Bacterial Toxin Surprises Scientists

Release Date:
Alisa Zapp Machalek, NIGMS
(301) 496-7301

Toxic proteins in black widow spider venom and some sea anemone toxins share the same strategy--they punch huge holes in the cell membranes of their victims. Pharmaceutical companies harness this strategy in the potent antifungal medications amphotericin and nystatin. Yet until recently, researchers have known little about how such membrane-piercing proteins work.

Now, using an innovative technique, NIH-funded researchers at New York City's Albert Einstein College of Medicine are closing in on how one such protein operates. Their work may shed light not only on the deadly power of certain toxins, but on diseases like cystic fibrosis that result from defects in natural membrane channels.

"If we could understand how this protein works, we might understand more about how proteins are normally moved across and inserted into membranes," said the study's lead scientist, Dr. Alan Finkelstein, a professor in the Department of Physiology and Biophysics and the Department of Neuroscience. "Why do we care about that? Because it is relevant to all the proteins that end up inserted in the cell's membrane [such as hormone receptors] as well as to any kind of excreted protein [like digestive enzymes]."

In a paper published in the March 1996 issue of the Journal of General Physiology, the researchers mapped out a rough structure of two conformations of colicin Ia, a toxin produced by some strains of E. coli bacteria to kill competing bacteria. From their previous work, the scientists knew that a portion of colicin Ia changes from a harmless cluster of about 175 amino acids into a lethal structure that harpoons the inner membrane of its bacterial victim. The resulting hole, or channel, allows the entrance of foreign particles, the escape of intracellular components, and, perhaps most devastating, loss of the vital electrical potential stored in the membrane. The researchers determined the parts of the protein that lie on each side of the membrane when colicin Ia is in its innocuous, closed-channel state and when it is in its harmful, open-channel state (see images).

Closed Image

Model of the colicin Ia protein channel in its closed and open states.

The protein is denoted by a thick black line, with numbers indicating specific amino acids. The membrane is denoted by the round balls with squiggly tails. The "cis" side of the membrane corresponds to the "outside" of a bacterial victim.

In the closed state, one end of the protein (containing amino acid 594) is anchored in the membrane. The rest of the protein lies harmlessly outside the membrane.

In the open state, a segment of protein from amino acids 541 to 474 is translocated into and across the membrane.

Open State Image

The scientists discovered that to form an open channel, an enormous chunk of colicin Ia--about 70 amino acids--thrusts into and through the victim's inner membrane. Most scientists thought that similar channels, which respond to voltage changes in the membrane, open and close due to a subtle rearrangement of a few amino acids. Never before has anyone seen such a large portion of protein cross to the opposite side of a membrane to open a channel. This protein movement is especially surprising because it requires a long stretch of water-soluble amino acids to cross a membrane made up of water-repelling lipids.

"This result is so counterintuitive that we've spent a lot of time documenting exactly what's going on to make sure we're not dealing with some massive hallucination," said Dr. Finkelstein.

It may eventually be possible to take advantage of this large protein movement as a transport device for the delivery of drugs into human cells, Dr. Finkelstein said. But the researchers are not yet pursuing such applications. Instead, they're trying to figure out exactly how the channel forms and what regulates it.

For example, although one end of the protein is already anchored to the membrane by two transmembrane stretches (see images), the protein is able to form an open channel by transporting just one more large stretch of amino acids across the membrane. This is unexpected because most protein channels thread themselves across the membrane seven or more times. The mystery has led some researchers to consider a hypothesis previously thought unlikely--that the channel walls are not made exclusively of protein, but contain parts of the membrane.

Despite their efforts, the researchers also remain baffled by exactly which parts of the protein are responsible for triggering channel opening and closing.

As with much scientific discovery, the study's findings raise more questions than they answer. But as they continue to study colicin Ia, the researchers hope they will improve general understanding of the critical interactions between proteins and membranes.

Please acknowledge partial funding for this work from the National Institute of General Medical Sciences (NIGMS), a component of the National Institutes of Health that supports basic, non-disease-targeted research.


The researchers developed a method to determine the parts of colicin Ia that are on each side of an artificial membrane in the channel's open and closed states. They created a family of labeled proteins by changing one specific amino acid in each protein to a cysteine (colicin Ia has no naturally occurring cysteines). They then attached the vitamin biotin to the cysteines. Using changes in voltage, they stimulated the protein to open or close. By adding strepavidin--which binds strongly to biotin--to one side of the membrane, they were able to determine whether the biotin-labeled residue was on the cis ("outside") or trans ("inside") of the membrane.



Qiu X-Q et al. Major Transmembrane Movement Associated with Colicin Ia Channel Gating. J. Gen. Physiol. 1996;107:313-28.
Cohen FS. Voltage Drives a Long Segment of Polar Residues of Colicins across Membranes (Commentary). J. Gen. Physiol. 1996;107:307-10.


All available at (718) 430-3169
Dr. Alan Finkelstein
Dr. Karen Jakes
Dr. Paul Kienker
Dr. Stephen Slatin
Dr. Xiao-Qing Qiu

To contact scientists available for comment, call the NIGMS Office of Communications and Public Liaison at (301) 496-7301.