The sense of smell – A milestone to understanding the brain

This essay was written by Ed Bracey and was first published in the 2011/12 Mill Hill Essays.

The sense of smell is vital for the survival of many animals; odours allow the distinction between foods and poisons and they can mark the difference between prey, predators and potential mates. Some animals can smell not only if a prospective mate is in season but also how genetically similar they are. The extraordinary olfactory capacity of some animals can be used to our advantage. Dogs, for example, have historically been used to follow the scent of criminals and for sniffing out drugs. Security companies have recently developed a system that uses the highly tuned olfactory ability of bees to detect explosive substances such as TNT. But this sense can be useful to us in a number of other ways. The olfactory system, which underlies the ability to sense odours, has for many years been a focus of research including the study of sensory perception and repair of brain injury.

We perceive the world in terms of models of our surroundings that our brains create. Sensory neuroscientists are interested in how the brain generates such models, in which combinations of neurons must be activated and in what order to create various sensations; the sight of a cricket pitch, the sound of running water, the feel of the sun on our skin, the smell of freshly cut grass.

In order to generate a pattern of brain activity that we perceive as a smell, the olfactory system must first overcome the problem of translating a stimulus in the environment into electrical activity in nerve cells. Life on earth has evolved an astonishing array of specialised sensors to detect happenings in the world around us. It is the role of these sensors, called receptors, to respond to specific stimuli such as heat, light or touch, and translate them into electrical activity in nerve cells. Like the other sensory systems, the olfactory system uses receptors to detect stimuli. Olfactory receptors are tiny proteins found on the surface of specialised neurons in the nose called olfactory receptor neurons (ORNs) (Figure 1a,b). When an odour molecule diffuses into the nose, the olfactory receptor physically binds it, recognising its molecular shape in the way a lock recognises a key. This causes the receptor to change shape slightly, activating tiny protein machines inside the cell, which in turn open pores in the cell membrane called ion channels. When ion channels are opened, positively charged sodium and calcium ions move rapidly through them into the cell. This makes the inside of the cell more positively charged than the outside of the cell in a process called depolarisation. If the cell becomes depolarised enough, it triggers an action potential – an electrical signal that shoots down a specialised thread-like part of the neuron called the axon (Figure 1b). The axons of ORNs carry action potentials from the nose into a structure in the brain called the olfactory bulb, where the first stage of processing odour-evoked electrical activity occurs (Figure 1a, c).

One special feature of ORNs is that when they are exposed to an odour for a few hundred milliseconds, olfactory receptors stop working and cause the frequency of action potentials ORNs send to the olfactory bulb to decrease. This process, called desensitisation, is one of the ways in which the olfactory system is thought to cause smells that are constantly present in the environment to fade into the background, allowing new odours that are encountered to be preferentially focused upon. This explains why we often stop noticing smells after just a few moments of exposure. Leave a room full of cigarette smokers, come back once your olfactory system has recovered from desensitisation, and you will be surprised at how smelly the place seems to have become since you left.

Detecting odours is trickier than detecting other stimuli such as sound, light and colour. These phenomena are all detected by a small number of receptor types. The visual system, for example, can use the combination of various levels of activity in just three distinct types of receptor that respond to red, green and blue light to represent all the colours in the visual spectrum. The auditory system represents the full range of audible sound frequencies with just one type of receptor called a hair cell. But there are a bewilderingly large number of different odour chemicals in the natural world – many thousands of which could be relevant for survival. A food odour could contain hundreds of different volatile odour molecules and only one of those may indicate if the food is rancid or not. A single olfactory receptor “lock” that could bind hundreds of thousands of different odour “keys” would be improbably complex and highly unlikely to have evolved. Evolution has instead solved the problem by generating hundreds of different types of olfactory receptor lock, each type responding preferentially to just a few odour keys. Animals that are highly dependent on smell have many olfactory receptor types. Mice have more than 1000 distinct types of olfactory receptor while we humans, being primarily interested in visual phenomena, use only about 400. Each ORN contains the genetic blueprints necessary to make hundreds of different olfactory receptors but it expresses only one type of olfactory receptor on its surface. This means that each type of olfactory receptor is expressed by only a few thousand of the millions of ORNs in the nose. Exactly how each ORN comes to select a particular olfactory receptor, and what ensures that equal numbers of olfactory receptors are expressed is still being investigated.

