Sophisticated Precision in Fruit Fly Sensory Systems
Last week, we considered the signal “wave” that controls development of the compound eyes in fruit flies and the motor neurons that control their rapid zigzag turns in the air. Pilots have to learn the forces of lift, drag, and thrust, and know how to prevent stalls. They know that rapid banking turns vastly increase G-forces and come with a high risk of stalling. Aerobatic know-how comes built-in for fruit flies. The same is true for all unrelated animals that perform powered flight, whether mammal (bat), reptile (pterosaur), or bird.
Smell Sense
The miniature insect flyers around us also have good olfactory organs that help them smell where to go. They smell good (or well, for you grammarians).
A news item from Ruhr University Bochum in Germany tells, “How Fruit Flies Smell CO2.” This commands our attention. Could fruit flies offer solutions for climate change monitoring? The scientists are inspired by the fly’s sensing ability.
The new findings are to be incorporated into the development of a CO2 biosensor, which the Bochum team is researching in cooperation with the Institute of Aircraft Systems in Stuttgart. “This should enable us to detect CO2in liquid media, which is something that as yet can’t be done,” says Störtkuhl. CO2 sensors are used on the International Space Station, for example, where they must consume as little energy as possible. Given that physical measurement methods are not very energy-efficient, a biosensor could be a great alternative.[
Notice the higher rank bestowed on biological sensors. A biosensor might be able to detect other volatile molecules. And so, curiosity rises about how fruit flies smell carbon dioxide, and why they need to. Some readers may be aware that mosquitoes follow the CO2 in human breath to find their victims. Fruit flies smell CO2 emissions from fermenting fruit for their needs.
Since CO2 is ubiquitous in the atmosphere, flying insects must be able to detect elevated concentrations along a gradient. The tiny insects’ expertise at finding and following carbon dioxide plumes in the air leads to suspicions of sophisticated systems for detecting the “odorless” gas that we humans exhale with each breath. (Thank goodness it is odorless to us.) So, how do they do it? We look at the paper in PLOS One for answers.
Answer: They’re not sure. The team found two receptors on the third segment of the fly’s antennae that respond strongly to bicarbonate CO2 emissions when those receptors are encoded in frog eggs. The response, however, depends on the mix of receptors.
We found that application of sodium bicarbonate evokes large inward currents in oocytes co-expressing Gr21a and Gr63a. The receptors most likely form hetromultimeric [sic] complexes. Homomultimeric receptors of Gr21a or Gr63a are sufficient for receptor functionality, although oocytes gave significantly lower current responses compared to the probable heteromultimeric receptor.
They also found that citronellol blocks the receptors — good to know for manufacturers of insect repellants.
Taste Sense
In fruit flies, the gustatory and olfactory senses overlap. Dr. Roman Arguello at Queen Mary University of London has been “delving into the genes and cells behind their delicate noses and tongues,” finding that the insects are able to adapt their senses to their environments. He likens it to one fly responding to a ripe peach as if it “tastes and smells like tangy vinegar to one fly, but like a burst of summer to another.” It’s quite common in fruit flies, he says, which inhabit a variety of habitats. But is this evolution?
“It’s like finding hidden islands of diversity within a vast ocean of uniformity,” says Dr Arguello. “These changes in gene expression tell us about the evolution of new smells, new sensitivities, and even new ways of using scent to navigate the world.”
Again,
“These findings open up exciting new avenues for understanding how sex differences evolve and how they impact animal behavior,” says Dr Arguello.
But does this research published in Nature Communications really provide “valuable insights into the general principles of how sensory systems evolve,” as they claim?
All they found were changes in gene expression — not mutated genes that were naturally selected as Darwinian theory requires. The paper only mentions mutations four times (pleiotropic mutations, at that), but “evolution” 144 times. As for “selection,” most of the 14 mentions involved stabilizing selection (i.e., keeping things the same), not the positive selection required for novelty and innovation. The only mentions of “positive selection” were from another study that contradicted this team’s finding of predominantly “evolutionary constraint.” They had no word on how taste receptors originated in the first place.
Directional Sense
Returning to the fundamental aspect of flight (flies do fly), another paper, this one in Nature, discusses the directional sense in these tiny aerobatic insects. Mathematicians will enjoy the abstract from this paper:
To navigate, we must continuously estimate the direction we are headed in, and we must correct deviations from our goal. Direction estimation is accomplished by ring attractor networks in the head direction system.However, we do not fully understand how the sense of direction is used to guide action. Drosophila connectome analyses reveal three cell populations (PFL3R, PFL3L and PFL2) that connect the head direction system to the locomotor system. Here we use imaging, electrophysiology and chemogenetic stimulation during navigation to show how these populations function. Each population receives a shifted copy of the head direction vector, such that their three reference frames are shifted approximately 120° relative to each other. Each cell type then compares its own head direction vector with a common goal vector; specifically, it evaluates the congruence of these vectors via a nonlinear transformation. The output of all three cell populations is then combined to generate locomotor commands.
For non-mathematicians, we can just recall the coach’s advice in many sports to turn your head in the direction you need to go. Fruit flies automatically know this, because specific cells are doing vector calculus and nonlinear transformations, then giving commands to the legs and wings. If building a robot, this would require some engineering know-how:
Accurate navigation requires us to fix a goal direction and then maintain our orientation towards that goal in the face of perturbations. This is also a basic problem in mechanical engineering: how can we keep the angleof some device directed at a target? One solution to this problem is to use a resolver servomechanism to measure the discrepancy or error between the current angle and the goal angle.
The eight researchers describe how fruit flies keep focused on the goal. They create a model and compare it to live data from fruit fly behavior. Why is a transformation needed? Because, they say, motor commands must convert their coordinates from allocentric space (other-directed) to egocentric space (self-directed).
This poses a coordinate transformation problem. Here we describe a network that solves this problem. This network creates two opponent copies of the allocentric head direction representation, with equal and opposite shifts (θ ± shift). Each copy is then separately compared with an allocentric goal representation, to measure congruence with the goal. The difference between the two opponent congruence values becomes an egocentric motor command.
The PFL3R and PFL3L cell populations take care of this. But there was a surprise finding:
At the same time, our results highlight the unexpected role of PFL2 cells. These cells provide a solution to a classic problem — namely, the fundamental tradeoff between speed and accuracy. High feedback gain allows a system to converge quickly towards its goal, and so it makes sense that gain should be high when error is large, that is, when there is a large discrepancy between the system’s current state and its goal. However, high gain can cause overshooting of the goal, especially when error is already small. We show that PFL2 cells effectively adjust the system’s gain, depending on the magnitude of the system’s current error.
Wow. That should be enough to make us all pause before swatting. CO2 sensors on the antenna, taste sensor gene regulators that adapt to the environment, and vector calculus in the head and legs — how does all this sensory engineering fit into such a tiny fly? The authors had almost nothing to say about evolution, leapfrogging over it with this statement, “the idea that a neuron’s inputs are adjusted (over development and/or evolution) to fit into some standard dynamic range dictated by the biophysical properties of a typical neuron….”
While admiring the humble fruit fly, let us take a moment to marvel at the human mind that can figure these things out. The next time you see a tiny fly or gnat, imagine inventing tools to study those itty-bitty eyes, wings, and antennae — to say nothing of the genes and proteins regulating them — and figure out how they work, using models and mathematical functions. Our sensory capabilities are well-designed, too.
No comments:
Post a Comment