One of the things that you have to get used to in science (and especially biomedical science) is the constant possibility that something that Everybody Knows will turn out to be wrong. Today’s installment is the way that Everybody Knows that the brain is an obligate user of glucose for fuel. You can find this in every textbook; it’s been known for years. That’s why there are so many glucose transporters are up there, why gluconeogenesis kicks in under glucose deprivation conditions, why you don’t see lipid droplets in neuronal cells, and so on.
But the study of a set of rare diseases, the hereditary spastic paraplegias (HSPs) has undermined that confidence. There’s a long list of those, and they’re associated with an equally long list of underlying mutations. But one of them (HSP54) seems to be driven by mutations in an enzyme called DDHD2, which is a lipase. In fact, it’s a neuron-specific triglyceride lipase, which makes you wonder why neurons would need their own pathway for breaking down the triglycerides that they’re not really supposed to be storing or using for much or anything. That question becomes more acute when you look at the neurons of HSP54 patients and find that they are full of (unhydrolyzed) triglyceride droplets.
Now, in any other metabolically active tissue this would be no mystery. But brain tissue has long been thought to be the exception: high metabolic turnover, but not through mitochondrial beta-oxidation of fatty acids. This new paper shows that DDHD2 is very active at synaptic terminals and monitoring ATP production in axons shows that DDHD2 activity is directly upstream of it. Inhibition of it quickly leads to torpor in mice, as does inhibition of fatty acid import into mitochondria, and both of these show accumulation of triglyceride droplets as they take effect. It’s hard to escape the conclusion that at least some neurons do indeed use triglycerides for energy storage and rely on oxidation of the resulting fatty acids for a good part of their energy needs. Indeed pre-loading such neurons with palmitic acid makes them much more resistant to glucose starvation conditions. So the observations that lipid droplets are present in very low abundance in neurons were correct, but that missed the fact that it’s that way because they are constantly turning over and being used for fuel (!) We’re all going to have to adjust our thinking!
I wanted to take a moment, in the light of this work, to conduct a little thought experiment. I’ve been hearing an awful lot recently from various highly placed and deeply uninformed people that AI is going to pretty much solve all our outstanding scientific problems Real Soon Now (there will be another blog post just on that). So let’s ask ourselves: would AI have told us this? Remember, when people say AI they are about 95% saying “machine learning”, so would it really have told us about this after having been trained on years and years of the medical literature telling it that neurons are obligate glucose users and don’t really have a role for triglycerides? Of course not. And this is why I keep saying (and I’m sure not the only one) that we simply don’t know enough to teach the machine learning algorithms yet. Not to mention that some of what we’d be teaching them is just wrong to start with. . .