IN FEBRUARY, Duke University biochemistry professor Homme W. Hellinga and his coauthors retracted protein design papers published in Science (2004, 304, 1967) and the Journal of Molecular Biology (2007, 366, 945). Because these papers represented an important milestone in attaining the coveted goal of being able to design an enzyme able to catalyze any particular reaction, some in the community were concerned the retractions could hurt the field.
Three months later, despite the vitriolic and anonymous chatter on the blogosphere, the protein design community questions whether the retractions will have any effects beyond those on Hellinga's lab.
In the original papers, Hellinga and coworkers claimed to use computational methods to design proteins called NovoTims that catalyze the same reaction catalyzed by the enzyme triosephosphate isomerase (TIM). Although the reported kinetic values for the best NovoTim weren't as efficient as the natural TIM enzyme, the work offered hope that scientists would someday be able to design proteins capable of catalyzing any reaction. C&EN initially reported the research in 2004 (July 12, 2004, page 24).
The designed enzymes caught the attention of John P. Richard, a biochemist at the State University of New York, Buffalo, who studies the natural TIM enzyme. He wanted to profile the designed TIM's activity more closely because, he says, such experiments "might tell you something about why it didn't have as high an activity as the wild-type protein."
Richard and his coworkers used the method published by Hellinga's group to make NovoTim in bacteria and then purify it. But the protein they isolated had much higher kinetic values than the Hellinga team had reported for the best of its designed enzymes, Richard says. He and his coworkers suspected that this activity resulted from contamination with wild-type TIM from the bacteria used to produce NovoTim. When they used a different purification strategy to isolate the expressed protein, the protein they harvested showed no TIM activity.
Disappointed that he had spent so much time trying to reproduce published procedures, Richard wrote a letter to Hellinga in July 2007 and sent copies to editors at Science and the Journal of Molecular Biology, urging further explanation of the discrepancies. In response, Hellinga repeated the experiments with Richard's purification method. The results verified that the designed protein had no TIM activity. Hellinga retracted the Science paper on Feb. 1 and the Journal of Molecular Biology paper on Feb. 23.
Also in February, Nature reported that Duke had launched an inquiry of misconduct allegations in September 2007 against Mary A. Dwyer, the graduate student who was the first author on the 2004 Science paper. Nature also reported that on Feb. 4, "Dwyer was notified that she had been cleared." Dwyer and Loren L. Looger, another former graduate student coauthor, declined to be interviewed for this article.
Duke also declined to comment on the status of this inquiry and whether an investigation of Hellinga is pending. In an e-mail statement, Douglas J. Stokke, assistant vice president of Duke's medical center news office, said: "Duke University and its faculty are committed to conducting the highest quality science and research. We are aware the retraction by Dr. Hellinga has generated debate in the scientific community. Duke continues to follow this debate and is evaluating various points that are being raised."
Members of the enzymology community are concerned that the retractions fail to address several issues with the paper. Jack F. Kirsch, a biochemist at the University of California, Berkeley, laid out his objections in a letter to Science published online on March 10.
"The retraction only admitted to contamination with wild-type enzyme," Kirsch tells C&EN. "That doesn't explain the very low KM values that they reported for NovoTim." KM, also known as the Michaelis constant, is a reaction parameter that defines the substrate concentration at which the reaction reaches half its maximal velocity. If the wild-type enzyme is the contaminant, Kirsch points out, "it's very hard to think of a way you could get a KM value that is much lower than that of the wild-type enzyme."
In his letter, Kirsch further pointed out that some of the reported results would only make sense if the design had actually succeeded. For example, Hellinga's team reported that as each of the three critical active-site residues in the designed protein was replaced with alanine, the mutants became less active. Double mutants are less active than single mutants, and triple mutants are the least active of all. "That's not what you would expect if there was random wild-type contamination," Kirsch says.
Kirsch also questioned the decision not to explain the original in vivo results in the retraction; NovoTim was reported to restore TIM activity in an Escherichia coli mutant lacking the bacterial wild-type TIM. He notes that it was this particular in vivo result that made the original papers so convincing and that it can't be explained by wild-type contamination.
Nobody knows what actually happened to generate such incorrect data about the designed proteins. The possibilities range from experimental carelessness to selective choice of included data to outright fabrication, but no one is willing to make such accusations.
Richard and John A. Gerlt, a biochemist at the University of Illinois, Urbana-Champaign, argue that the problems go deeper than contamination. "There was a fundamental flaw in all of this," Gerlt says. "Not an experimental flaw but a design flaw."
THAT DESIGN FLAW manifests itself in the reaction's stereochemistry. TIM catalyzes the interconversion of dihydroxyacetone phosphate (DHAP) and D-glyceraldehyde 3-phosphate (GAP). However, the NovoTim protein, as Hellinga's group described it, abstracts a proton that results in the formation of L-GAP. In a letter to Hellinga, Richard explained the situation in more detail: "Any protein designed to catalyze suprafacial transfer of the pro-S hydrogen of DHAP via a cis-enediol[ate] intermediate would form L-GAP and would, therefore, not show activity using the standard enzymatic assay for isomerization of DHAP. This is because L-GAP is not a substrate for the coupling enzyme [glyceraldehyde 3-phosphate dehydrogenase] used in the assay for isomerization of DHAP." Therefore, even if the designed protein had worked, the assays shouldn't have given a positive response.
Hellinga tells C&EN that he decided not to address such questions because he believes the fact that the designed enzyme ultimately didn't show TIM activity made many other points moot. "I didn't see a reason to go into the design methodology because the experiment clearly didn't work," he says. "By inference, obviously the design was wrong."
People in the protein design community remain confident that the retractions will not harm progress in the field of protein design. Shortly after the Science retraction was published, for example, a team led by David Baker at the University of Washington published papers describing designed proteins that catalyze retro-aldol reactions (C&EN, March 10, page 13; Science 2008, 319, 1387) and Kemp elimination reactions (Nature, DOI: 10.1038/nature06879).
"The bottom line is that [Hellinga's] molecules don't work. Is that an indictment against the field? I don't think so," says Stephen L. Mayo, a biology and chemistry professor at California Institute of Technology who also develops computational methods to design proteins. He describes the enzymes designed by the Baker group as "unimpeachable."
Hellinga acknowledges that Baker's results bode well for the health of the protein design field. "My lab may have had a real problem, but the field does not," he says.
William F. DeGrado, a biochemist and protein designer at the University of Pennsylvania, suggests that in the future, problems could be avoided by requiring authors to include more extensive characterization of designed proteins.
"What's absolutely critical is not only putting the computational methods together but also putting the experimental methods together in such a way that there's absolutely no doubt that what you've done is right," DeGrado says. "Structural validation is an important issue that increasingly needs to be included in design papers."
Barry L. Stoddard, a structural biologist at the Fred Hutchinson Cancer Research Center and the University of Washington, Seattle, believes that protein design is an "incredibly important problem" that "demands an enormous amount of caution and care." He hopes that the field now realizes that it needs to "proceed slowly and carefully," especially when claiming such large advances.
Mayo points out that the retractions don't negate the underlying computational methodology of designing proteins de novo from a pantry of amino acids. Methods similar to Hellinga's have been successfully integrated in software written at Caltech and the University of Washington. "The retraction is a retraction of experimental results. The retraction doesn't reach back into the methodology," Mayo says. "I think there are compelling data to suggest that the fundamental methodology for developing software to design amino acid sequences is there."