By K Meechan
When you stop and think about it, our sense of smell is actually very impressive. We can identify a vast array of odours: say, a burning bonfire, freshly cut grass, or that distinctive smell of milk that has been in your fridge for far too long. Each of these odours is made up of a complex array of molecules in the air, present at different concentrations – yet somehow our brain must detect and interpret this information to identify what we are smelling (out of all the many possibilities) and the appropriate response.
This response also relies on our brain’s ability to integrate olfactory information with that from our other senses. E.g. if something smells delicious, but looks like it’s growing a plethora of moulds – then we really shouldn’t be choosing to eat it. Our previous experience is also factored into these decisions – for example, we learn that a gas leak will smell of rotting eggs [as companies add the compound ‘mercaptan’ to it]. Therefore, when we smell rotting eggs, our previous knowledge should tell us that this situation is dangerous, and allow us to respond accordingly.
Studying our sense of smell therefore offers a window into how a sensory system functions: how do we sense odours? How is this information processed by the brain? How is it integrated with information from our other senses / from memory and learning?
The power of the fruit fly
Our model of choice is the fruit fly – Drosophlia melanogaster. It too uses its sense of smell to detect a wide range of odours, and responds with a number of complex behaviours.
For example, fruit flies avoid sites that smell of Leptopilina wasps – these wasps are a great danger as they inject their eggs into larvae, resulting in the wasp offspring consuming them from the inside. This response is therefore vital for Drosophila survival. It’s an example of an ‘innate’ behavioural response to an odour i.e. one that is ‘hard wired’ in the fly, which doesn’t need to be learned or practised. Due to the stereotyped nature of these responses, they are thought to be (at least at some level) genetically specified.
But, just as humans can draw on their memories and past experience when interpreting odours, so can flies. For example, in the lab we can train flies to associate an odour with a reward (e.g. sugar) or a punishment (e.g. an electric shock). Flies will then approach or avoid these odours respectively. These ‘learned’ behaviours are the result of experience, and are caused by plasticity in the brain – i.e. the ability of neural circuits to change their responses with experience.
So, by studying the fly olfactory system, we can address questions about how this sensory system functions, and also how innate and learned behaviours interact. All of this with the many advantages that Drosophila offers as a model system.
A Crash Course in Fly Olfaction
So what do we know about the fly olfactory system?
This system has been studied for many years, so the first and second order neurons are well described. Intriguingly, the organisation of the fly olfactory system has many similarities to that of mammals, including humans.
First order neurons – ORNs
The equivalent of the ‘nose’ for a fly, is its antennae and maxillary palps. These head structures are covered with tiny sensory hairs (known as sensilla) that contain the dendrites of a particular type of neuron called an ORN (Olfactory Receptor Neuron).
There are around 410 olfactory sensilla on the antennae, and around 60 on the maxillary palps. These sensilla are divided into various morphological and functional classes, each typically housing two ORNs (although this does vary). This results in a total of around 1200 ORNs on the antennae, and 120 on the palps.
Different ORNs are specialised to detect different odour molecules, dictated by the types of odorant receptor protein they express. Most ORNs express one of around 60 different odorant receptor proteins present in Drosophila, although some express multiple receptors. Binding of odour molecules to these receptors triggers depolarisation (‘firing’) of the ORNs.
Each receptor protein (and therefore each ORN) has a specific odour response profile; i.e., different receptors bind to different sets of odour compounds with varying degrees of affinity. Some receptors are narrowly tuned, in that they will only respond to a small number of odours. Conversely, some have much broader tuning, responding to a wide variety of different odours. Various studies have looked systematically at the odour response profiles of olfactory receptors – we now know the response profiles of almost all receptor types in Drosophila.
Studying these response profiles reveals the combinatorial coding strategy used in olfaction – each olfactory receptor is activated by multiple odour molecules, and conversely each odour molecule will activate multiple olfactory receptors. Therefore, any particular odour is represented in the brain by the combined activity pattern of a number of ORNs. It’s thought that this combinatorial representation is what allows the brain to recognise a vast number of different odours, with only a very limited number of different receptor types.
The primary olfactory centre – the Antennal Lobe
After ORNs have been activated by various odours at the antennae and palps, these signals are passed down ORN axons to the primary olfactory centre of the brain – the antennal lobe (AL). This is the equivalent of the olfactory bulb in mammals. This centre has an interesting organisation, whereby it’s partitioned into different anatomical and functional domains known as ‘glomeruli’ [approximately 52 for Drosophila]. ORNs expressing the same olfactory receptor, while distributed broadly on the antennae / palps, will send their axons to converge in the same glomerulus of the AL. This results in different regions of the AL representing information about different sets of odours.
Second order neurons – the PNs
The second order neurons of the fly olfactory system are known as Projection Neurons (PNs). These are analogous to the mitral/tufted cells in mammals. Again, there are many different types of PN, each having dendrites specific to a particular glomerulus. In most cases, several PNs will innervate each glomerulus, receiving connections (synapses) from the ORNs there. This allows ORNs detecting different odours to activate very specific subsets of PNs – maintaining the distinct channels set up by the different olfactory receptors on the antennae/palps.
Various other neurons are also present in the Antennal Lobe – for example, a network of local neurons forms connections between themselves, and also to ORNs and PNs. These have various effects on the responses of both ORNs and PNs.
Higher order processing – the Lateral Horn and the Mushroom Body
From the antennal lobe, PN axons then relay the signal to two major brain regions – the lateral horn (LH) and the calyx (dendrites) of the mushroom body (MB). The LH is thought to be involved in mediating innate responses, while the MB is a massively parallel processing structure important for associative learning and memory.
While we know a lot about the first layers of the olfactory system, the processing in the LH, MB and beyond is less well understood. Using a connectomic approach, our lab aims to reconstruct these regions at synaptic resolution (i.e. down to individual neuron morphologies and synaptic connections).
Part 2 of this post will look at the organisation of the LH and MB, and the open questions that our group is working to address
Ebrahim SAM, Dweck HKM, Stökl J, et al. Drosophila Avoids Parasitoids by Sensing Their Semiochemicals via a Dedicated Olfactory Circuit. PLOS Biol. 2015;13(12):e1002318. doi:10.1371/journal.pbio.1002318.
Grabe V, Sachse S. Fundamental principles of the olfactory code. Biosystems. 2017. doi:10.1016/j.biosystems.2017.10.010.
Wilson RI. Early olfactory processing in Drosophila: mechanisms and principles. Annu Rev Neurosci. 2013;36:217-241. doi:10.1146/annurev-neuro-062111-150533.
Laissue P.P., Vosshall L.B. (2008) The Olfactory Sensory Map in Drosophila. In: Technau G.M. (eds) Brain Development in Drosophila melanogaster. Advances in Experimental Medicine and Biology, vol 628. Springer, New York, NY doi:102-114. doi:10.1007/978-0-387-78261-4_7.
Fiala A. Olfaction and olfactory learning in Drosophila: recent progress. Curr Opin Neurobiol. 2007;17(6):720-726. doi:10.1016/j.conb.2007.11.009.
Image Credit: Adult brain and segmented neuropils – Template brain created by Arnim Jenett (Janelia Research Campus), Kazunori Shinomiya and Kei Ito (Tokyo University) [Ito et al., 2014] Then transformed to the EM template FAFB [Zheng et al., 2017]