The Molecular Interference Effect: Does AMPK Really Switch Off mTOR When You Do Cardio?

The molecular interference effect is the idea that cardio switches on a cell-level energy sensor called AMPK, which then dampens mTORC1, the signal that tells muscle to build protein. It is a clean story. It is also much less clear in trained humans than the gym internet makes it sound.

This page is the biology deep dive. If you want the practical programming side, start with our guide to cardio and concurrent training, then come back here for the “why”.

What is the molecular interference effect?

Your muscle cells run two broad adaptation programs. Lifting heavy loads pushes the growth program, centred on mTORC1. Endurance work pushes the energy-and-endurance program, centred on AMPK and mitochondrial building. In cells, the two programs push against each other.

The logic goes like this. Cardio drains energy inside the fibre. AMPK senses that and turns down expensive jobs like building new protein. AMPK has been shown to phosphorylate raptor, a core part of mTORC1, and to act on TSC2, another upstream brake (Gwinn et al., 2008). In that work, the raptor step was needed for mTORC1 inhibition under energy stress in cells.

Notice the source. That raptor paper used cultured cells, not people doing leg press. A lot of the early evidence for this model came from cells and rodents, and that is the first reason to hold it loosely.

Does AMPK really block mTOR in human muscle?

This is where it gets interesting, because researchers tested exactly that with muscle biopsies.

In one study, ten moderately trained men did heavy leg press, either alone or followed by 30 minutes of cycling. Phosphorylation of mTOR rose about 2-fold and S6K1, a downstream growth marker, about 14-fold, with no difference between the two sessions (Apró et al., 2013). The authors concluded that signaling through mTORC1-S6K1 after heavy lifting was not inhibited by later endurance exercise.

A second study made it harder for the lifters. Eight trained men did interval cycling first, which raised AMPK activity, then lifted. S6K1 activity and the rest of the growth markers still rose with no difference from lifting alone (Apró et al., 2015). So AMPK was up, and mTORC1 signaling still went ahead.

Then a 2021 crossover in eight male cyclists compared lifting alone with lifting plus moderate or high-intensity cycling. The authors reported no support for a molecular interference effect under those controlled conditions, and no intensity-dependent change in AMPK (Jones et al., 2021).

A review of this literature reached a similar conclusion: human studies had not observed the molecular interference after acute concurrent exercise that could explain compromised hypertrophy (Fyfe et al., 2014).

Fair caveat: these studies are small (8 to 10 men), short-term, and measure signals a few hours after one session. A signal is not growth. Also, they do not rule out that repeated sessions over months behave differently.

If the signal is intact, why do some studies still show interference?

Good question. The 2015 study offers a clue. Cycling before lifting did not hurt S6K1, but it raised gene and protein expression of MuRF1, a marker linked to protein breakdown, which lifting alone did not (Apró et al., 2015). The authors suggested that interference may act partly by influencing protein breakdown rather than only by blocking synthesis.

That fits a broader view. Chronic interference may be less about one molecular switch and more about the program: how much endurance work, how often, and what kind. It may also come from plain fatigue that makes your lifting sessions worse.

What do the real-world training studies show?

Zoom out from the biopsy and look at the meta-analyses.

A 2012 meta-analysis of 21 studies (422 effect sizes) found mean hypertrophy effect sizes of 1.23 for strength training, 0.85 for concurrent training and 0.27 for endurance training alone (Wilson et al., 2012). Concurrent training was clearly better than endurance alone, but numerically lower than lifting alone. The same paper found that running, not cycling, was linked to decrements in hypertrophy and strength, and that higher endurance frequency and duration correlated negatively with outcomes.

A newer analysis of 43 studies in healthy adults came out more reassuring. Compared with identical lifting alone, concurrent training had a standardised mean difference of -0.01 for muscle hypertrophy (95% CI -0.16 to 0.18) and -0.06 for maximal strength, both non-significant, with a small negative effect on explosive strength (-0.28) (Schumann et al., 2022). Explosive strength was more attenuated when both were done in the same session than when sessions were at least 3 hours apart. Running versus cycling, frequency, training status and age did not significantly change the results.

The same research group then looked at individual muscle fibres across 15 studies. They found a small negative overall effect on fibre hypertrophy (SMD -0.23, 95% CI -0.46 to 0.00), and a larger negative effect on type I fibres when the aerobic work was running (SMD -0.81) but not cycling (Lundberg et al., 2022). The authors called that running finding preliminary.

So what do I do with this? A practical read

The molecular story and the training studies point the same way: the issue is dose, not the existence of cardio. Here is the plain-gym translation.

Lever	What the evidence suggests	Practical move
Endurance volume	Higher frequency and duration linked to larger decrements (Wilson 2012)	Keep weekly cardio moderate and add it gradually
Mode	Running linked to more interference than cycling in two meta-analyses, but not in Schumann 2022	Cycle or incline walk if growth is the priority
Same session or not	Explosive strength blunted more when same session, not when 3+ hours apart	Split sessions if convenient
Order	Similar AMPK and mTOR responses in one study of 18 trained men	Do the priority session first
Intensity	No intensity-dependent AMPK change in 8 cyclists	Choose what you can recover from

The order point comes from a study of resistance-trained men comparing lifting alone, lifting then endurance, and endurance then lifting. AMPK, mTOR and related signals responded similarly, and the authors said they could not tell which sequence was most favourable (Jones et al., 2016).

A simple weekly setup for lifters who run or cycle

Most people reading this are in a gym three to five days a week and want to keep some conditioning. A sensible, hedged template:

• Lift first on days you do both, then do easy cardio.
• Use low-impact cardio (cycling, incline walking) for most of your weekly minutes. See incline treadmill walking for lifters and cycling vs running for muscle preservation.
• If you want to run longer distances, plan around it using programming concurrent training for 5K and 10K running.
• Use our heart rate zone calculator to keep easy days easy.
• Track your key lifts. If squat or bench numbers stall for several weeks while cardio rises, cut cardio first.

In an Indian context, heat matters. A 45-minute run at noon in summer adds a lot of fatigue and sweat loss. Early morning or evening, or a stationary bike indoors, gives you the cardio benefit with less recovery debt. Dal, paneer, curd, eggs and soya chunks all help you hit protein, and enough rice or roti keeps total calories from sliding.

If the interference concern is really about being in a deficit, our piece on mTOR and AMPK in calorie deficits versus surpluses goes deeper, and timing cardio and weights covers the scheduling questions in detail. The wider hypertrophy picture is in the muscle hypertrophy guide for natural lifters.

What we still do not know

Be wary of anyone who sounds certain. Biopsy studies are short and small, mostly men, and look at hours after one session. Meta-analyses mix very different programs, and many participants were not highly trained. We do not have long-term biopsy data on a lifter doing months of high-volume endurance work. And individual response varies, so some people may be more sensitive than the averages suggest.

Also: if you have a heart condition, chest pain, unexplained shortness of breath or dizziness, talk to a doctor before raising your training load. Nothing here is medical advice.

The bottom line

The molecular interference effect is a real mechanism in cells and a shaky explanation in trained humans. In practice, keep cardio moderate, favour cycling or walking if you are guarding muscle, split sessions when easy, and fuel well. For most lifters, cardio does not turn muscle growth off.