monitoring. Anyone stacking this with another vasodilator, with a beta-agonist, or with anything that drops blood pressure is stacking on top of a measured effect rather than a theoretical one.
The mechanism has an unavoidable off-target: this is an AMP mimic, and AMP is not a niche signal. ZMP accumulates wherever adenosine kinase is expressed, which is essentially everywhere, and AMP-sensitive enzymes beyond AMPK — fructose-1,6-bisphosphatase in gluconeogenesis, glycogen phosphorylase in glycogenolysis — respond to it too. The erythrocyte finding Bosselaar 2011 shows that the compartment taking up the most drug is the one with no mitochondria to benefit. A drug that mimics the cell's universal low-energy signal is not tissue-selective by design.
What the human safety record actually consists of. Three controlled infusion studies in healthy volunteers and patients Koistinen 2003 Babraj 2009 Bosselaar 2011, each lasting hours, with a combined sample in the dozens. There is no long-term human safety data for AICAR at any dose, no repeated-dose study, and no study of the route this compound is sold for. That is a different kind of ignorance from ‘the trial was run and found a problem’ and it should be described as what it is.
It is a prohibited substance in sport. AICAR sits on the World Anti-Doping Agency's prohibited list as a metabolic modulator, and it is detectable — it is an endogenous purine-pathway intermediate, so the analytical approach is a concentration threshold rather than simple presence. Any tested athlete should treat that as settled.
And the honest summary of the risk, which is unusual. For most compounds in this Vault the danger is that they work in a way nobody has characterized. Here the arithmetic says a 5–25 mg subcutaneous dose cannot approach any concentration at which AICAR has ever been shown to do anything, so the dominant risk is not toxicity — it is spending money and injections on a concentration two to three orders of magnitude below the published working range, and attributing to it whatever the training block would have produced anyway. The toxicity risk becomes real only if somebody, reading the arithmetic above, decides to scale the dose toward the gram range that was studied — at which point the hemodynamic effects are guaranteed and were only ever observed under continuous monitoring.
Sources read for this page
- Merrill GF, Kurth EJ, Hardie DG, Winder WW. AICA riboside increases AMP-activated protein kinase, fatty acid oxidation, and glucose uptake in rat muscle.. Am J Physiol 1997 · PMID 9435525
- 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. AMPK and PPARdelta agonists are exercise mimetics.. Cell 2008 · PMID 18674809
- Koistinen HA, Galuska D, Chibalin AV, Yang J, Zierath JR, Holman GD, Wallberg-Henriksson H. 5-amino-imidazole carboxamide riboside increases glucose transport and cell-surface GLUT4 content in skeletal muscle from subjects with type 2 diabetes.. Diabetes 2003 · PMID 12716734
- Rantzau C, Christopher M, Alford FP. Contrasting effects of exercise, AICAR, and increased fatty acid supply on in vivo and skeletal muscle glucose metabolism.. J Appl Physiol (1985) 2008 · PMID 18032581
- Babraj JA, Mustard K, Sutherland C, Towler MC, Chen S, Smith K, Green K, Leese G, Hardie DG, Rennie MJ, Cuthbertson DJ. Blunting of AICAR-induced human skeletal muscle glucose uptake in type 2 diabetes is dependent on age rather than diabetic status.. Am J Physiol Endocrinol Metab 2009 · PMID 19190259
- Bosselaar M, Smits P, van Loon LJ, Tack CJ. Intravenous AICAR during hyperinsulinemia induces systemic hemodynamic changes but has no local metabolic effect.. J Clin Pharmacol 2011 · PMID 21148051
AICAR — safety, predicted from mechanism
Predicted from mechanism, not from a human safety trial. How that reasoning works →
What the mechanism predicts
Derived from the molecule, not a trial.
- These act on mitochondrial function, NAD+ availability, sirtuin signaling or cellular clearance. The honest prediction is that acute harm is unlikely and the interesting risks are theoretical and long-range — which is a different shape of risk, not an absence of one.
- Growth and clearance signals cut both ways. Anything that improves the efficiency of cell survival is also improving it for cells you would rather not keep. Anything that pushes clearance hard is doing so indiscriminately.
- The near-term, practical ones: NAD+ precursors and infusions commonly cause flushing and a strong sensation if pushed fast, and several compounds in this class are stimulating enough to disrupt sleep.
What has actually been reported
- Generally well tolerated at studied doses. NAD+ infusion discomfort is rate-dependent and resolves by slowing down.
