Why don’t hummingbirds get diabetes?

Question: why don’t hummingbirds get diabetes? 

That is, hummingbirds drink straight sugar water all day, every day. Any other animal who tried to do that would get insulin resistance and diabetes (yes, even birds). So why not hummingbirds?

I have an answer for you, but we’ll need to do a refresher first. And then the answer will reveal a mystery. You will be left hungry for more, much like someone with, well…some disease. Ready? Let’s begin!

Refresher first: when you’re hungry and you eat a hamburger, your body can’t use the hamburger directly for fuel. Instead, it has to convert the calories in the hamburger to usable fuel. Usually, this is through converting the calories to glucose through gluconeogenesis, which ends up being what we call “blood sugar”.

Once we have blood sugar, our body can use it immediately for fuel through glycolysis, store it short-term as glycogen in the liver and skeletal muscles, or store it long-term as fat1. Insulin is the main hormone that determines what happens to blood sugar, and insulin resistance (i.e. poor response to the insulin signal), caused by high levels of blood sugar for extended period of time, is what leads to type 2 diabetes. I explored the weird relationship between insulin resistance and weight loss in a previous post2.

If the glucose does go through glycolysis, the product of glycolysis, pyruvate, is then used for aerobic respiration in the mitochondria, at least in mitochondrion-carrying organisms like people. This then eventually forms ATP, which is the final form of immediate, short-term energy.

Alright, done with the refresher. Add in a bunch more detail and make you memorize all the intermediate steps, and you’d be ready for the AP Biology exam. But, that stuff isn’t particularly fun. Let’s talk about the fun stuff: the mystery.

Glycolysis is upstream of aerobic respiration. So, what would happen if we do a ton of glycolysis? Well, you might think, we’re going to have a ton of pyruvate and we’re going to do a ton of aerobic respiration.

And that’s exactly what happens! In a neat paper from this year, scientists took a look at hummingbirds to try to answer our original question, “Why don’t hummingbirds get diabetes, given how much sugar they’re consuming?”

As these scientists found out, it’s because hummingbirds don’t do gluconeogenesis, only glycolysis. That is, they never make glucose; they just use it. Hummingbirds can get away with this because they can do glycolysis directly from fructose, unlike any other animal that we know of.

Hummingbirds don’t do gluconeogenesis because, over the course of evolution, they lost fructose-bisphosphatase 2 (FBP-2), which is necessary for gluconeogenesis. In vitro, when these same scientists knocked out FBP2, they found gluconeogenesis went down, glycolysis went up, and mitochondrial respiration went up. So far, so good!

They also found something a bit more interesting: FBP2 knockdown resulted in mitochondrial biogenesis (i.e. an increase in the number of mitochondria). This isn’t shocking in light of the whole increase in mitochondrial respiration thing, but it is interesting for me. I’m used to thinking about mitochondria in the context of neurodegeneration, where losing mitochondria is a really bad thing, and spinning up and down mitochondria at the metaphorical drop of a hat is strange to think about.

Still, not crazy. FBP-2 knockout causes loss of gluconeogenesis, increase in glycolysis, increase in mitochondria, increase in mitochondrial respiration in hummingbirds observationally, and in vitro experimentally. Cool. I wonder what happens if we do the opposite? What if we upregulate FBP-2?

Well, then we increase gluconeogenesis, inhibit glycolysis, mitochondrial biogenesis, and mitochondrial respiration, at least in cancer cells. Cool, that works. This is all making sense. So, what if we just instead go ahead and inhibit glycolysis directly? That shouldn’t make a difference.

And…shit. Direct inhibition of glycolysis during blood vessel formation via siRNA increases mitochondrial respiration (and vice versa). That’s awkward.

Now, this isn’t inexplicable. In fact, the authors have a great explanation: cells need alternative pathways to produce energy. If there’s no glycolysis, cells still need to get energy somehow, so they end up working overtime in less efficient pathways to still get energy. And the authors do show that this has real consequences in that blood vessel formation is way more difficult when glycolysis is inhibited.

But I do have sort of a problem with this. The FBP-2 upregulation explanation makes sense, and the direct inhibition explanation makes sense, but they both do seem difficult to coexist. To put in formal logic terms, we’re saying sometimes A causes B, and sometimes A causes not B, and it’s just the luck of the draw which one is which. It doesn’t seem super scientific.

Now, I should probably point out that these scenarios aren’t exactly comparable. The upregulation of FBP-2 links to a paper about sarcomas that were using the hummingbird’s FBP-2 knockout trick to grow unnaturally fast, while the blood vessel formation paper is talking about a normal physiological process. So, I could unify these two with an analogy like candy bars and weight loss. If someone’s using candy bars (FBP-2 knockout and increase in glycolysis) as a pre weight lifting snack that allows them to lift more and lose more weight (increase mitochondrial respiration), taking away the candy bars will cause them to gain weight (decrease mitochondrial respiration). But, if someone’s using candy bars (increase in glycolysis) in day to day life, taking away the candy bars will cause them to lose weight.

But, as this analogy suggests, there is a missing step somewhere. Even though we thought we could focus on candy bars to explain weight gain or loss, we actually needed to think about the entire energy budget to see what the candy bars were being used for. This is not nearly as easy as just looking at someone’s candy bar consumption and weight loss. We went from two numbers that can be taken at a specific point in time to monitoring someone throughout their day or week.  But, if this analogy holds, then in some sense, this is apparently how the cell is “thinking” about it. There’s some sense in which a cancer cell or a blood vessel knows what energy it needs and can take alternative paths to get it if one path is blocked. Cancer cells using FBP-2 tricks to grow really quickly do not think the same about energy as normal cells, and so the way they modulate their mitochondria is different.

And now that I’ve gone thoroughly out on a limb, I might also say that this suggests a way in which cells can mess up in misbudgeting their energy requirements. I don’t mean this in just the sense of cancer, but in the sense of somehow generating too much energy or not enough and then, presumably dying or generating harmful waste products that other surrounding cells will have to deal with. Hell, maybe that’s even at play in some of the neurodegenerative diseases I was discussing earlier.

But, as I mentioned, this is definitely a limb that I am climbing out on. I really don’t know. The mystery remains.


Not an exhaustive list!


In the previous post, I talked about how the relationship between insulin resistance and weight loss confuses me. My confusion was as follows: once you are insulin resistant, then insulin resistance has a pretty instantaneous correlation with weight loss. If you are insulin resistant and you start weight loss, insulin resistance goes away very quickly. If you are insulin resistant and you start gaining weight again, insulin resistance comes back very quickly. This all occurs even though the process of gaining weight or losing weight is slow, and even though the initial buildup of insulin resistance with weight gain is very slow.

Even more weirdly, none of the usual suspects that you would want to blame for these quick changes in insulin resistance seem to matter. Specifically, you’d probably want to look at lipolysis, the destruction of fat cells, and lipogenesis, the creation of fat cells, as causes for these rapid changes in insulin resistance. But, both lipolysis and lipogenesis result in decreases in insulin resistance, so it’s not clear why gaining weight back would cause insulin resistance to increase again.