Why We Smell Like Bugs: A Case Study of How Evolution Sculpts the Brain

In a recent post I explained why it’s awesome to study the brains of invertebrates, like fruit flies or worms. I bet by now you’re convinced that doing experiments in these tiny creatures can teach us lots of things about the fly or worm brain. But what most people care about is the human brain.1 Can invertebrates really teach us anything about what’s going on inside our own heads?

I mentioned earlier how fly and worm brains don’t look anything like mammalian brains. And I don’t just mean because they’re a zillion times smaller—even when you zoom in on them, their shape and structure looks pretty different.

brain comparison

Comparison of mammalian brains (left) and a fruit fly brain (right), which has been stained in red and greatly magnified. (left image credit: Wikimedia Commons)

Nevertheless, invertebrate brains actually don’t function so differently from our own. Brain circuits often have certain organizing principles and motifs that are shared even among diverse organisms.

This makes sense, because all animals, from worms and flies to humans, need to solve most of the same problems. From an evolutionary perspective, the main goals of any species are to survive and reproduce, thereby passing on our genes to future generations. Survival requires activities such as finding a nice place to live, locating food, and not getting eaten by someone else. Reproduction involves things like finding a good mate and caring for your offspring so they don’t get eaten either.

While different species might eat different foods or live in different places, we all still need to accomplish the same basic tasks. So it’s really not surprising that evolution has hit upon some of the same solutions time and time again.

Everyone smells the same

The olfactory system, which mediates our sense of smell, is a great example of how the brains of totally different animals can work in very similar ways. Specifically, the insect and mammalian olfactory systems share an amazingly similar organization even though they’re thought to have evolved independently.

I’m going to spend today’s post discussing the parallels between the insect and mammalian olfactory systems and why they may have evolved in this way. In future posts I’ll discuss similarities between other organisms, like worms and mammals, and in other brain circuits.

Yeah, I know I’ve already talked about smell a couple times, which may partly reflect my bias working in an olfaction lab. But smell is isn’t as lame as you might think, as I’ve discussed before: it’s super important in our lives and even more crucial for many other animals, like rodents or dogs, who use smell as their main sense in exploring the world.

What I haven’t discussed before is how your brain actually interprets what’s going on in your nose, which is the most interesting part. If you think about it, it’s pretty amazing: how can the simple act of chemicals binding to receptors in your nose provoke intense feelings such as hunger, disgust, or nostalgia?

Comparing noses


Antennae of various insects (credit: Chris huh via Wikimedia Commons)

Smell starts in the nose. An insect’s “nose” mainly resides within its antennae, which are the two pointy things that stick out from an insect’s head. You may not think an antenna looks much like your nose, but on the inside they’re pretty similar. No, I don’t mean that an antenna is full of boogers. Both insects and mammals have a bunch of neurons in their noses that detect smells through tiny receptors that recognize odor molecules.

Since there are tons of smells in the world that we have to detect (the exact number is a matter of debate, as I’ve discussed), most animals have many different odor receptors to detect them with—around 60 in flies and hundreds in mammals. But an individual olfactory neuron in the nose actually has only one type of receptor, so it can only detect certain kinds of smells. Different neurons can have different receptors, so if there are 60 types of receptors then there will be 60 different types of neurons. For the sake of redundancy there are multiple neurons with the same type of receptor; these “sister neurons” are scattered all throughout the nose and don’t usually reside near each other.

Already, the parallels between the insect and mammalian nose are rather uncanny. It didn’t need to be the case that each olfactory neuron has only one type of receptor (this isn’t true for taste neurons, for example). And you might not have expected that sister neurons would be dispersed throughout the nose instead of being grouped together. But these features are indeed common to both flies and mammals, suggesting that they’re important for optimizing our sense of smell.

