I used Neuropixels probes to record from 2121 well-isolated neurons across 22 regions in 29 mice navigating a virtual linear corridor. The corridor consisted of two sensory identical halves with two types of landmarks (Figure 1). The visual contrast of landmarks in the corridor varied across trials. In a subset of trials the landmarks were also the source of an auditory stimulus. The gain of the running wheel varied to decouple physical and virtual position. I quantified the response of each neuron in the corridor as a function of virtual position, sensory intensity, reward, and running speed, using linear ridge regression.
Regions included visual, somatosensory, retrosplenial, and motor cortex, the hippocampal formation, the dorsal thalamus, striatum, and midbrain. Activity of neurons was modulated by position in the corridor (Figure 2), sensory intensity (Figure 3), reward, and running speed. There was a positive correlation between the number of neurons modulated by virtual position or sensory intensity and the number of neurons that were modulated by both.
These results indicate that spatial coding is widely distributed across the brain (objective 1), and that the proportion of neurons integrating sensory and spatial information increases with a higher percentage of neurons modulated by either (objective 2). These results have been disseminated on international and European conferences, such as SfN (society for neuroscience), iNav (interdisciplinary navigation), DNM (Dutch Neuroscience meeting), and UCL (University College London) Neuroscience symposium, and will be part of a manuscript for publication.
To establish the hippocampus as the potential origin of spatial signals in sensory cortex (objective 3), we worked on the development of different tools. First, we designed a genetically modified mouse line to temporarily, completely, and specifically shut down the hippocampus. This mouse line was then proposed in a grant for the creation of a genetically modified mouse line by the PIs of my host laboratory, which was awarded in January 2023. Note that this is not a monetary grant, but a grant to develop a mouse line cost-free. Once this mouse line becomes available, we aim to use it in the same corridor as described above to test the effect of hippocampal shutdown on spatial coding in the sensory cortex (Figure 4a). We expect that the development of this mouse line will have a broad impact to the field, as specific and complete hippocampal shutdown will reveal the role of the hippocampus not only in spatial navigation, but also other cognitive processes such as memory. Many collaborators across Europe and the United Kingdom already expressed their interest in this line.
Second, we developed tools for longitudinal Neuropixels recordings. Given the complex shape of the hippocampus, shutdown techniques such as chemogenetics (injection of agonist) are more suitable than optogenetics (requires implantation of multiple light sources), especially when combined with recordings. To efficiently track activity of single neurons across shutdown and control experimental blocks, we need to be able to longitudinally record from the same neurons. In preparation for these experiments, we developed software to do this. The manuscript describing this software is currently in revision (already available via bioRxiv), and the software was made available online via Github.
