The process is called flipping, but in this case, it's not selling a stock or real estate rapidly at a high profit, it's a method of getting things into cells.

It calls for attaching whatever is desired to a helical protein that will hide in the nearest lipid bilayer when the going gets tough - or rather, acidic.

The protein, which its discoverers have named pHLIP, for pH (low) insertion peptide, is neither a helix nor particularly interested in inserting itself into membranes when the environmental pH is neutral or basic. But as the researchers described in the April 25, 2006, issue of the Proceedings of the National Academy of Sciences, under acidic conditions, pHLIP "becomes rigid (as a syringe needle) when it inserts into a lipid bilayer, and it forms a transmembrane helix and injects molecules into cells."

Donald Engelman, a professor of molecular biochemistry and biophysics at Yale, told BioWorld Today that the method is "an absolutely accidental discovery," made in the course of basic research on how membrane proteins fold. The researchers had been studying bacterial protein variants to see how they inserted into cell membranes and found that one of them had the unusual property of having only weak interactions with lipid membranes at neutral pH, but inserting itself into the membrane at more acidic pH, which, as it happens, is a feature of some disease states.

Engelman said it was his former post-doc, Yana Reshetnyak, a co-author on the paper who is now an assistant professor at the University of Rhode Island, who followed up with the idea of using that peptide, which was derived from the bacterial protein rhodopsin C, for drug discovery. That follow-up included figuring out which end of the peptide goes inside the cell during insertion, and adding a cysteine residue to which drug payloads could be added that would be delivered across membranes and then cut once inside a cell to create pHLIP.

The researchers first excluded the possibility that pHLIP was delivering its payload via other mechanisms: "It works in pure lipid vehicles, so it can't be receptor-mediated," Engelman said. "And it works at four degrees, which means it is basically temperature-independent," which also excludes endocytosis as a possible mechanism.

In their PNAS paper, the researchers described the results of adding several types of payload to the pHLIP helix. First, they attached peptide nucleic acids (PNA); relatives of DNA and RNA, PNA has a protein-like polypeptide backbone. "But instead of amino acids, you stick on bases," Engelman said. The bases are more stable than DNA and RNA but are able to interact with intracellular gene and RNA sequences.

"They can be used, and have been used, to turn genes on and off," he said, but "PNA technology is not new. The fact that we can put them into cells in a nontoxic and controlled way is." Other strategies, such as electroporation, are "much more damaging to cells."

The scientists also tested whether pHLIP could transfer the fungal toxin phalloidin into cells; a potent cell toxin, phalloidin can, to date, be delivered into cells only by permeabilizing their membrane. The scientists found that they could "flip" phalloidin into cells, where it promptly interacted with the cytoskeleton in ways that suggested it enhanced apoptosis. Engelman said that "already, the vehicle we have is a potential therapeutic," but he added that phalloidin is certainly far from ready for the clinic.

In general, Engelman said that the pHLIP helix can deliver "fairly polar compounds" with a molecular weight of up to 1,000 into cells - a group that "embraces most drugs."

Engelman, Reshetnyak and their colleagues have preliminary animal data showing that in the short term, at least, the peptide is not toxic to mice. They also have shown that it accumulates in tumors in a mouse model, because tumors, like a number of other diseases, promote acidic conditions within a cell. Animal studies are ongoing, as is more basic research into pHLIP's properties.

"Ultimately, the design [of the carrier] might not be a peptide," Engelman said. "We think of this as a lead rather than a candidate."