Sugar On The Brain Circuit: Mice Seeking Sweets May Hold Key To Compulsive Overeating

You know the feeling: you're tired, cranky, low or just have a serious, relentless desire for something sweet. Part of your brain cries out, "No, don't do it, this will end badly." But another (louder) part wants what it wants and won't let up until that pint of Cherry Garcia, or red velvet cupcake or Caramel Macchiato is in plain sight. It's an itch that must be scratched.

Now, brain scientists at MIT say they've identified a specific neural circuit in mice that can increase that compulsive overeating of sweets, but doesn't interfere with normal eating patterns necessary for survival. More specifically, turning on this set of neurons drove mice to seek the reward of a sugary drink even in the face of punishment (a shock to the foot); and compelled them to eat voraciously even when full.  When the researchers shut down this pathway, however, the compulsive sucrose-seeking decreased.

Why does this matter? The new research, published in the journal Cell, may ultimately provide a target for the treatment of compulsive overeating and sugar addiction in humans, without undermining the clearly critical drive of eating to live, the scientists say.

"Imagine if I told you that in the future, we could change the way our neural circuits communicate in a way that I did not want to binge on sweets, but still allowed me to eat healthy foods when I’m hungry?" says Kay Tye, the study's senior author and an assistant professor in the Department of Brain & Cognitive Sciences at MIT. "Obviously there is a ton of work that needs to be done to make this vision a reality, but our study suggests that it is possible."

A Binge-Free Future?

As obesity rates have spiked in recent decades, experts say that overeating in general and consuming too much sugar in particular are major threats to human health.

But Tye says "the real underlying problems are the cravings that lead to compulsive eating, and the behavior of compulsive overeating itself."

To tease out what might be driving that compulsion, Tye looked to a particular set of neurons in the mouse brain.

She and her colleagues showed that when mice perform reward-seeking actions enough that they become habits, that activates neurons connecting two key areas: a brain region called the lateral hypothalamus (an area important for hunger, feeding and homeostasis) and the ventral tegmental area (a brain region important for motivation and reward).

"If we want to understand how the brain gives rise to these feelings, thoughts and actions, we need to know more than what they are saying, we need to know who they are talking to," Tye said. The team used so-called "optogenetic projection-defined phototagging" [essentially using laser light to activate or silence neurons] to see "which neurons...were saying what...and who they were talking to..."

These neural communications are quite distinct, Tye said; for instance, it's important to distinguish between two types of reward-seeking behavior: binge-eating and drug addiction: "You don't need cocaine to survive, you need food to survive," she said.

The "Wanting" Neurons

Tye says that one of the biggest challenges with treating the obesity that comes from compulsive overeating disorders is that "most treatments are just a band-aid — treating the symptoms instead of the core problems.  Gastric bypass for example, is something that just makes it harder to eat, it doesn’t always change a person's habits and eventually many people relapse and regain the weight." Again, she theorizes that it's the craving embedded in the brain that drives the compulsive behavior. She says there may be a distinctive set of  "wanting neurons" as opposed to "liking neurons."

Tye offers more context:

The treatment of food addiction has to be more delicate because you want to shut down the compulsive overeating, but you need to keep the desire to eat healthy food to survive intact.  Our study suggests this is possible.... we hope that our work inspires more effort in developing circuit-based treatments using deep brain stimulation or non-invasive methods like focal ultrasound or transcranial magnetic stimulation, perhaps in combination with cognitive behavioral therapy.  So far our study tells us that this type of treatment is possible because they are different circuits, and it gives us a starting point of where to look."

How About A Milk Shake?

MIT graduate student Edward Nieh, the study's lead author, offers more of the technical nitty gritty in an email:

We used a novel viral technique to identify two populations of neurons based on how they are connected with each other and discovered that they encode different parts of reward.

The first population, which are LH [lateral hypothalamic] neurons that send information to the VTA [ventral tegmental area], are activated when the mouse approaches the reward port where it can drink a sugar reward. In a human, this would be like these cells firing when you approach the counter at a fast food restaurant to get your milk shake, i.e. “reward seeking”.