Sorting in the brain

The olfactory neurons in the nose transmit information through long projections called axons, which extend all the way from the nose to the brain. Now, something amazing happens here, in both flies and mammals. Even though sister neurons are scattered throughout nose, they send out axons that somehow cluster together in a nice little bundle once they get to the brain, effectively sorting themselves out from the scores of other axons that are around.

olfactory neuron sorting

Olfactory neuron axon sorting. Circles represent olfactory neurons, and circles of the same color are sister neurons that have the same receptor. Sister neurons are scattered throughout the nose but cluster their axons together in the brain.

Seriously guys, this is super impressive! In mice, for example, there are around 1000 types of olfactory neurons completely interspersed in the nose—but their axons sort themselves out perfectly in the brain, each type separating itself from the 999 other types. This would be like if you had thousands of people each wearing a hat with a random number from 1 to 1000, and we all put our arms in a box and were able to blindly grasp hands with people who had the same number.

Interestingly, this sorting process involves totally different mechanisms in flies and mammals, even though the end result is the same. This is a common finding when you’re comparing the brains of very different animals—the circuits may share a similar organization or function even though the underlying mechanisms are different. Evolution only cares about the end result. In this case, it’s apparently very important that the axons of sister neurons cluster together in the brain.

Why is this so important? Well, we don’t know for sure, but we can guess. Remember that sister neurons detect the same smells (since they have the same type of receptor), whereas neurons of other types detect different smells. So by sorting out the axons, the brain is keeping different kinds of olfactory information separate. This might help your brain keep track of what your nose is smelling. It’s like when you want to keep track of different projects at work: you keep separate folders for each project. Organizing your information makes things more efficient.

How the brain smells

Ok, so now we’ve boldly ventured from the nose into the brain. Here, the little bundles of axons from the nose transmit information to a new set of neurons, called projection neurons, which represent the second step in the olfactory pathway. The projection neurons then send olfactory information to multiple parts of the brain.

projection neurons fly mouse

Projection neurons send information through their axons to multiple higher brain regions. In flies (left), a single projection neuron is shown sending axon branches to two different brain regions (circled). In mice (right), a bundle of projection neurons is shown in pink (bottom image) sending axons to several different regions which are labeled in the top right. (fly image from Rachel Wilson via hhmi.org; mouse image from Sosulski et al., 2011)

In both flies and mammals, these different brain regions serve different functions in olfaction. One of these brain regions controls innate responses to smells, meaning responses that you’re born with and don’t need to learn—like when a mouse is afraid of the smell of a cat, or when the pesky fruit flies in your kitchen are attracted to the banana you left out. A separate brain region controls learned responses to smells: through experience we may come to savor the aroma of freshly brewed coffee or become repulsed by the smell of a tequila shot (it’s not just me, right?).

It’s pretty cool that the olfactory pathway splits up in the same way in both flies and mammals, with separate circuits for innate and learned responses to smell.2 But what’s even cooler is that these parallel brain regions in flies and mammals share a common structural organization.

The brain region controlling innate responses to smell is relatively well-organized. The information that it receives from projection neurons is structured logically: different types of projection neurons send information to different parts of this region, while projection neurons carrying similar information stick together.

In contrast, the brain region controlling learned responses is super disorganized, with information from the projection neurons getting completely jumbled. This seems weird—you’d think mixing everything up would make it harder for the brain to keep track of which smell is which. What’s going on? Did evolution just screw up here?

Relating structure and function

The answer to that question is almost always no: evolution didn’t just screw up, because if it screwed up then you’d be dead. Plus the fact that things are so similar in insects and mammals suggests that there’s a reason why olfactory information is logically organized in some brain regions and completely jumbled in others.

Let’s start with the brain region for innate responses. An innate response to a smell is a response that an animal will display without ever having experienced that smell before. So this response must be generated by a hardwired brain circuit that’s been programmed by your DNA. For this to work, your brain obviously needs to keep track of which odors are supposed to cause which responses, which works best if the olfactory information is organized and not jumbled.

