In the Jan 31, 2008, issue of Nature, two research groups presented what each believes to be the binding mechanism of the anti-influenza drug amantadine to its target, the M2 channel. While the two groups agree on the basic structure of the channel, that's about where it ends. Not only do they present different amantadine binding mechanisms, but they also disagree on whether their data are incompatible or not.

The M2 channel is the target of one major class of anti-influenza drugs, the amantadines, which consist of amantadine and rimantadine.

The other major group, of which Tamiflu is the most prominent member, targets the viral neuraminidase. And one could argue that neuraminidase inhibitors are the only remaining major class of anti-influenza drugs.

Over the past three years, due to an unfortunate confluence of increased use and viral reassortment mechanisms, resistance to the amantadines has developed explosively. "Right now, the balance of power has clearly shifted to the virus," James Chou told BioWorld Today.

The goal of the groups was to determine the structure of the M2 channel with amantadine or rimantadine, respectively, bound to it, gaining insight into the drug's mechanism of action and ideas for targeting influenza A - a group of flu viruses that includes H5N1, the avian flu strain that many public health experts believe is the most likely candidate for an influenza pandemic.

One group, led by senior author William DeGrado, professor of biochemistry and biophysics at the University of Pennsylvania, used X-ray crystallography to determine the binding mechanism of amantadine.

Their paper appears to confirm the dominant idea of how amantadine acts. Their data showed that something - in their interpretation, amantadine - binds directly inside the M2 channel, acting as a plug and preventing protons from passing through.

In the other paper, senior author James Chou, assistant professor of biological chemistry and molecular pharmacology at Harvard Medical School, and first author Jason Schnell used nuclear magnetic resonance spectroscopy to solve the structure. They come to a starkly different conclusion about how and where amantadine binds.

"Before, most people - including us - had this very intuitive model" of amantadine's binding to the M2 channel, Chou told BioWorld Today - namely the plug model championed by DeGRado's group. "But that is not true - the drug is binding a pocket right outside the channel." In his view, the binding changes the channel's shape so that it is more likely to be in a closed state.

The two groups don't just disagree on the binding mechanism; they also disagree on whether their data are incompatible.

"I don't think there is a conflict at all," DeGrado told BioWorld Today. "Channels have many different states. We solved a structure at low-pH and intermediate-pH, and James Chou solved a high-pH form."

Chou was less conciliatory about the differences in the papers. "I believe that their structure is an artifact of the crystallization process," used by DeGrado's group, he told BioWorld Today.

His explanation for the electron density observed by DeGrado's group is that it stems from part of a detergent used during crystallization.

Other researchers, though they avoided picking a winner, appear to agree with Chou that the two binding mechanisms cannot be reconciled easily.

The author of a "News and Views" article accompanying the two papers called the proposed binding mechanisms "incompatible."

And Mei Hong, assistant professor at Iowa State University, when asked whether the disagreements could be reconciled, said that "my first instinct is that they really disagree." Hong and co-author Sarah Cady published a paper in the early online edition of the Proceedings of the National Academy of Sciences this week that uses solid-state NMR spectroscopy to study the amantadine/M2 channel complex.

While methodological limitations of solid-state NMR restricted Cady and Hong to looking at an eight amino acid stretch on the channel, Hong said that her results were "more in line with the crystal structure paper" published by DeGrado's group. She also noted that the mechanism proposed by Schnell and Chou implies that amantadine is binding on the inside of the channel - which is, though not impossible, not the most parsimonious mechanism either.

Which of the papers' models the channel in the most realistic way depends on what you look at.

Cady and Hong studied the channel embedded within a lipid bilayer, which, she said, is "closest to the cell membrane."

On the other hand, both Hong's and DeGrado's groups used shorter peptides than Schnell and Chou, and Chou said that those shorter peptides are not able to accurately model all states of the channel.

Not surprisingly, the different binding mechanisms inferred would have different implications for what sort of drugs to design next. DeGrado noted that his group had identified a region near the M2 channel that appears to have a very low mutation rate.

"That would be a very good spot to address future efforts in drug design," DeGrado said.

Chou had different suggestions for designing new drugs targeting the M2 channel. Amantadine is "like a molecular wedge, if you will," wedging the channel closed - the virus, in turn, reacts by mutating the binding pocket so that the channel is more likely to be open.

His proposed solution? "Either design molecules that can wedge the gate better, or design molecules that can go in and block the channel" - which in DeGrado's group's view, amantadine already does.