A second population, which are LH neurons that receive information from the VTA, are activated by a cue that predicts the sugar delivery and the sugar itself. In a human, this would be like these cells firing when you see the sign for the restaurant (e.g. a big golden arch) and drinking the milk shake itself.

So the first thing we did was characterize in detail how this part of the brain processes reward. Once we knew how this part of the brain worked, we could move on to changing how it functions and manipulating animals’ behavior.

What would happen if we could artificially activate or silence these cells that fire when you approach the counter to get your milk shake? Could we make an animal more or less likely to seek a reward?

To study reward seeking behavior, we used a task where the mouse was placed in a chamber in which it had to cross a shock floor giving mild electric foot shocks in order to obtain a sugar reward. Indeed, when we activated the LH-VTA pathway, the mice were willing to endure more shock in order to obtain the sugar reward. In contrast, when we turned the signaling in this pathway off, the mice were willing to endure less shock for their reward. Importantly, turning off the signaling in this pathway didn’t affect normal feeding behavior.

It's Complicated

I asked Mark Andermann, an assistant professor at Harvard Medical School and Beth Israel Deaconess Medical Center, in Boston, (and not involved with this research) to offer an assessment of the work.

He emailed me a long, thoughtful answer which I can sum up this way: The research is tantalizing, but much more must be done before it can be deemed a slam dunk.

Some of Andermann's comments, via email, are here:

Motivational circuitry is complex...which likely reflects the fact that motivated decisions are almost always dependent on many contexts so that we can meet current needs with minimal risk to safety, as well as anticipation of future needs that might be met by taking action.

...The problem, historically, in studying this circuitry is that it wasn't possible to record from specific groups of neurons within an area, or to stimulate or silence these specific groups. The current studies in Cell are using ever-more-refined methods that involve recording from, and turning on or off of, different genetically-labeled cell types within motivation-related areas deep within the brain. In this way, these studies provide key insights into how to carve up individual areas involved in motivation into specific subtypes of neurons with increasingly specific contributions to a given behavior.

In general, this type of 'circuit-mapping' approach is appealing in the context of developing more specific therapies for brain disorders. For example, if neighboring groups neurons either facilitate or deter a motivated behavior, you'd want to figure out the signature of each group, and then target your drug or other therapy to one group but not the other, rather than stepping on the gas and the brake at the same time.

In terms of developing rational therapies for compulsive eating, the field likely has a long way to go. While the authors argue for a specific, causal role of specific lateral hypothalamus neurons in compulsive sucrose seeking, they also note that much of the data is consistent with "increases in motivational drive of the urge to seek appetitive reinforcers". More experiments will be required to tease apart whether different areas involved in motivated behavior can be fully separated into compulsive (eating sweet foods even when sated) and need-based (eating when hungry) circuits.

Further, the lateral hypothalamus has classically been viewed as an area that is involved in the approach to the object of a motivation. Thus, the contribution of the lateral hypothalamus to the "urge to seek appetitive reinforcers" could be quite general to whatever appealing object is currently available (e.g. a sugar treat, a mate, a friend, etc.), which might make it difficult to target with therapies for overeating while avoiding a blunting of other healthier 'compulsive' behaviors (such as social behavior). This work provides the impetus for future studies to investigate the degree of specificity of these neurons to seeking of food vs. other rewards. Regardless, the findings of Tye and colleagues provide a fascinating new window on how two brain areas critical for coordinating motivated behavior communicate with each other."

In a related paper also published in Cell, researchers led by Garret Stuber of the University of North Carolina School of Medicine, also used "an optogenetic approach in mice to identify neurons in the lateral hypothalamus that control both feeding and reward-seeking behavior. By imaging the activity of hundreds of individual lateral hypothalamus neurons as the mice freely explored an area with food or worked to obtain a sweet reward, they further uncovered distinct subsets of neurons that either mediate food-seeking behavior or respond to reward consumption," says the news release.

Don't forget to watch the video, in which researchers evoke some serious gnawing behavior by a mouse. "As soon as we activate the inhibitory component of the pathway, the animal starts to lick the floor," notes MIT's Nieh. "And at one point in the video, you can actually see the animal make the motions of picking up an object and gnawing on it, when no object is actually present."