We have successfully developed a reliable experimental pipeline that allows in situ subtype identification with unprecedented power. This method uses the expression of 72 different genes to assign in vivo recorded cells to a fine subtype. Using this pipeline, we recorded and molecularly identified more than 1000 inhibitory neurons in mouse primary visual cortex.
We found that the visual properties of inhibitory neurons in the primary visual cortex differed greatly between the major Families (Pvalb, Sst, Vip, Sncg and Lamp5). However, the fine subtypes composing these Families of inhibitory cells shared very similar visual responses. In sharp contrast, the modulation by the animal’s state greatly varied even between fine subtypes of the same Family. Strikingly, we found that a single axis of variability, defined using only gene expression, could predict this state modulation across all subtypes we have recorded from. This single axis also correlated with the spike properties of the subtypes as well as with their morphology. These results were collected and published online in a pre-print article (doi:
https://doi.org/10.1101/2021.10.24.465600(odnośnik otworzy się w nowym oknie)) which was presented at different international conferences (COSYNE, FENS forum, SfN annual meeting) and has just been accepted in principle for publication by the peer-reviewed journal Nature.
We are currently still exploring how the connectivity of these fine subtypes can determine their tuning properties. In particular, we are using retrograde tracing to decipher whether excitatory cells with different modulation by the animal’s arousal are connected by diverse neuronal subtypes and how this relates with their in vivo properties.
