Labs Fail to Reproduce Protein's Appetite-Suppressing Effects

See allHide authors and affiliations

Science  09 Jul 2004:
Vol. 305, Issue 5681, pp. 158-159
DOI: 10.1126/science.305.5681.158

This article has a correction. Please see:

Two years ago, Stephen Bloom and his group at Imperial College London touched off a scramble in labs around the world with the latest discovery of how the body regulates appetite. The investigators reported in the 8 August 2002 issue of Nature that a molecule called peptide YY3-36 (PYY3-36), when injected into rodents, dampens appetite for 12 hours or more. But, like previous hot results in the quest for diet drugs, this one has cooled upon closer inspection. In an unusual joint letter in Nature this week, more than 40 scientists announced that they cannot reproduce Bloom's central findings.

Endocrinologist Bloom and his colleagues reported that PYY3-36 is formed in the digestive tract after eating and travels to the brain, where the peptide controls short-term appetite. (A different loop involving the hormone leptin regulates long-term body weight.) Researchers pigeonholed PYY3-36 with a seemingly related stomach peptide called ghrelin, discovered in 1999 by Kenji Kanagawa's team at the National Cardiovascular Center Research Institute in Osaka, Japan. In 2000, Matthias Tschöp, now at the University of Cincinnati in Ohio, linked ghrelin to hunger stimulation in the brain. Today, Tschöp is the leader of the PYY3-36 skeptics. Citing data from 1000 rodents, obtained in 12 labs, he and others say they cannot replicate Bloom's work. “We thought this information should be shared with the public,” Tschöp says, “because there are so many experiments going on with PYY3-36, costing millions of dollars.”

Bloom remains steadfast. Other labs “are doing the experiments badly,” he suggests. He thinks the problem is stress. If an animal is handled roughly and jabbed with a needle, Bloom says, it reacts to danger by not eating. The appetite loss can negate PYY3-36's appetite-suppressing effects. He says, “This is ignorance of animal behavior.”

In the 2 years following publication of Bloom's original paper, Tschöp's group and others tried to reproduce the initial findings in rodents. When they injected PYY3-36 into the bodies of young rats, as Bloom did, the animals showed no difference from controls in the amount or how often they ate. Neither did their body weight change. In contrast, Bloom's group tracked an average decrease in weight gain of about 20% in animals given PYY3-36 twice daily for 7 days, compared to controls given saline. Bloom also published a follow-up report last September in The New England Journal of Medicine, showing that PYY3-36 reduced appetite by 30% in 12 obese and 12 lean volunteers.

Food fight.

Matthias Tschöp (right) reports that 12 labs can't confirm that PYY3-36 suppresses appetite in mice. Stephen Bloom (left) says they are doing the experiments badly.


Frustrated, Tschöp began talking to other researchers and learned that they were experiencing similar troubles. The investigators decided to collaborate. They bought rodents from the same vendor as Bloom, purchased food from the same supplier, and injected animals at exactly the same time of day with the same kind of syringes filled with PYY3-36 manufactured by the same chemical supply house. “Every single bit of information we got, we followed up on,” Tschöp says. “And still we weren't able to reproduce the results.”

But at least one scientist has corroborated Bloom's results in rodents: Andrew Young, vice president and senior research fellow at Amylin Pharmaceuticals, headquartered in San Diego, California. His firm has a financial stake; it intends to market a drug based on PYY3-36. Young says that he, Bloom, and Tschöp began presenting contrary findings, back to back, at the same meetings. “Tschöp would say, ‘We can't get PYY3–36 to work,’” Young says. “And then I would get up and say, ‘We've tried and tried and can't get it not to work.’” Asked if he knows the reason for the discrepancy, Young says, “I can't explain it. Maybe it's [a difference in] the peptide.”

Tschöp's consortium claims to have shown that their peptides share the same molecular weight and basic chemistry as Bloom and Young's. And in answer to the suggestion that their animals were too stressed, Tschöp says that in his lab controls showed no loss of appetite when injected with saline, compared with animals that were not injected at all.

Despite the controversy, Amylin intends to test the drug in human trials. Tschöp admits it's possible that PYY3-36 could emerge as a powerful antiobesity drug. (His group has done no studies in people.) And one pioneer in the field, Bruce Schneider, associate vice president for clinical research at the Association of American Medical Colleges in Washington, D.C., says that conflicting data should not present a regulatory problem. The main concern is that a new drug be “reasonably” safe before entering clinical trials, says Schneider, a former medical officer at the U.S. Food and Drug Administration. But he cautions that “companies often take big risks” financially when they test antiobesity drugs—in part because scientists don't understand the causes of most forms of obesity, and any drug that affects metabolic systems can have surprising side effects.

Bloom is not fazed by the challenge. His findings have come under fire in the past, he says. Researchers criticized Bloom's initial data on a hormone called vasoactive intestinal polypeptide and a molecule called glucagon-like peptide 1. In both cases, Bloom says, the effects he reported turned out to be correct, although he had to go through years of extra work and funding struggles until the science ironed out. “Of course I believe in scientific process and crosschecking,” he says. “And I recognize that people make mistakes. But it kind of gets wearing and, in terms of an individual scientist's career, is a little bit destructive.”

The “more important issue,” says longtime obesity researcher Jeffrey Friedman of Rockefeller University in New York City, “is to determine what criteria need to be satisfied before one accepts a new molecule as playing a critical role in regulating food intake or body weight.” Indeed, when Friedman's group discovered leptin in 1994, they set the groundwork for those criteria. Researchers need to be careful before hyping such findings, Friedman cautions. After all, leptin failed as a drug even though it passed Friedman's stringent criteria.

Tschöp favors one simple precaution: Open up venues in which researchers can publish negative data that counter the initial claims. Just to challenge three initial papers about PYY3-36, he says, his team had to perform hundreds of experiments: “Right now it is much easier to find a needle in a haystack than to prove that in a huge haystack there isn't a needle inside.”

View Abstract

Stay Connected to Science

Navigate This Article