Keep it clean

A suppression hierarchy determines the order in which a fly cleans the different parts of its body.

Andrew - So, the idea was that if we were to cover the body of the fly in dust, the fly would be faced with a decision on which of those body parts to clean first. What we found was that they will immediately focus on cleaning parts of their head, particularly, their eyes and then they would move on to clean their antenna, and then at a certain point, they would groom the posterior parts of the body. So, over time, they have a higher probability of performing specific movements going from the head to the back of the body. 

Chris - Julie, have you been able to find out the origin, neurologically, of these particular movements and how are you attacking that?

Julie - One of the wonderful things about working with fly is that we have genetic tools that allow us to drive expression of proteins of our choice in particular groups of cells in the nervous system. And each one of these stocks or lines of flies allows us to target a different discreet group of neurons in the nervous system of the fly. We use this collection of lines to determine which of those clusters disrupt or trigger aspects of grooming behaviour.

Chris - And when you do this, Andy, what do you see?

Andrew - Well, one of the big surprises when we activated these different neural subsets, we could activate very specific grooming movements. So, we found that the grooming behaviour could be almost divided up into its distinct component parts and when we've looked at the neurons that were involved in driving the specific movements, we found that they were sensory neurons and neurons within the brain that can drive very specific grooming movements. 

Chris - Would you therefore say they're almost like a pattern generator, and that you activate that pattern generator and you get that particular pattern of grooming in that particular body part?

Andrew - That's correct. 

Chris - So, Julie, does this mean then, that these pattern generators are connected together, and if you've got a sort of sequence that Andy's saying, you start with the head and work your way backwards on the fly. The one which is at the back is furthest down the pecking order and is turned on by the one before it, but suppressed by all the ones before that.

Julie - I would say that we're activating the behaviour upstream of the pattern generators, whose identity is still not known, and what's happening in our assay is that we're activating all of the triggers for the different behaviours and an upstream behaviour suppresses the ability to execute any of the downstream ones. When you put dust on the head, at the top of the hierarchy, they clean their heads and then they stop. That behaviour does not trigger any of the downstream behaviours.

Chris - So, with that observation in mind then, what do you think is going on? How do you account for that? If one thing's not triggering the next, how do you get that effect?

Andrew - So, if you go back to the original beginning experiment where we put dust on the whole body of the fly, what these essentially does, if you consider these different movements to be independently triggered, each body part has its own compliment of sense organs that can detect the dust. So, if you have dust on your eyes, in flies, they have bristles that can detect that dust on the eyes and then direct movements to the eye. The same goes for the wings and the abdomen; they all have their own independent way of sensing the dust. So, when you completely cover the body of the fly in dust, essentially what you're doing is making the flies want to groom all of their body parts, so in essence, the different movements are in competition with each other, so there is a hierarchy of suppression. The eyes, which are groomed first, that has the ability to suppress everything that follows it in the sequence. Whereas, something like he abdomen which comes after the eyes would not be able to suppress eye grooming, and it would have to essentially wait until eye grooming is completed. So, it's a competition, and it's almost as if the rules of the competition are pre-determined by some internal mechanisms that we have yet to figure out.

Chris - And if one thinks about animals like a cat which does have a very stereotypical way of grooming itself, so does members of the rodent family, mice and rats. Do you think you can use what you're seeing in the fly as a model for what we see in higher organism, and even in humans?

Julie - I think the idea that there are many different schema for how you produce behavioural sequences is a useful general insight. We look around and we see many different behavioural sequences, things like bird song, things like human language, could all be considered sequential organization, and there would be lots of different ways you might produce those sequences, but the way that we have found is quite different. All of the things are triggered at the same time, but one can suppress the other, and so, this suppression hierarchy results in sequential execution. This is an alternative way to make a sequence and may indeed turn up lots of behaviours, now that we know to look for it.