Three minutes of all-out effort, split into six 30-second sprints, changed nearly a quarter of the proteins circulating in the blood, while 90 minutes of steady moderate cycling shifted less than a quarter of one percent. The finding, reported August 13, 2026 by a team at Rockefeller University in New York, was published in the journal Cell Reports Medicine.
The result does not mean sprints replace endurance training. It suggests that intensity, not just total minutes, drives a distinct set of biochemical signals the body uses to adapt.
What the researchers measured
The Rockefeller team compared two protocols in the same participants: six 30-second sprints separated by short rests, and 90 minutes of continuous moderate cycling. Blood was drawn immediately after each session and again over the following hours.
Sprinting triggered changes in more than 200 metabolites, along with a surge of proteins tied to blood-vessel growth, tissue remodeling, and hormonal signaling. Moderate cycling produced only minor early changes.
Also read: 8 early warning signs of endometriosis that take years to get diagnosed
To gauge whether those shifts matter for long-term health, the researchers cross-checked the sprint-responsive proteins against health data from more than 53,000 people in the UK Biobank.
Of 33 proteins linked to lower cardiovascular and metabolic disease risk in that dataset, 32 moved with sprinting. Only three moved with moderate exercise. More than a quarter of the identified proteins were also associated with slower biological aging.
The effect persisted after eight weeks of training, arguing against the idea that the surge is simply the body's shock at an unfamiliar stress.
What the researchers said
"What's exciting here is that just a few minutes of intense exercise can trigger a significant molecular response," said Paul Cohen, head of the Weslie R. and William H. Janeway Laboratory of Molecular Metabolism at Rockefeller.
Also read: Taking the stairs may cut your risk of dying from heart disease by 39 percent
Luke Olsen, the postdoctoral fellow who led the work, wrote that the molecular mechanisms linking these intensity-dependent adaptations "have remained largely elusive." His group's work suggests exerkines, the proteins and metabolites released into the bloodstream during exercise, may be highly sensitive to intensity and could be the main drivers of the health effects seen with short, hard efforts.
Where it fits with current guidance
Federal physical-activity guidance recommends at least 75 minutes a week of vigorous-intensity aerobic activity for adults, or a longer stretch of moderate activity. The new findings do not change that target. They add molecular evidence that short, hard efforts activate pathways that steady sessions barely touch.
Interval training already has a track record. A 2025 systematic review in Frontiers in Physiology, pooling nine randomized trials with 666 participants, found that both high-intensity interval training and sprint interval training reduced body fat compared with controls, with sprint work looking especially useful for people who are already fit. Other work has linked even very small daily doses of exercise to a lower risk of early death.
The Rockefeller study measured what happens in the blood, not what happens on a race clock or a body-composition scan. Endurance training builds capacities, such as mitochondrial density and cardiovascular efficiency, that a three-minute sprint session cannot deliver on its own.
Also read: What your hair could be telling you about your thyroid, iron and hormones
The clearest reading of the paper is that a small dose of high-intensity work can supplement, not replace, regular activity, and that even brief efforts may carry biological weight for people short on time.
Anyone new to intense exercise, or living with heart or blood-pressure conditions, should check with a clinician before adding all-out sprints, since maximal efforts sharply raise short-term cardiovascular load.
Also read: Tomatoes may have a surprising benefit for liver health
This article is made and published by Jesper Bengtson, who may have used AI in the preparation.