As clear as the benefits of exercise are, explaining its effects at the molecular level has been scientifically strenuous. That’s especially true for irisin—a hormone that appears in some animal studies to increase energy expenditure after physical activity and promote healthy metabolism. Experiments to measure irisin in exercising people have set off a bitter scientific dispute, leading some to believe humans don’t produce it at all. Now, the lab that discovered the hormone is coming to its defense, aiming to win over critics by showing its presence in human blood with a more sophisticated, commonly used analytical technique.
Irisin—named for the Greek messenger goddess Iris—first revealed itself in the lab of Bruce Spiegelman, a cellular biologist at Harvard Medical School. His lab was trying to identify key proteins produced by muscles during a workout. “We’re all told to exercise,” he says. “In the end, presumably the [effects] will all come down to a series of molecules.”
The group identified a fragment of a protein called FNDC5 that was secreted into the blood of physically active mice. This fragment, irisin, appeared to be a powerful messenger that boosted metabolism by prompting energy-storing white fat cells to behave more like energy-burning brown fat cells. The team also found evidence for the molecule in human serum in people undertaking endurance training, and saw increased levels after exercise. The hormone, first reported in a 2012 Nature paper, became an enticing lead for treating obesity and metabolic disease, and Harvard and Spiegelman licensed the discovery to a company co-founded by the researcher in 2011 called Ember Therapeutics.
But that idea was soon under fire. Other labs looking at people failed to see increases in irisin after exercise. And two especially critical papers suggested humans produce little or no irisin at all. In one, biologist Steffen Maak of the Leibniz Institute for Farm Animal Biology in Dummerstorf, Germany, and colleagues claimed that the commercial antibodies used to detect irisin in blood were prone to false positives and that many studies measuring the hormone in humans were likely invalid.
Another, led by diabetes researcher Juergen Eckel at the University of Düsseldorf in Germany, examined the DNA sequence that signals the start of the irisin gene. In humans, this sequence has an unusual mutation, which Eckel’s group suggested produces a truncated protein. They found that this version of the gene produced 100-fold less full-length irisin than those found in other animals, and they concluded that any health benefits of the hormone were unlikely to hold for humans.
To defend irisin, Spiegelman’s group has now turned to mass spectrometry—a common analytical technique he calls “the most accurate and definitive way to measure the existence of anything.” The researchers snipped up proteins from human blood into their component pieces—known as peptides—and exposed them to electrical current to create charged fragments. When they passed these fragments through a magnetic field, their paths curved at different angles depending on their masses, allowing the team to identify specific peptides based on where they ended up. Another group had reported an irisin-specific peptide in human blood based on this technique, but it hadn’t determined how much of the hormone was present.
Spiegelman’s group used a more sophisticated process that could quantify the irisin-derived peptides by comparing levels of current they generated with those of reference peptides that were chemically identical. That method turned up irisin peptides in all of 10 research subjects, the group reports today in Cell Metabolism. And the results suggest that the human protein is not truncated: Among the peptides detected was one encoded right near the mutated starting sequence of the gene, suggesting that the body can still produce a full-length protein using that DNA.
To reinforce the link between irisin and exercise, the group also compared sedentary and active subjects. Six people who had been on a 12-week aerobic exercise regimen had an average of 4.3 nanograms of irisin per milliliter of blood serum, according to the mass spectrometry analyses; four people with no such training had only about 3.6. Although it’s not clear what effects on cells and tissues the hormone has at those levels, Spiegelman and colleagues point out the serum concentration of insulin, an important regulator of blood sugar, is in a similar range.
The new technique offers other groups a reliable—if expensive—way measure irisin, which could be used to validate the controversial antibody-based tests, says Francesco Celi, an endocrinologist at Virginia Commonwealth University in Richmond who led the previous mass spectrometry work on irisin. The paper will elicit “a big sigh of relief” from the field, he says.
For irisin’s skeptics, the mass spectrometry result itself is hard to dispute. But it’s also very hard to reconcile with the previous finding that irisin is produced at dramatically lower levels in humans than in other animals. “I’m still suspicious,” says biochemist Harold Erickson at Duke University in Durham, North Carolina, and first author on the paper that questioned the antibody tests. University of Oslo nutritional biologist Christian Drevon, who took part in the study suggesting lower human production of irisin, says the new approach could be an “improved method” for detecting the protein if validated in other labs. But he sees no convincing evidence that irisin is sensitive to exercise.
That’s because the exercise component of the study is weak, he says. Other researchers concur. The differences between trainees and sedentary people could be the result of factors other than irisin, they caution, and the experiment should have observed the same subjects both before and after exercise. Spiegelman concedes that the new work wasn’t a rigorous exercise study, but he says his lab will continue to probe the effects of irisin on human brain, bone, and fat tissues.