Carbon Dioxide “Alarm System” Might Help Explain Anxiety Disorders


by Carl Sherman

March 10, 2010

For most people, a little stale air isn’t much of a problem—a lot of carbon dioxide has to build up before they start to panic. But for some, inhaling even a whiff or two of CO2 can provoke an immediate sense of dread. In fact, a prominent psychiatric theory holds that an overly sensitive detection mechanism for the gas, or “suffocation false alarm,” makes these folks particularly susceptible to panic disorders and other anxiety problems.

Now, nearly a century after this phenomenon was first noted, researchers have discovered the first possible biological mechanism for such a hair-trigger alarm system—an ion channel found in particularly high numbers in the brain’s amygdala. The work, researchers say, could provide insights into causes and possible treatments for many anxiety disorders.

The amygdala is a central structure in the brain that plays a substantial role in processing emotional memories, including those relating to fear and anxiety. The region is known to have particularly high concentrations of ASIC1a, an acid-sensing ion channel found throughout the body in small quantities but most common in the brain.

Scientists know that the channel is vital to the processing of normal fear. But the new research suggests that the amygdala has a much more active role in the process than previously thought, using ASIC1a channels to directly detect internal changes caused by dangerous situations.

Acid response

In particular, carbon dioxide acts like an acid when dissolved in water; this is what gives soft drinks and soda water their characteristic “bite.” The researchers hypothesized that elevated CO2 levels help acidify the bloodstream, activating ASIC1a in the amygdala and in turn causing nerve activity that leads to panic.

To test this, researchers led by Adam E. Ziemann of University of Iowa looked at the responses of both regular and mutant mice to air containing 10 percent CO2. The normal mice showed classic fear behavior, such as freezing up, avoiding open areas, and staying away from the chamber where they had been exposed to the CO2. But mice bred to lack the ASIC1a gene showed no such responses.

Other experiments confirmed the importance of pH. When the researchers injected mice with bicarbonate, a buffer that helps keep pH stable, they froze significantly less when exposed to CO2- rich air. The mutant mice showed no such difference.

 The researchers went on pinpoint the site of the action. When they injected an acidic solution into the amygdala, it induced fear behavior in normal mice, but not in those without ASIC1a. And when the scientists injected a virus that carried the ASIC1a gene directly into the amygdala of the mutant mice—restoring the channel in specific regions of the amygdala but nowhere else in the brain—normal fear responses to CO2 returned.

“It’s intriguing, because the field of psychiatry has for decades hypothesized the existence of  such a CO2 chemosensor but never found it,”  says Kerry Ressler, a neuroscientist at Emory University who was not involved with the new work. “This study not only identifies the sensor, but locates it in the part of the brain most involved in fear-related behavior. It may help us better understand panic disorder and abnormalities underlying it.”

Wider implications

According to University of Iowa neuroscientist John Wemmie, senior author of the paper reporting the research in Cell, the findings support the idea of a suffocation alarm. But they also “raise the possibility” that rather than CO2 itself, “pH dysregulation, or an abnormality in acid-sensing channel function, could be central to the pathophysiology of panic.”

Regardless, the team says, the work may have important implications for both understanding and treating conditions involving increased or unwarranted anxiety, such as post-traumatic stress disorder. In earlier studies, for instance, the team had found that ASIC1a appeared necessary for the normal expression of fear. It didn’t matter if the fear was innate—such as the response to the smell of a predator—or learned, for example, if the animals began to associate the sound of a bell with the risk of an electrical shock.

To test the implications of the new findings, the researchers administered bicarbonate to mice while they underwent fear conditioning. The next day, the conditioned fear response was significantly reduced in these animals, compared to others who hadn’t been pretreated. Bicarbonate also weakened freezing behavior triggered by a chemical that resembles predator odor.

 ASIC1a channels are present at synapses between brain cells, Wemmie says. For this reason, the team suspects that acidic blood not only directly contributes to a fear response but may also alter the synaptic plasticity that leads to longer-term fear memories.

“The notion that the amygdala can sense changes in extracellular pH and translate that into fear is a really interesting discovery,” says Stephen Maren, who researches the neurobiology of learning and memory at the University of Michigan but was not involved with the new research. Mostly, scientists have thought of the amygdala as a receiver of sensory information from visual and auditory systems that arouse fear, but “this represents a new way that the amygdala can respond to stress, by directly detecting internal states.” He adds, “I don’t think ASIC is essential for fear conditioning, but the system seems to be modulatory—it might lower the threshold.”

The idea that a system that warns of suffocation might underlie more general anxiety responses is conjectural but makes sense, he says. “Mammals are under constant oxygen threat—if you can’t breathe, it signals imminent death—and the notion that the response has been co-opted by other mechanisms to use in the service of learning fears” is “fun to speculate about.”

The researchers “without a doubt showed that acid regulation is important for all sorts of amygdala functioning,” Ressler adds. “They were able to show that systemic bicarbonate blocks fear conditioning, but I’m not sure what to make of that. Context is a complex thing,” and factors other than acidity might be involved. “I’d like to look in a little more detail at how much pH actually regulates fear memory.”

Maren agrees that it would be “interesting” to test whether anxiety-provoking stimuli can, like CO2, change amygdala pH enough to activate ion channels. “If any aversive experience, whether a yelling boss or public speaking or a car accident can do it, it broadens quite a bit how we think about the synaptic physiology underlying plasticity and learning.”

Despite the work that remains to be done, the research “has a lot of potentially clinically relevant implications,” Ressler says. The idea of manipulating pH to enhance anxiety treatment merits exploration, he adds, and the ASIC mechanism itself could be a worthy focus. “If it’s shown that blocking the acid-sensing channel decreases fear expression, a small-molecule ASIC1a- specific antagonist might give the pharmaceutical field a new target to aim for.”

Maren suggests that ASIC research might also improve psychotherapeutic interventions, such as exposure therapy, that work by extinguishing conditioned responses. “The same receptors in the amygdala that are important for fear learning are also involved in fear suppression,” he says. “I’d be curious to know if ASIC is involved in that process as well, whether it plays a complementary role.”