The virtues of exercise are myriad: better cardiovascular health, decreased risk for diabetes, boosted mood, and even perhaps a leaner physique. But aside from such macro links and knowledge about the heart rates, blood–oxygen levels and hormonal responses related to exercise, scientists have a relatively cursory understanding of the chemical mechanisms at work in the body during and after physical activity.

A new study, published online May 26 in Science Translational Medicine, presents a thorough profile of exercise's impact on the human body's metabolites in plasma—and reveals vast biological differences among more- and less-fit individuals. The findings also reinforce links between exercise and insulin sensitivity as well as point to new ways to enhance exercise for both healthy individuals and those suffering from cardiovascular or metabolic diseases.

"We've known for a long time that exercise protects against metabolic and cardiovascular disease," says Robert Gerszten, director of Translational Research in the Cardiology Division at Massachusetts General Hospital, and co-author of the new study. "But how these effects occur is not entirely clear. Exercise physiology is surprisingly very poorly understood."

Predictive patterns
Gerszten and his colleagues measured more than 200 metabolites, small chemicals involved in the body's metabolism, using mass spectrometry in individuals before and after exercise. The team found that people who were more and less fit had vastly different metabolite profiles across a wide range of exercise durations—from 10 minutes of treadmill running to completion of the Boston Marathon.

And even in those who exercised for just 10 minutes, their metabolite profile was still being affected an hour later, which suggests that exercise sparks "long-term signals [and] that we're just beginning to scratch the tip of the iceberg," Gerszten says.

Across all groups, after exercise the researchers saw an increase in certain metabolites, such as glycerol, that are involved in the breakdown of fatty tissue as well as a decrease in allantoin, which is linked to oxidative stress. But not all changes were equal. Glycerol levels increased much more in fitter people (as assessed by peak oxygen uptake). In leaner subjects (with body mass indexes, or BMIs, less than 28), levels of niacinamide, a metabolite that increases sensitivity to insulin, increased more than twice as much as it did in larger individuals.

Not all of these changes were a surprise, however. "Many of [them] were not a surprise and have been shown before" in muscle, says Robert Hickner, director of the PhD Program in Bioenergetics and Exercise Science at East Carolina University in Greenville, who was not involved in the study. "It's not that surprising that you see the higher glycerol response" in fit people, he notes, because they have been shown to have higher rates of fat breakdown.

Gerszten and his team also found that some of the metabolites that were changed with exercise also could be applied to a broader population correlated with cardiovascular health. "Some of those metabolic changes in acute exercise also turn out to be predictive of fitness in a large epidemiological cohort," he says. Of the dozen of these metabolites that were measured in cohorts from the longitudinal Framingham Heart Study (the long-term 1948 study to identify the common contributive factors of cardiovascular disease by following its development in a large subject group) levels of glycerol turned out to be "significantly correlated" with resting heart rate, another indicator of fitness, as measured in the study.

The next question to answer will be whether these biomarkers also predict long-term disease and mortality risk, Gerszten says. And if they do, he notes, they could help doctors do a better job at finding at-risk populations and make suggestions for preventive health interventions.

"There's certainly room for a good test to predict disease," Hickner says. And the quest for new biomarkers continues: "The question is can we find better ones," he says. "I don't think they've come upon any in this study."

Deeper mechanisms
Beyond the fluctuation in metabolite levels, the new study also found that they can work together to influence gene expression. By exposing muscle in vitro to a combination of metabolites, the researchers found that transcription factor nur77, which has recently been linked to lipid metabolism and glucose use, was upregulated (but had remained unchanged when exposed to only singular metabolites). "This shows you that metabolites are going to act in very unanticipated ways," Gerszten says.

He cautions, however, that, "this is an early study, but it does give us a lot of insight into what some of the functional players might be." Gerszten and his colleagues are now looking into the role of genetics in determining an individual's plasma metabolite profile.

Exercise enhanced

The findings of the new study have implications for "optimized training programs as well as the development of nutritional supplements to enhance performance," Gerszten says. By comparing metabolite profiles of unfit people with fit ones, he notes, researchers could find the metabolite lags that fitness drinks or nutritional supplements might replenish.

He and his colleagues are already at work trying to locate which metabolites are most important for regulating pathways relevant to boosting overall performance. "We're going to look over time to see which metabolites change over various interventions," Gerszten says. "This gives us a more granular look into the pathways that are changing."

But the performance enhancement research is not just about helping athletes or gym junkies up their game, Gerszten says. Rather it is "most importantly for people with metabolic and cardiovascular disorders" to help them be able to exercise better and perhaps improve their condition.

Whether or not a metabolite-enhanced drink or supplement could vastly increase performance is still up in the air, Gerszten says. "It's too early to tell."

But it will not be as simple as plugging in metabolites that seem to be lower in those needing extra help exercising. "It's very hard to find supplements that improve performance other than water, glucose, carbohydrates," Hickner says. "It might be a stretch from this study to get into that realm of taking the metabolic profile and determining what supplement you should give."

The real challenge, Hickner says, is not finding the ideal formula for boosting performance, but wading through the data itself. "The ability to study many metabolites at once is becoming valuable," he says. But "to sort through the data is our challenge in the coming 10 years."