Some fruit fly species sniff out just about any fruit; others are pickier, sticking to just one kind. These behavioral quirks are thought to reflect neural changes that evolved to help different species adapt to distinct environments, but how those changes came about has remained a mystery.
Some parts of the Drosophila olfactory circuitry have evolved more than others, according to a new comparison of two fly species’ connectomes. Certain characteristics, including neuron number and type, are strongly conserved between the species, but others, such as the balance of excitation and inhibition in the circuit, differ.
The findings, described in a preprint posted on bioRxiv in June, begin to reveal the evolutionary changes in the brain that may have helped the two species develop different olfactory preferences and adapt to their particular environments, says principal investigator Lucia Prieto-Godino, group leader at the Francis Crick Institute.
Numerous studies have compared gross neuroanatomy across species, but the new work is one of the most complex cross-species network comparisons, says Greg Jefferis, group leader at the MRC Laboratory of Molecular Biology, who was not involved in the work but collaborates with Prieto-Godino on other projects. The preprint authors “make a pretty strong case that these differences that they see in the connectome are actually meaningful for the behavior of the animal,” he adds.
Still, the paper’s conclusions are based on just two connectomes, says Alexander Bates, a neurobiology postdoctoral research fellow in Rachel Wilson’s lab at Harvard Medical School. Bates has worked with the authors of the preprint before but was not significantly involved in this project. Because each connectome captures a snapshot of a single fly’s brain, there is no guarantee that the differences spotted between the connectomes are at the species level and not the individual level, Bates adds.
T
he new work compares the antennal lobe connectomes of larval Drosophila melanogaster, a commonly used model organism, and Drosophila erecta, a non-model species. The two are closely related, but the latter breeds on the fruit of a woody plant called the pandan, while the former opts for a wide range of fruits.
D. erecta “has a very different lifestyle from melanogaster, so then we thought that that probably had changed the olfactory circuit,” Prieto-Godino says.
The larval Drosophila olfactory system is well suited for the comparative analysis, in part because it is small, comprising just 21 olfactory sensory neurons.
The size of the larvae allows for “a comparative analysis on a synaptic level of the highest possible resolution, which is not yet possible in other systems. I think that that’s very exciting,” says study investigator Christoph Giez, a postdoctoral fellow in Prieto-Godino’s lab.
“If you imagine doing this in a mouse, or even a bigger brain, it’s going to be a total mess,” says Katrin Vogt, group leader at the University of Konstanz, who was not involved in the work. “I think they are really a sweet spot, because they are not too closely related species, and one is a specialist.”
Another advantage of using flies is that their olfactory neurons and circuits are organized similarly to those in humans and many vertebrates, Prieto-Godino says. “We can hopefully learn general principles about how central neural circuits evolve generally, but also things that might be transferable across olfactory systems.”
Prior work had already used electron microscopy to map every single neuron in the olfactory system of D. melanogaster. Prieto-Godino’s team created a comparable map for D. erecta, which took about two weeks of continuous scanning with a customized electron microscope, she says.
Not only did the two species show identical cell types and number of cells, but the connections between different interneuron types within the antennal lobe were also similar.
“That defines kind of a scaffold, and this is probably the general circuit blueprint that is required to process olfactory information,” Prieto-Godino says, adding that this prediction needs further validation.
O
ne way to test if this truly is a general circuit blueprint for olfaction is to look at more species and see if the same characteristics are conserved, Prieto-Godino says.
“I would love to actually go farther because we didn’t find any new cell types,” she says. “I would like to now look at the species that are more far away related, to see if we can find novel cell types and see how these integrate into the circuit.
D. melanogaster and D. erectus differ in subtle ways, such as the ratio of excitatory versus inhibitory synapses on olfactory sensory neurons and projection neurons, the preprint suggests.
“Even within a single dendritic field of a neuron, evolution can target different synaptic elements differently,” Prieto-Godino says.
Because the species the researchers picked are so similar, establishing a baseline for individual variation within each species would strengthen the claims the paper makes, Bates says. “The first step in my mind would have been to establish the baseline.”
That said, connectomes are still quite costly to produce, and relatively few full connectomes exist to compare, Bates says. And even as more connectomes are published, comparing them can be difficult because not all connectomes come from similar individuals. Ideally, those being compared would be from flies that are the same age and weight and have similar behaviors, among other characteristics.
“In the future, we’ll do more connectomes, more rigorously, and we’ll be able to make these kinds of experimental comparisons with a better n size,” Bates says.
To control for possible individual variations in a single connectome, Prieto-Godino and her colleagues compared the connectomes bilaterally, which can catch individual miswirings because the two hemispheres are expected to be symmetrical.
The “ultimate goal” is to understand how behaviors change, and understanding neural circuits is one step in that direction, Prieto-Godino says.
“The way we did the analysis hopefully will serve as the basis for people doing cross-species comparative connectomics,” Prieto-Godino says. “Kind of a guide of, ‘What are the kind of things that we can find? What are the kind of things that we can look at?’ And hopefully that will go beyond our findings.”