Loading ...

Timothy Ryan, PhD, runs a lab at Weill Cornell Medicine in New York City, where the team focuses on studying synapses and adenosine triphosphate (ATP). Like most teams, they kick around lots of what-ifs. One day, a postdoctoral researcher, Mukesh Kumar, PhD, asked Ryan if fat could serve as a fuel to run a synapse.

Other parts of the body break down fat to make ATP, so why doesn’t the brain? After all, the brain is nearly 60% fat.

“My immediate reaction, honestly, was just that I had a bias because every review article says it doesn’t look like it does that,” Ryan said.

Well, guess what.

In their new study in Nature Metabolism, Ryan and his team showed that a triglyceride-filled lipid droplet in the synapse of a mouse is broken down by neurons into fatty acids, which are sent to mitochondria to produce ATP.

The process is similar to how muscle in other parts of the body uses fat to make ATP. It’s a somewhat hidden process because it does not occur when at rest (lipid droplets are present before exercise but not afterward).

Like most discoveries, it began with the question. But the path to the discovery meant delving into decades of dogma.

Ryan initially told Kumar it’s going to be a pretty high bar to prove the brain uses fat as fuel.

“But then I started reading the background literature of where it came from,” said Ryan. “And it really started in 1933.”

Nearly 100 Years of Dogma On the Line

Back then, the foundational research demonstrated respiration in muscle tissue when given fat for fuel, but the same wasn’t apparent when the researchers tried brain tissue.

“The problem is that when you chop up brain, you make a really nonfunctional tissue. The cells can’t do anything anymore. I’m not blaming them. In 1933, we didn’t have that level of sophistication,” Ryan said.

But decades of subsequent laboratory approaches also used brain tissue and increasingly sophisticated techniques — looking for fat-fueled brain tissue to produce radioactive CO₂ or examining processes at the cellular level.

Another reason it seemed unlikely that the brain metabolizes fat is because electron microscopy of brain cells has never shown lipid droplets. Combined with metabolism experiments, the case seemed settled — other parts of the body metabolize fat but the brain didn’t seem to do it in lab experiments, and lipid droplets didn’t appear on high resolution images.

Neurologist John K. Fink, MD, said he wasn’t shocked by Ryan’s findings “because I don’t tightly hold to the dogma that neurons cannot utilize triglycerides as a source of energy. Are other people shocked? Only to the degree that they hold tightly to dogma.”

“This exposes not so much new areas of neurology but how cautious we have to be about overzealous acceptance,” said Fink, a professor at the University of Michigan Medical School, Ann Arbor, Michigan, who also practices in the area of spastic paraplegia. These latest findings, Fink said, offer an important basis for further exploring brain function in low-glucose conditions.

“Now that we know that it can happen, we need to know to what extent does it happen,” he said.

What Does It Mean to Not See Something?

The question of does the brain use fat persisted in Ryan’s lab, which is well-known for its 2014 work demonstrating that synapses don’t store ATP and it’s produced on demand. The on-demand nature is important — when ATP can’t be made on demand, things start to go wrong.

“Keeping synapses working is the most important thing in the brain. I say that as a person who studies synapses for a living, but I don’t think anyone would argue that that’s the business end of the brain. It’s synaptic communication,” said Ryan. “That was our motivation — to learn all the rules we can about what controls how well we do this job.”

A primary question that needed to be answered in that exploration was: What fuel is used to make ATP?

Glucose is the standard, Ryan acknowledged, but “the brain doesn’t tolerate not having enough fuel because it needs to use it all the time. And so that brought up the question: What other things might you use?”

Ketones were something they considered, partly driven by the fact that a severe ketogenic diet is sometimes effective as a last recourse for treating children with epilepsy. The link between ketones and the brain is unclear.

“Ketones are made probably in the liver and get shipped to the brain, and neurons use the ketones. So we have studied this, but we don’t quite understand why that ends up resetting things well,” Ryan said.

Another fuel the brain may use is fat, the team hypothesized. Muscle also needs on-demand fuel and uses glucose.

“But muscle famously is also well-known to use fat,” Ryan said, noting that it’s a very now-you-see-it, now-you-don’t process.

“Actually, if you look at resting muscle in a biopsy, you discover that resting muscle has lipid droplets,” he said. “So if you look at fat cells, which we all have — and we all have more and more of as we age — those cells are full of big lipid droplets. They’re giant. That’s how fat is stored, in lipid droplets.”

The lipid droplets are perfect spheres. But after exercise, most lipid droplets are gone because they were used as fuel to make ATP.

“So what does it mean to not see lipid droplets in the brain? It could well mean that part of the brain is resting,” Ryan said.

What if the Brain Stops Burning Fat?

Ryan and the team also leveraged understanding of genetic susceptibilities for diseases like Parkinson’s disease and spastic paraplegia using existing research about how mutations affect enzymes involved in lipid metabolism. A DDHD2 mutation — which is linked to hereditary spastic paraplegia type 54 — can drive a huge buildup of lipid droplets in the brain, as shown in previous work by Benjamin Cravatt, PhD, of Scripps Research in La Jolla, California.

“We knew that the neurons were accumulating lipid droplets, but that’s where we left it,” said Cravatt, a chemical biologist.

The implication, Ryan said, is that if the brain doesn’t have the enzyme made by DDHD2, it can’t take apart lipid droplets to make fatty acids, which can’t then be given to mitochondria to make ATP.

“Luckily for us, Cravatt also made — he’s a chemist — a small molecule inhibitor, so we didn’t just have to rely on knockout mice. We could do really fast experiments, where we could block this enzyme with a drug that he made…in 10 hours, you could see a huge buildup of lipid droplets in the neurons.”

Turns out the inhibitor Cravatt made was reversible.

“That allowed us to pull all the tricks my lab is good at and to show that now you could run a synapse with absolutely no glucose whatsoever,” Ryan said. “If you would build up lipid droplets, you could skate by as if nothing was wrong without any fuel — like if you did that, you would be insulin resistant because it wouldn’t matter if your glucose drops because for a little while, you could run a synapse without any fuel whatsoever because it had its fuel depot.”

It’s a possibility, Ryan said, that as we age, we become more reliant on this backup fuel source for the brain, posing potential areas of future research that intersect with dementia and other neurologic diseases.

Cravatt called Ryan’s work “impressive.”

“What we’re seeing in the DDHD2-disrupted humans that get spasmodic hyperplasia is obviously an extreme nonphysiologic outcome of losing the enzyme,” Cravatt said. “But what does the enzyme normally do? And is it normally relevant? Is a triglyceride hydrolase activity relevant for brain energetics, brain signaling? This paper really speaks to that quite well — there appears to be an intriguing role for lipid metabolism as a source of energy and fuel in neurons normally as part of their signaling.”