In Surprise, Research Suggests Lipid Bilayer Uninvolved, Challenging Prevalent Theory
General anesthesia at clinically relevant concentrations induces unconsciousness by changing the function of proteins on the cell membrane, not the lipid bilayer, as previously believed, researchers have concluded. The findings challenge a century-old concept of how anesthetics work, and may ultimately help guide the development of novel, highly targeted anesthetic agents.
“Our lab is keenly interested in how inhaled anesthetics work,” said first author Karl Herold, MD, PhD, a senior research associate in the laboratory of Hugh C. Hemmings Jr., MD, who is the senior author of the paper (Proc Natl Acad Sci U S A 2017;114:3109-3114) and chair of anesthesiology at NewYork-Presbyterian/Weill Cornell Medicine, in New York City. “We’ve been using anesthetics for quite some time now, and it’s not really known how they work, which is interesting, scary and fascinating. We know how much to give and what happens when we give them, but we really don’t know how they work on a molecular level.”
Indeed, since the first successful public demonstration of ether to induce surgical unconsciousness in 1846, anesthesiologists have been trying to understand how general anesthetics exert their clinical effects. In the 19th century, two pharmacologists hypothesized that anesthetic potency correlates with drug solubility in lipids. Their suspicions were supported by experiments in which they dissolved anesthetic agents in olive oil, and their findings produced the prevailing scientific explanation for anesthesia that has held sway for over a century: General anesthetics work by altering lipid bilayer properties and disrupting neuronal function, leading to unconsciousness.
A Controversy Develops
“In the 1970s, scientists started probing deeper into possible targets of general anesthesia,” Dr. Herold said. When researchers began to suggest that proteins were the targets of general anesthesia, a major controversy began.
“Over time, more and more studies suggested that anesthetics interact directly with proteins, in particular with membrane proteins on the surface of the cell, such as ion channels, which are important for cell-to-cell communication,” Dr. Herold said. “This protein theory contradicted the original lipid hypothesis, which stated that anesthetics affect neuronal function by changing properties of the cell membrane itself.”
To determine the biological mechanism behind anesthesia, Dr. Herold and his colleagues tested 13 anesthetic agents using a technique developed by Olaf S. Andersen, MD, professor of physiology and biophysics at Weill Cornell Medical College. Unlike much previous research on lipid bilayers, however, the investigators used clinically relevant drug concentrations. “We wanted to test if anesthetics change the bilayer at concentrations that are relevant in the clinic,” he explained.
The investigators formed a “model cell” composed of a thin lipid membrane in which they placed a reporter protein, known as the gramicidin ion channel. They placed a fluorescent dye inside the model cell, which would change its visual signal when ions entered the model cell through the channel. Because the movement of ions through gramicidin channels is very sensitive to changes in lipid bilayer properties, Dr. Herold and his colleagues were able to measure whether anesthetics alter lipid bilayer properties by monitoring the signal of the fluorescent dye (Figure).
These form a dimer and allow ions to pass through. Using the antibiotic gramicidin, which does not allow ions to pass in the unassembled state (spirals shown on left), a scientist can measure how many ions flow through the formed gramicidin channel (shown on right). Many drugs can affect the rate at which these channels form. In this study, it was shown that the anesthetics did not affect the rate at which these channels formed.
Results a Surprise
They were surprised to find that when using clinically relevant concentrations, none of the anesthetics affected ion movements through the channels, indicating that they did not affect lipid bilayer properties. “We were surprised by the absence of general anesthetic effects on the bilayer,” Dr. Herold said. “We thought they would have some effect, particularly because this is the mechanism that many others have proposed.”
“That was a very surprising result,” Dr. Andersen added. “When we started conducting the experiments, I was convinced we would see some effect on the bilayer. The fact that the results are as clean as they are was to me really amazing.”
The researchers also confirmed their findings with a variety of anesthetic classes beyond inhaled anesthetics. “With ketamine, for example, you can go up to 100 times the dose that you would ever give a human before an effect on the lipid bilayer can be detected,” Dr. Herold said.
Lipid bilayer changes could be documented in the study, but only at dangerously high drug concentrations. “When we got to fourfold clinical concentrations, some drugs did change the bilayer,” Dr. Herold explained. “And this was good to know because it confirmed that we were measuring the changes correctly. What’s more, this might explain why some of these drugs have side effects at very high concentrations.
“In previous experiments,” he said, “we tested several diabetes drugs, some of which were taken off the market because of side effects. These were the ones that changed the bilayer much more strongly than others that have fewer side effects and are still on the market.”
For the Weill Cornell researchers, the findings are the culmination of years of work. “We’re very happy because we’ve worked very hard on this study for the past three years,” Dr. Herold said.
Important for Drug Development
Yet even with such remarkable findings under their collective belts, the researchers have already turned to their next objective: a specific protein target for general anesthetics. “There’s a question regarding what protein is the most important target for anesthetics,” Dr. Herold said. “Some people believe it’s the GABAA receptor, while others say it’s the NMDA [N-methyl-D-aspartate] receptors. We believe that the sodium channel plays a key role, since it initiates and propagates action potentials, which are crucial to communication between cells. If these sodium channels are blocked by a drug, there will be limited communication between cells.”
Developments such as these may help pave the way for pharmaceutical companies to develop safer drugs by providing a screen for potentially undesirable lipid effects. “Drug companies can design drugs that could only lead to a specific protein target without interacting with lipids, making the drugs more specific and with fewer side effects,” Dr. Herold said.
Emad Tajkhorshid, PhD, the J.W. Hastings Professor of Biochemistry in the Beckman Institute for Advanced Science and Technology at the University of Illinois at Urbana-Champaign, agreed that the downstream effects of this research could be significant. “I think these findings are going to focus the research on designing new anesthetics,” Dr. Tajkhorshid said. “Before this research, it was thought that general anesthetics needed to be highly lipid soluble. But now we know they have these additional properties, so we can design and go after molecules that interact with specific channels.” Indeed, in a study published earlier this year (J Biol Chem 2017;292:9480-9492), Dr. Tajkhorshid and his colleagues showed that a membrane-embedded pathway is responsible for delivering general anesthetics to two interacting binding sites in a bacterial ion channel.
Another benefit of this research, Dr. Tajkhorshid added, is the potential to explain and perhaps alter the synergistic effects of other classes of drugs with anesthetics. “There are other drugs that work with the same classes of proteins,” he said. “Therefore, when they both bind to different sites of the same protein, they might be able to potentiate or inhibit each other’s effects.”