AICAR

AICA riboside is taken into cells by adenosine transporters and phosphorylated to ZMP, which structurally resembles AMP and therefore activates AMPK without the cell actually being energy-depleted. Active AMPK switches the cell from anabolic to catabolic mode: it increases glucose uptake via GLUT4 translocation, stimulates fatty acid oxidation by inhibiting acetyl-CoA carboxylase, suppresses lipogenesis and gluconeogenesis, and over time drives PGC-1alpha-mediated mitochondrial biogenesis. The famous 2008 result was that sedentary mice given AICAR ran roughly 44% further than untreated controls. AMPK activation also directly opposes mTOR, so it is a poor choice during a hypertrophy phase - the mechanism that makes it good for endurance and fat oxidation actively suppresses muscle protein synthesis.

Targets: AMP-activated protein kinase (AMPK), Acetyl-CoA carboxylase (inhibited), GLUT4 translocation, PGC-1alpha, mTORC1 (inhibited)

• · 0.1 mg/kg/min infused during cardiac surgery is the dosing that generated the human safety data. It bears no relationship to how the compound is used recreationally.
• · 500-1000 mg daily is the commonly quoted range (shown here in micrograms). It is expensive, unvalidated for this purpose, and the rodent studies that inspired it used doses that do not scale to humans.

Work out your exact syringe units →

• commonElevated uric acid— AICAR is a purine intermediate; raising it can precipitate gout in susceptible people.
• commonInjection-site reaction— Large volumes at high doses.
• commonHeadache and dizziness
• commonSuppression of muscle protein synthesis— Not a side effect so much as an unavoidable consequence of AMPK activation - it will work against a bulking phase.
• uncommonTransient hypotension or bradycardia— Reported with infusion dosing.

• Gout or hyperuricaemia - the purine load is a direct problem.
• Significant renal impairment.
• Pregnancy and breastfeeding.
• Any competitive athlete subject to drug testing - this is an explicitly named prohibited substance.

• redundantmetformin — Both act on AMPK signalling; the combination adds GI intolerance more than effect.
• conflictigf-1-lr3 — AMPK activation inhibits mTOR, which is precisely the pathway IGF-1 is being used to switch on.
• synergyslu-pp-332 — Complementary exercise-mimetic mechanisms, with correspondingly stacked unknowns.

• · Serum uric acid before and during use.
• · Renal function.
• · Fasting glucose - AMPK activation improves glucose disposal and this is one of the few effects you can actually measure.

Never approved as a medicine. Explicitly prohibited at all times in sport by WADA as a metabolic modulator, with validated detection methods in place.

• Narkar et al. 2008, Cell - AMPK and PPARdelta agonists are exercise mimetics (preclinical)
• Acadesine cardiac surgery trials and pooled safety analyses (trial)
• WADA Prohibited List, section S4 metabolic modulators (guideline)

AICAR is a prodrug and everything interesting about it happens after it gets inside a cell. Adenosine transporters carry it in, adenosine kinase phosphorylates it to ZMP, and ZMP is close enough to AMP structurally to bind the cystathionine-beta-synthase domains on the AMPK gamma subunit. That does three things simultaneously: it promotes phosphorylation of Thr172 on the alpha subunit by LKB1, it inhibits dephosphorylation of that site by protein phosphatases, and it allosterically activates the already-phosphorylated enzyme. Active AMPK then executes the energy-stress programme - phosphorylating ACC to release malonyl-CoA inhibition of CPT1 so fatty acids enter mitochondria for oxidation, phosphorylating TBC1D1 to promote insulin-independent GLUT4 translocation, phosphorylating and activating PGC-1alpha to drive mitochondrial biogenesis, and inhibiting mTORC1 through TSC2 and raptor. That last one matters enormously for anyone reading this class for hypertrophy: AMPK activation directly suppresses mTORC1, which is the opposite of what an anabolic agent does. AICAR is not a muscle-building compound and is mechanistically antagonistic to one. Narkar's Cell paper is the reason it is famous: four weeks of AICAR alone, in sedentary mice, induced metabolic genes and enhanced running endurance by 44%, and PPARdelta agonism plus training was synergistic. That is a striking result and it generated the phrase 'exercise in a pill'. Goodyear's accompanying NEJM commentary, titled 'The exercise pill - too good to be true?', is the honest counterweight and is worth reading alongside it. The gap between the mouse result and human use is enormous: the doses used are grams per day for a human, the compound is poorly bioavailable orally, it is cleared within an hour, and it raises urate.

The dose arithmetic is the thing that ends most AICAR plans. The clinical cardiac studies infused 0.1 mg/kg/min continuously - for an 80 kg person that is 8 mg a minute, roughly half a gram an hour, delivered intravenously because the compound clears within an hour and does not absorb well orally. The anecdotal performance protocols of 500 mg to 1 g daily are neither the clinical route nor the clinical exposure, and the mouse studies used weight-adjusted doses that scale to grams. So the realistic situation is that most people taking oral AICAR are getting an inadequate dose of a compound that needed continuous infusion to work in humans, and paying a lot for it. The second problem is uric acid: purine pathway loading raises urate and can precipitate gout in anyone predisposed. The third is the mTORC1 point, which almost nobody in this space appreciates - AICAR actively suppresses the pathway that builds muscle. And the fourth is cost, because grams per day of anything is expensive, and AICAR is one of the more costly compounds in this catalogue on a per-dose basis.

