Engineering a Memory with Optogenetics

Memories are mysterious and can be irritatingly illusive – is there anything more frustrating than the feeling of knowing something, of it being on the tip of your tongue, only for it to remain stubbornly inaccessible, hidden away deep in your brain? Understanding memory, even ‘finding’ memories in the brain, has been the subject of intense research for decades. A recent paper published in Nature Neuroscience represents a major advance, using optogenetics to ‘reverse engineer’ memories in mice.

Optogenetics is a method which inserts opsins, light sensitive proteins originally isolated from microbes, into the genome. Numerous different opsins exist and distinct families have different effects in response to light: bacteriorhodopsins (which pump protons out of the cell) and halorhodopsins (which pump chloride ions into the cell) are inhibitory by making it harder for neurons to fire action potentials whereas channelrhodopsins allow positively charged ions to enter and stimulate neurons by increasing their firing rate. The key advantage of this technique is the ability to place opsins under the control of genes that are only expressed in specific neurons and then directly control their activity.   


Deisseroth, K. Nature Neuroscience. 2015 Sep; 18(9): 1213–1225.

In this study, Vetere et al (2019), used optogenetics to generate a fully artificial memory. First, they showed that an odour – acetophenone – can be paired with a mild foot shock to produce an aversive associative memory; after conditioning mice exposed to acetophenone will then try to avoid the odour. Next they used optogenetics, inserting channelrhodopsin2 (ChR2) into olfactory sensory neurons with a specific odorant receptor – M72 – which is strongly activated by acetophenone. Pairing the stimulation of these neurons with the foot shock was able to generate the same memory as the actual experience of smelling acetophenone. However, this is only one part of the circuit, stimulating these neurons replaces the smelling sensation but does not remove the foot shock required to form the memory. To replace this component, they targeted neurons projecting from the lateral habenula to the ventral tegmental area, which are involved in mediating avoidance and aversion from stimuli (optogenetics was used to find this out, too) Again, they inserted ChR2, but this time they delivered the gene using a virus and then implanted an optrode – to deliver light – above this brain region.

With both limbs of the circuit now in place, researchers paired stimulation of the M72+ sensory neurons and stimulation of the lateral habenula projections. 24 hours later, when mice were exposed to acetophenone for the 1st time, they avoided the odour although they had never previously experienced either the odour, or a foot shock associated with the odour. Importantly, the researchers also showed that they were able to create the opposite memory – an attractive memory –  by pairing stimulation of the same M72+ sensory neurons with projections from a different brain region which mediates rewarding signals (the laterodorsal tegmental nucleus). Stimulating this new circuit had the same effect as pairing acetophenone with a food reward. 

Vetere G et al Nat Neurosci. 2019 Jun;22(6):933-940

Diagram showing the neural circuits targeted with optogenetics and results. CS + US is where both the sensory neurons and value neurons were stimulated compared to US | CS where the two populations of neurons were not paired. In the pink, where the lateral habenula projections were paired with the M72+ neurons, the mice avoided the odour whereas in the blue, when the laterodorsal tegmental nucleus was paired with the M72+ neurons the mice preferred the odour.  

The researchers here bypassed experience entirely and implanted a specific memory in mice by directly stimulating specific populations of neurons within the brain. The reverse engineering of a memory in this way leads to fascinating questions. Does stimulation of the sensory neurons induce a smelling sensation in the mice? Does stimulation of the aversion or reward signalling neurons induce a pleasant or unpleasant sensation? Can these implanted artificial memories override experience induced memories? If only the mice could tell us…

The olfactory system because of its stereotyped organisation and spatial mapping, as well as the relative simplicity of the stimulus (it was just a smell, after all), was uniquely suited to this approach. Hopefully, optogenetics will power future advances, it is certainly a powerful tool. Unravelling memories in this way may improve our understanding of dementia and identify crucial pathways in addiction, paving the way for new treatments.

Jack Gordon

Biomedical science undergraduate interested in genetics, neuroscience, cancer research and the pharmaceutical industry.