Compass protein attracts heap of criticism

Debate grows over a molecule implicated in animal navigation.

The red-spotted newt and many other animals can somehow sense Earth’s magnetic field. Credit: George Grall/NGC

For decades, scientists have wondered how animals can navigate huge distances using the weak signals of Earth’s magnetic field. So, interest was piqued in 2015 when two teams released papers in quick succession describing the functions of a protein found in animals that seemed to sense magnetic fields. But the claims have proved controversial, and questions have been piling up.

The basic science behind the discovery was reported by Xie Can, a biophysicist at Peking University in Beijing, and his colleagues. In a paper in Nature Materials¹, they claimed that a protein in animal cells forms a structure that responds to magnetic fields, and so might help in navigation. In the same year, a group led by Zhang Sheng-jia, then at Tsinghua University in Beijing, had published a paper in Science Bulletin² reporting that the same protein could offer a powerful means of controlling brain cells.

An academic battle has long raged between Xie and Zhang, but mounting evidence has cast doubt on both of their discoveries. Several researchers have challenged Xie’s claims that the protein reacts to magnetic fields. And last month, Xie co-authored a paper in Frontiers in Neural Circuits³ disputing Zhang’s work on the protein’s potential to magnetically control cells.

This has all given rise to serious questions about the role of the molecule at the centre of the dispute. In their 2015 paper¹, Xie and his colleagues reported that a protein called IscA1 forms a complex with another protein, Cry4, that explains how organisms pick up magnetic cues. The study found that this complex incorporates iron atoms, which gives it magnetic properties, and has a rod-like shape that aligns with an applied magnetic field.

Two months earlier, Zhang had described using IscA1 to control neurons and muscle cells in worms². Zhang learned of IscA1’s properties and obtained his IscA1 samples from Xie, and so the fact that his team published first was an early source of tension in what quickly became a bitter dispute. Officials from both Tsinghua University and Peking University asked Science Bulletin to retract Zhang’s paper. And that November, Zhang lost his position at Tsinghua — for reasons that the university did not specify.

The data are what they are. This may expand our knowledge of molecular magnets.

Doubts about Xie’s research have emerged since then. Michael Winklhofer, a geophysicist at the University of Oldenburg in Germany, examined Xie’s data and found that the complex would be too weakly magnetic to sense Earth’s field⁴. Markus Meister, a biophysicist at the California Institute of Technology in Pasadena, raised similar concerns: Xie had reported that the complex would contain only 40 iron atoms, but Meister argues that the smallest known naturally occurring iron-based magnet has 1 million iron atoms packed into a smaller space⁵.

David Keays, a neuroscientist at the Institute of Molecular Pathology in Vienna, has also questioned the study. He says that IscA1 and Cry4 are found throughout many tissues, whereas one would expect them to be sequestered in specific areas if they were functioning as parts of a magnetic-field receptor. “Sensory receptors, whether they be taste, hearing or photo-receptors, tend to have a restricted expression pattern,” he says.

Collaborators of Xie say that they have been able to reproduce some of his findings, and Xie told Nature that he stands by his results. He disputes the contention that the magnetic properties of IscA1 would be too weak by saying that Cry4 might boost its effect. “The data are what they are,” he says. “This may expand our knowledge of molecular magnets.”

The challenge to Zhang’s paper has been more pointed. Zhang claimed to have transferred IscA1 into worm neurons and then used a magnetic field to induce the cells to take up calcium. The ability to manipulate such a basic cell function could promise neuroscientists a powerful tool that is less invasive than opto-genetic techniques, which use light-sensitive proteins to control neurons in living animals.

But last month, Xie, Tsinghua University neuroscientist Lu Bai and Lu’s student Pang Keliang reported³ carrying out experiments under various conditions, including some almost identical to those used by Zhang. They found no change in calcium flowing into cells in any of the cases. The authors conclude that the “findings cast serious doubts” that IscA1 alone could influence the activity of neurons, as Zhang had claimed.

Several scientists outside China also told Nature that they could not reproduce Zhang’s results. Nature tried to reach Zhang through multiple e-mails and phone calls to Shenzhen University in China, where he now has a position, but he did not respond to requests for comment. (Neither Nature Materials, which is editorially independent from Nature’s news team, nor Science Bulletin responded to requests for comment about criticism of the papers.)

Meanwhile, even as his critics become increasingly aggressive, Xie says he has convincing data that demonstrate the reaction of an IscA1 complex to a magnetic field, and that he plans to publish them within a year. “We are more and more confident — 100% sure — that we are right about this,” he says.