• AMP-activated protein kinase (AMPK), via intracellular ZMP — ZMP is a lower-affinity AMP mimetic; it requires substantial intracellular accumulation to activate AMPK, which is why the doses are so large

  Allosteric activation plus promotion of Thr172 phosphorylation and protection from dephosphorylation. Drives fatty acid oxidation, GLUT4 translocation, mitochondrial biogenesis via PGC-1alpha, and mTORC1 inhibition.

• Adenosine kinase — Substrate rather than target

  Converts AICAR to ZMP. Cells with low adenosine kinase activity will not respond, which is a source of individual variability nobody accounts for.

• mTORC1 — Not a direct target

  Inhibited downstream of AMPK, through TSC2 and raptor phosphorylation. This is why AICAR opposes hypertrophy rather than supporting it.

• Purine biosynthesis and adenosine signalling — ZMP is an intermediate in de novo purine synthesis

  Off-target metabolic consequences including raised uric acid, which was a recognised finding in the clinical infusion studies.

Acute AMPK activation happens within minutes of achieving adequate intracellular ZMP, and disappears equally fast given the clearance. The adaptive changes - mitochondrial biogenesis, fibre-type shift, endurance capacity - took four weeks of daily dosing in the mouse work. In humans, no timeline has been established for any performance outcome.

The historically important stack is AICAR with a PPARdelta agonist such as GW501516, which is exactly what Narkar tested and found synergistic with training. It is also the combination that made both compounds notorious in cycling. Understand what that paper actually showed: AICAR alone worked in sedentary mice, and the drug-plus-training combination was synergistic - so the honest framing is amplification of training, not replacement of it. The interaction to take seriously is with metformin, which also activates AMPK, and with any insulin-sensitising or glucose-lowering agent, because AMPK-driven GLUT4 translocation is insulin-independent and adds to whatever else is lowering your glucose. The combination to avoid on mechanistic grounds is with anything you are taking for hypertrophy: AMPK activation inhibits mTORC1, so running AICAR alongside IGF-1 analogues, GH or a heavy resistance-training block is working against yourself. If you want endurance adaptation and hypertrophy simultaneously, AICAR is on the wrong side of that trade.

Against SLU-PP-332: both are exercise mimetics, but they work at different levels - AICAR acutely activates a kinase, SLU-PP-332 activates the transcription factor that drives the oxidative gene programme. SLU-PP-332 has better and more reproducible recent preclinical data; AICAR has actual human exposure through cardiac surgery trials and therefore some real safety information, which is not nothing. Against metformin, which also activates AMPK and costs almost nothing and has decades of human data, the argument for AICAR is that it activates AMPK directly through ZMP rather than indirectly through complex I inhibition - a real mechanistic distinction that has never been shown to matter for performance. Against actual endurance training, Narkar's own data showed the combination beat either alone, which is the fairest reading. And against everything else in this body-composition class, AICAR is the odd one out: it inhibits mTORC1, so it is an endurance and metabolic compound that works against hypertrophy rather than for it.

$300–$1200/month. Gram-scale daily dosing makes this one of the most expensive compounds in the class. Research-grade AICAR typically runs 40-100 USD per gram, and anecdotal protocols consume 15-30 g a month. The clinical infusion exposure would cost several times that again.

• ?No human oral bioavailability figure was resolved. The clinical use of continuous infusion strongly implies it is poor, but the number is not published in the sources reviewed here.
• ?No human study has tested AICAR for endurance performance or body composition at any dose.
• ?The relationship between the mouse doses in Narkar's work and any human dose has never been established by allometric scaling in a published source.
• ?Clearance, volume of distribution and protein binding are not published as numeric values in what was resolved here; the half-life figure of roughly 45 minutes in the Core record reflects the infusion-based clinical protocols rather than a formal PK study.
• ?Whether chronic AMPK activation impairs resistance-training adaptation in humans, as the mTORC1 mechanism predicts, has not been directly tested.
• ?Blood-brain barrier penetration is marked partial on the basis of nucleoside transporter expression at the barrier rather than a direct measurement for AICAR.
• ?Long-term consequences of chronic purine pathway loading beyond urate elevation are uncharacterised.

• AMPK and PPARdelta agonists are exercise mimetics 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, Cell, 2008 · PMID 18674809

  The famous paper. Four weeks of AICAR in sedentary mice induced metabolic genes and enhanced running endurance by 44%, and PPARdelta agonism synergised with training. Everything AICAR is sold on comes from here.

• The exercise pill - too good to be true? Goodyear LJ, New England Journal of Medicine, 2008 · PMID 18946072

  The sceptical commentary published alongside the excitement. Read it with Narkar rather than instead of it.

• Combined pharmacological activation of AMPK and PPARdelta potentiates the effects of exercise in trained mice Manio MC, Inoue K, Fujitani M, Matsumura S, Tsuruta Y, Physiological Reports, 2016 · PMID 26997622

  The follow-up in trained rather than sedentary animals - the more relevant model for anyone already training, and a more modest result.