All animals have a preferred temperature range, whether they’re crocodiles basking in the sun or a naked-mole rat burrowing underground. Being able to accurately sense environmental temperature (thermosensation) is a life or death scenario: too hot and the animal risks overheating whereas too cold can bring about hypothermia. Understanding how insects sense temperature is particularly interesting, given the range of environments they inhabit and the fact that they’re poikilothermic (or “cold-blooded”). This means that studying their temperature preference behaviour is more tractable than those pesky homeothermic animals (“warm-blooded”, like mice or humans).
Antennae are more commonly understood as structures for olfaction (smelling), but insects also use them for thermosensation and sensing how humid an environment is (hygrosensation). In Marin et al. (2020), we reconstructed the peripheral neurons responsible for thermo- and hygro-sensation in the vinegar fly (Drosophila melanogaster) using connectomics. But to find out how these neurons might affect behaviour, we went two steps deeper into the brain.
For the second-layer, we found 32 different neuron types in the antennal lobes (olfactory bulbs) of both hemispheres. Compared to olfactory neurons, these second-layer “VP PNs” (ventroposterior projection neurons) have more diverse shapes as a population and project to many more areas of the brain. What does this mean? Temperature and humidity can affect the processing of many different types of behaviour.
One region the VP PNs target is called the lateral accessory calyx (lACA), a thermosensory centre that is part of a memory-forming structure called the mushroom body. The calyces are where sensory information is sent to before memories are formed about that sensory information. So, we also decided to reconstruct all the neurons receiving information from VP PNs in the lACA, just to see what types there were and where else in the brain they might take this thermosensory information.
Interestingly, we saw that three similar neurons connected the lACA to the accessory medulla. Two of these neurons were identified as components of the circadian system (i.e. the fly’s body clock). This leads to an interesting question, can a blind fly still tell the time? We can feel that the day gets hotter and cooler, and so this connection could explain how temperature establishes circadian rhythm, even in the absence of light. Alternatively, perhaps this connection changes the fly’s preferred temperature during day vs. night.
Finally, we found the shortest sensorimotor circuit ever described in Drosophila’s brain: VP4 or ‘dry sensing’ PNs synapsed onto a descending neuron (DN). DNs take information from the brain to the ventral nerve cord, where they connect to interneurons responsible for coordinating the fly’s muscles. This results in movement and behaviour. This means that only three synapses separate the environment from muscle coordination in this circuit. Being too dry is probably one of the most dangerous situations for a fly (or even more so, its offspring). Maybe it’s equivalent to touching a hot stove and yanking your hand back – a reflex response.
Altogether, we find that the circuitry involved in thermo- and hygro-sensation is a lot more complicated and diverse than previously thought. Connectomic studies, such as Marin et al., 2020, generate “bottom-up” circuit level hypotheses for not just one circuit, but the entire downstream network of an interesting neural population. Using connectomics as a guiding hand, experimentalists can then tease apart the function of a circuit and also explore circuit avenues that otherwise could have been overlooked.
Marin et al., 2020. Connectomics analysis reveals first, second, and third order thermosensory and hygrosensory neurons in the adult Drosophila brain.