The brain must then piece together the electrical information from ORNs to create a perception of smell. It first starts to process this activity in the olfactory bulb. All the ORNs that express a particular type of olfactory receptor are randomly dispersed throughout the nasal lining but when their axons reach the olfactory bulb, all axons from the ORNs expressing that receptor type converge on one region in the olfactory bulb called a glomerulus (Fig 1c). A glomerulus is a sphere-like bundle of nerve fibres and connections between neurons (synapses) around 50-100um in diameter. A typical mouse olfactory bulb contains about 1800 glomeruli in which ORNs make synapses with various other neurons which process the electrical activity entering the bulb. Here, ORNs also make synapses with the output cells of the olfactory bulb, the mitral and tufted cells. The output cells carry the information processed by the circuitry of the olfactory bulb to higher centres of the brain. Different odours activate different subtypes of ORNs and thus activate different combinations of glomeruli. In the same way that various combinations of electrical activity in an LED display board can be used to generate different light patterns, the ‘wiring’ of the olfactory bulb means that each of the thousands of chemicals in the natural world is represented by a specific pattern of activity on a two-dimensional map of glomeruli (Fig. 1c). Even simple odours, such as those used in food containing just a single or a few types of odour molecule, activate several glomeruli. More complex odours, such as those found in the natural environment of a forest or the bouquet of a wine elicit more complex patterns of spatial activity in the olfactory bulb.

After signals from ORNs have been processed by the olfactory bulb, they are rapidly passed on to another region of the brain called the olfactory cortex by the mitral/tufted cells (Fig. 1a).

Odour detection and representation by the olfactory system.
Figure 1. Odour detection and representation by the olfactory system.

A. Cross section of a mouse head. Odours drawn into the nasal cavity activate olfactory receptor neurons (ORNs, blue). The axons from ORNs project to glomeruli in the olfactory bulb where they connect with mitral cells (green) which carry signals to higher regions of the brain such as the olfactory cortex. Axons from the vomeronasal organ project into glomeruli in the accessory olfactory bulb.

B. Odours entering the nose are bound by olfactory receptors in the cell membrane of the ORN. Binding odours causes the olfactory receptor to change shape, activating protein machines (green) in the cell which cause ion channels (red) in the cell membrane to open. Ion channel opening causes positive sodium and calcium ions to enter the cell, depolarising it. If there is enough depolarisation, an action potential is triggered which is carried along the ORN axon to the glomeruli in the olfactory bulb.

C. All ORNs expressing the same receptor project to the same glomerulus in the olfactory bulb. Different odours activate different combinations of ORNs and thus different combinations of glomeruli, creating a unique map of glomerular activity in the olfactory bulb for each odour (highly active glomeruli are black, inactive glomeruli are pink).

Recent research suggests that the olfactory cortex may have an association function. Association is thought to link information from different brain areas. In humans, higher association areas allow us to correlate different perceptions. If you are asked to think of a purple apple, you can imagine it even though you may have never seen one. Parts of your association cortex allow you to link the concepts of purple and apple. The olfactory cortex may help our brains link smells to various pleasant or repulsive experiences.

The olfactory bulb and olfactory cortex are thought to have evolved far earlier than other sensory areas, and their circuitry is thus simpler. For example, the olfactory cortex has three distinct neuronal layers, each of which may perform different functions, while more recently evolved sensory areas such as visual or auditory cortex have six layers. The extra layers in more recent sensory areas dramatically increase the number of potential connections and complexity of cortical circuitry. In humans, this complexity allows us to perform intricate feats of language, thought and understanding yet makes scientific study of the brain areas that enable these tasks more difficult. The olfactory system is simple, but performs functions similar to those of newer sensory systems, such as representing stimuli with two-dimensional maps and association of odours with other stimuli. This makes it an attractive system to study; a complete understanding of the olfactory system may allow us to extrapolate processing rules to more complex, recently evolved brain regions.

The olfactory bulb also feeds directly into parts of the brain called the limbic system, which are involved in processing emotions and memory. All other senses except olfaction are first filtered and processed by an area deep in the brain called the thalamus before passing to the limbic system. The olfactory system’s hotline to the cortex and limbic system may be why odours more than any other sensation can be so hauntingly evocative; instantly and vividly conjuring up a feeling or a long-forgotten experience, a place, a person.

Another aspect of olfaction that has been well researched is the ability of some animals to use odours to communicate. Odours secreted by an animal that cause a change in another animal’s behaviour are called pheromones. There are many instances of pheromone usage in the animal kingdom, they signal various phenomena such as readiness to mate, territorial boundaries or aggressive intent. In most animals, pheromones are detected by the vomeronasal organ (VNO), which is found at the base of the nasal cavity (Fig 1a). The VNO operates on similar principles to the olfactory epithelium, binding odours with receptors expressed on ORNs which project axons into an area similar to the olfactory bulb called the accessory olfactory bulb (Fig 1a). The urine of some species contains peptides specific to each animal’s genetic make-up. The VNO can be used to detect these peptides and thus recognise the genetic identity of others. In the late 1950s Hilda Bruce discovered that pregnant mice abort fetuses of genetically similar males if a more genetically dissimilar (and thus more evolutionarily advantageous) male becomes available. This is known as the Bruce effect.

