Gut-brain dopamine pathway drives learned food preferences over time

If you can remember your first sip of coffee, the memory might not be a pleasant one. Coffee's bitter flavor is one we naturally steer clear of. So how do coffee fans learn to love it? 

The answer may lie in an understudied "slow" dopamine pathway connecting the gut and brain. A new study from UC San Francisco shows that this pathway helps the brain connect a flavor with the beneficial effects that follow ingesting it. 

Scientists have long suspected that dopamine plays a role in this type of learning. But most dopamine research has focused on fast signals that are triggered in seconds by events outside the body, like seeing a favorite food. 

The new study, published Nov. 4 in Nature, investigates a much slower dopamine pathway that responds to signals that come from within the body. 

We've shown that this slow dopamine system is what drives us to learn to like certain foods over time, and we can see where and how that happens."

Zachary Knight, PhD, Professor of Physiology and senior author of the study

Once your brain has learned to associate the flavor of your favorite brew with the uplift from caffeine, Knight said, "that coffee starts to taste a lot better."

The slow dopamine route from gut to brain 

Dopamine has been intensely studied because it helps guide our behavior by teaching us what is worth pursuing. Scientists know a great deal about its fast-acting pathways. 

But the slower side to dopamine is much more of a mystery, and we are much less conscious of it in our daily lives, Knight said. Sensors in the digestive system detect nutrients such as sugar and fat and send signals toward the brain. The resulting dopamine signal builds after food reaches the gut and can persist for many minutes. 

To trace that signal, Knight and James Grove, PhD, who was then a graduate student, delivered nutrients, either sugar water or a milkshake-like liquid, directly into the stomachs of mice while measuring dopamine release across the brain. Bypassing the mouth allowed them to isolate the brain's response to nutrients reaching the gut, taking the effects of taste out of the equation.

The nutrients triggered a strong and sustained release of dopamine in the anterior basolateral amygdala, or BLA. This area of the amygdala was known to release dopamine, but what that dopamine was doing was unknown. 

"We could see that this dopamine is part of a different system and begin to understand its role," he said. 

Next, the researchers tested whether the pathway actually controls learning. When they blocked it, mice continued eating normally but could no longer learn that a particular new flavor predicted calories. 

Then they reversed the experiment. When the scientists artificially stimulated dopamine fibers in the BLA after mice tasted a calorie-free flavor, the animals learned to prefer it. The stimulation had effectively substituted for nutrients. 

Human brain imaging by Knight's collaborators showed a similar pattern. 

Yale psychologist Dana Small, PhD, reanalyzed an earlier PET study of 13 men who consumed either a milkshake or a noncaloric drink and found that dopamine release concentrated in the left side of the BLA after the nutrients reached the gut. In a separate experiment, Marc Tittgemeyer, PhD, and colleagues used fMRI in 38 adults and found increased activity in the same region after they consumed glucose. 

In both mice and humans, the researchers saw that nutrients produced a much stronger dopamine response in the left BLA than in the right. This uneven response was unexpected, Knight said, because neural circuits controlling basic functions such as feeding are generally thought to operate symmetrically across the two sides of the brain. 

Why the system favors the left side is a question that Knight's Lab is currently trying to answer. 

Could food preferences be fine-tuned? 

Mapping the pathway gives scientists new places to investigate how food preferences are learned - and perhaps, eventually, how they might be modified. 

One intriguing question involves GLP-1 drugs. These medications mimic a hormone produced in the gut and can dramatically change eating behavior. Knight's team did not study GLP-1 drugs, and whether they affect the newly identified pathway is unknown. 

But Knight speculates that the drugs could potentially reduce nutrient-triggered dopamine signaling, making food less reinforcing over time. 

That raises a broader possibility: Rather than simply affecting how hungry or full someone feels, future treatments might someday influence the underlying learning process that makes some foods particularly desirable. 

"By understanding the steps between nutrient detection in the gut and learning in the brain, it might be possible to fine-tune them to subtly change our preferences for different foods," Knight said.