In contrast, learned responses don’t need to be hardwired into your brain, because these aren’t responses that you’re born with—instead, your previous experience teaches you how to respond to a smell the next time. Learning works by inducing changes in certain brain regions to link specific smells to specific responses. So in these brain regions, you don’t need to start with an organized system that dictates which odors lead to which responses. In fact, if you started with a super organized system it would just get all messed up by learning-induced changes anyway.

Don’t judge a brain by its cover

So there you go. You’ve got two very different types of animals with two very different noses and brains, but they actually smell things in a similar way. The practical importance of this discovery is that now we can be more confident that experimenting on flies will be useful in understanding how our own sense of smell works. As I discussed previously, there are numerous reasons why you might want to study flies instead of mammals.

Given that the insect and mammalian olfactory systems evolved independently, I still find it amazing that their structural and functional organization seems to be so similar. But this is something scientists have noticed time and time again: evolution often recapitulates certain themes to solve the same problems in different organisms. The deeper we look into other animals’ brains, the more facets of our own brain we see reflected back at us.



1. Personally, I don’t share this bias: my goal is not necessarily to understand the human brain, but to understand ANY brain. It’s amazing how even a fly brain allows a fly to sense its environment, perform a vast repertoire of behaviors, and make complex decisions. I would be thrilled to figure out how that works even if it didn’t end up having any relevance to humans.

2. In this post I wanted to describe the olfactory pathways as simply as possible, but here’s some more info for those of you who want specifics. The brain region mediating innate olfactory responses is called the lateral horn in flies and the cortical amygdala in mice. The region controlling learned responses to smell is called the mushroom body in flies and the piriform cortex in mice. In flies, the lateral horn and mushroom body are the only two pathways for smell processing in the brain. In mice there are a few additional regions that receive olfactory information (as shown in the figure above), but their functions are still unknown.

And for true aficionados, here are the major primary studies whose results I describe in this post:

de Belle JS, Heisenberg M. Associative odor learning in Drosophila abolished by chemical ablation of mushroom bodies. Science 263:692-695 (1994).

This was a foundational study showing that destroying the mushroom body in flies causes defects in learned responses to smell but not innate responses, thereby implicating two separate pathways in innate vs. learned olfactory responses.

Jefferis GS, Potter CJ, Chan AM, Marin EC, Rohlfing T, Maurer CR Jr, Luo L. Comprehensive maps of Drosophila higher olfactory centers: spatially segregated fruit and pheromone representation. Cell 128:1187-1203 (2007).

This study showed that the lateral horn, which mediates innate olfactory responses in flies, receives organized, structured information from projection neurons.

Caron SJ, Ruta V, Abbott LF, Axel R. Random convergence of olfactory inputs in the Drosophila mushroom body. Nature 497:113-117 (2013).

This study showed that the mushroom body, which mediates learned olfactory responses in flies, does not receive structured information from projection neurons; instead, the information is randomly mixed together.

Root CM, Denny CA, Hen R, Axel R. The participation of cortical amygdala in innate, odour-driven behaviour. Nature 515:269-273 (2014).

This study identified the mammalian brain region responsible for innate responses to smell, the cortical amygdala.

Choi GB, Stettler DD, Kallman BR, Bhaskar ST, Fleischmann A, Axel R. Driving opposing behaviors with ensembles of piriform neurons. Cell 146:1004-1015 (2011).

This study identified the piriform cortex as a brain region for olfactory learning in mice. The authors were able to induce learned responses in mice by artificially activating neurons in this region at the same time as giving a reward or punishment.

Sosulski DL, Bloom ML, Cutforth T, Axel R, Datta SR. Distinct representations of olfactory information in different cortical centres. Nature 472:213-216 (2011).

This study in mice showed that olfactory information from the projection neurons maintains a stereotypic organization in the cortical amygdala but not the piriform cortex. Two additional studies published concurrently, Miyamichi et al. and Ghosh et al., make similar conclusions.

Leave a Reply

Your email address will not be published.