Whether humans use pheromones is unclear. It is likely that we possess a vestigial VNO but studies have found that no axons wire it up to the brain. There is also no evidence of an accessory olfactory bulb in humans. The main olfactory bulb can detect pheromones in other species and could also theoretically do so in humans. No pheromones have been isolated and conclusively shown to cause physiological changes in human behaviour. Some studies have suggested that androstadienone, a component of male sweat, may have a number of effects on women such as increased sexual arousal, a calming or comforting effect and increased perception of pain intensity, but these have yet to be proven conclusively.

In the early 1970s a researcher called Martha McClintock suggested that women synchronise their menstrual cycles through pheromonal communication. Women who were exposed to the sweat of others at different phases of their menstruation seemed to change the length of their cycles. However, more recent experiments have been unable to repeat the effect and re-analysis of the original methods highlighted errors in both data collection and analysis. Most scientists now consider the “McClintock effect” an artefact of methodological error, although the idea remains widespread in society.

An altogether different facet of olfactory bulb function has recently become a focus for researchers investigating ways to heal injuries to the brain and spinal cord (the central nervous system). The axons of damaged neurons do not re-grow in the adult central nervous system because the support cells and tissues in the central nervous system express a number of inhibitory signals preventing it. Furthermore, when the nervous system is injured, local support cells called astrocytes are activated and start to form what is known as a glial scar. In the first instance the scar is useful as it seals up damage to the barrier between the spine and the rest of the body, protecting the central nervous system from infection. However, astrocytes also produce chemical signals that inhibit axon regrowth. In the spinal cord, axons above the area of damage cannot grow through the glial scar and so cannot reconnect with lower parts, preventing victims from recovering the use of limbs and other body parts below the area of injury. Many different treatments developed to try to overcome this inhibition have met with little success.

Olfactory ensheathing cells
Figure 2.
Olfactory ensheathing cells surround olfactory receptor neurons as they pass from the nasal epithelium, through a bone in the skull called the cribriform plate and into the glomerulus.

In humans, the olfactory system is unlike other sensory systems because its receptor neurons are continually replaced in the adult. Newborn ORNs send axons from the nose into the olfactory bulb happily uninhibited by the usual signalling factors that would stop other neurons in their tracks. The key to the ORN’s abnormal ability to pass through the nervous system lies with a special stem cell present in the olfactory epithelium called the olfactory ensheathing cell (OEC) (Fig 2). Unlike other cells in the adult, stem cells can divide indefinitely, creating daughter cells that either remain stem cells and continue dividing or differentiate into other cell types. Some stem cells, such as those in the heart or pancreas divide only under special circumstances, such as when tissues are damaged. Others, such as OECs or those in gut or bone marrow, regenerate throughout life to replace parts worn out or damaged. OECs are stem cells which wrap around the axons of newly born olfactory receptor neurons, providing them with growth hormones and chaperoning them through the unwelcoming environment of the olfactory bulb until they reach the glomerulus (Fig. 2). The discovery of OEC function led to the hope that introduction of OECs into injured brain areas could encourage axons to grow through damaged regions and reconnect with their targets. Results in mice and rats with injured spinal cords have been promising; animals injected with OECs recover some function in regions below the spinal injury. Few clinical trials have taken place in humans although some mild improvements after OEC transplants in damaged spinal cords have been reported. However, the technique is still mired in problems. Most studies that have shown a positive outcome were done in mice and rats. Small rodents may not make the best clinical models of spinal cord injury because unlike humans, they seem to have some innate regenerative capacity in their spinal cords. Also much research has not always checked rigorously that spinal cords have been completely severed before treatment with OECs. Some studies that claim to find axons regrowing through the damaged spine may simply have found old undamaged axons. It seems unlikely that transplants of OECs alone will be enough for full recovery of spinal tract injury. This may be disappointing but studying just how OECs promote axon growth in the CNS may yet give hope to victims of spinal cord injury and other victims of debilitating brain disorders such as stroke and Parkinson’s disease.


The olfactory system is a rich source of promising research. Understanding how OECs help axons travel through the nervous system may provide scientists and doctors with new treatments for those with spinal and brain injuries. Understanding how the comparatively simple brain areas of the olfactory bulb and olfactory cortex operate may offer a shortcut to decoding the workings of more complex sensory systems and even to the holy grail of neuroscience – the processes that underlie awareness and consciousness.

Major Kovalyov’s nose crossing Palace Square
Major Kovalyov’s nose crossing Palace Square, an illustration by Sergey Aleksandrovich Alimov for Gogol’s satirical tale The Nose.
By Сергей Алимов (Author) [FAL], via Wikimedia Commons

Leave a comment


email* (not published)