- The human evidence is mostly short trials with surrogate endpoints — a marker moved, not a life changed. That is worth knowing before you build a decade-long habit on it.
How to reduce the risk
Same mechanism as the prediction.
- Slow the infusion rate. Almost all NAD+ discomfort is rate, not dose. There is no prize for finishing quickly.
- Dose earlier in the day. Several of these are subtly stimulating, and sleep is where most of the repair you are paying for happens.
- Pick an endpoint you can actually measure, and take the baseline before you start. This is the class most prone to spending years on something with no way of knowing whether it did anything.
- Fix the basics first. Sleep, training and bloodwork move the same markers further than anything on this list, and they are free. A longevity compound stacked on four hours of sleep is a rounding error.
What it does to your bloodwork
A fact about the assay.
- There is no NAD+ assay worth ordering clinically. Judge this class on downstream markers — inflammatory markers, lipids, fasting glucose, and whatever your baseline panel showed as out of range.
Don't run this if
- Active malignancy, for the survival-signaling reasoning above.
- Pregnancy — uncharacterized.
The honest unknown
- Whether any of it extends healthspan in humans. Every honest person in this field is running on mechanism and animal data, and this catalog should say so rather than imply otherwise.
Not medical advice. If you take prescription medication or have a diagnosed condition, check this with a pharmacist or doctor.
AICAR — interference & stacking
Predicted from mechanism, not from an interaction study. How mechanism-predicted claims are made →
What AICAR moves on your bloodwork
Expected direction, not a measured one.
- hs-CRP (High-Sensitivity C-Reactive Protein) — ↓ expected to fall
Where these work, systemic inflammation is the plausible readout.
What to do: hs-CRP is cheap and moves. Baseline and 12 weeks. - HbA1c (Hemoglobin A1c) — ↓ expected to fall
Improved mitochondrial and metabolic function should show here if the effect is real at all.
What to do: The honest use of these markers is as a falsification test: if nothing moves in 12 weeks, the compound is not doing much for you. - Comprehensive Metabolic Panel (CMP) — ◆ worth watching
Liver and kidney function — the standard baseline for anything run long term.
What to do: Twice a year is enough on a stable protocol.
This class is where honest expectation-setting matters most: the markers above are how you find out whether anything happened, and for most of these compounds that question is genuinely open.
- How to work up to it, and when not to
- When to take it, and why that window
- Cycle length
- Time off between cycles
- Fasted or fed, and when in the day
- Needle gauge and injection site
- Coach Cam's personal notes
- Which compounds push the same lever, and why the dose adds up faster than people count
- What blunts it — the stacks that waste your money
- What compounds the risk, so a side effect arrives sooner than any one of them suggests
- Coach Cam's read on running it alongside the rest of your protocol
Everything above is free and stays free. Skool is where it becomes a plan — AICAR in an order, with the rest of what you're running.
Unlock in Skool — $10/mo →Bloodwork to run alongside AICAR
Baseline first, then again at 8–12 weeks.
| Marker | What it’s watching for |
|---|---|
| HbA1c (Hemoglobin A1c) | Where you started, so you can prove the change was real |
| Fasting Insulin | Moves years before HbA1c does — the earliest signal you get |
| Lipid Panel (Cholesterol, HDL, LDL, Triglycerides) | Rapid fat loss shifts triglycerides fast, in both directions |
| Comprehensive Metabolic Panel (CMP) | Liver, kidney and electrolytes while intake is restricted |
The Metabolic Health & Prediabetes panel covers these in one order — 8 markers, $81.45 with the discount applied.
Check results you already have → · All 103 markers A–Z
AICAR — frequently asked questions
What is AICAR?
AICAR (Acadesine) is a metabolic & fat loss research compound. AMPK activator — mimics the low-energy signal of exercise, driving glucose uptake and fatty-acid oxidation.
Is the full AICAR protocol on this page?
The reported research dose is on this page, along with how AICAR works and the evidence behind it. The protocol — how to work up to it, frequency, cycle length, time off, what not to stack it with and Coach Cam's notes — is inside Skool.
What is the half-life of AICAR?
AICAR has an approximate half-life of ~30-45 min, which is part of what determines how often it's dosed.
What's the evidence behind AICAR?
Current evidence level: Animal + research. AICAR is offered for research purposes only and is not an approved medicine.
What AICAR is used for
AICAR appears under 4 goals in the goal router.
Related Metabolic & Fat Loss compounds
Where this goes next
The pages here are the frameworks. The protocols — the dosing, the order to correct things in, the week-by-week schedule and what to retest — are inside Skool.