Analysis of in vivo grid cell activity

Csordás, Dóra Éva (2020): Analysis of in vivo grid cell activity: burst firing and spatio-temporal anticipation. Dissertation, LMU München: Graduate School of Systemic Neurosciences (GSN)

Abstract

The manner in which the brain encodes the position and movement of an animal as it navigates its environment has been a topic of intense study for decades. Interestingly, specific cell types have been identified that are thought to contribute to distinct aspects of spatial navigation. An important class, the so-called grid cells of the medial entorhinal cortex (MEC), fire spikes preferentially when the animal is near specific locations in its environment. For each grid cell, these locations form the vertices of a hexagonal lattice spanning the explored environment, which promoted the idea that the brain uses grid cells to measure the amount of distance travelled. On the other hand, many details about how the temporal firing behaviour of grid cells informs the brain about the animal's position and trajectory remain unclear. My thesis is based on two projects that address such issues of temporal coding in grid cells. In the first study we investigated grid cells based on spike-time autocorrelations and the existence of depolarizing afterpotentials (DAPs) following single spikes. Analyzing whole-cell data from mice running on virtual tracks, we found three different groups, "sparsely bursting cells", "bursty cells with depolarizing afterpotentials" and "bursty cells without depolarizing afterpotentials". Bursty cells with prominent DAPs were mostly stellate cells in Layer II of the MEC; their interspike intervals (ISIs) reflected DAP time-scales (5-10 ms). In contrast, neither the sparsely bursting pyramidal cells in Layer III, nor the high-frequency bursters in Layer II, showed a DAP. The ”bursty without DAP” cells had the earliest peaks in the ISI distributions, consistently around 4 ms. We hypothesized that these differences in the temporal characteristics could resemble differences on spatial coding. However, extracellular recordings from mice exploring real 2D arenas did not show strong differences in the tuning properties of the three cell groups. We next extended our analysis to non-grid principal cells in the MEC and found similar discharge characteristics. These findings suggest that depolarizing afterpotentials shape the temporal response characteristics of principal neurons in MEC with little effect on spatial properties. In the second study we asked whether grid cells encode the animal's current location or show some predictive behavior concerning the animal's future movement. We found that the grid cell firing rate is higher when the animal moves into a firing field of that neuron ("inbound" parts of the animal's trajectory) than when it travels in an outbound direction. We could eliminate this difference by shifting spikes ahead in time or in space along the momentary movement-direction or the head-direction vector or by a given distance along the trajectory. Optimal shifts ahead were around 170 ms and 2-3 cm, respectively. With these optimal forward shifts, we have shown that grid-cell activity indeed anticipates future movements. In short, in my thesis I have shown that DAPs influence burst firing characteristics of principal cells in the MEC and that grid cells exhibit anticipatory firing. Both findings add new aspects to the rich literature on the neural basis of spatial navigation.

• Dokumententyp: Dissertationen (Dissertation, LMU München)
• Themengebiete: 500 Naturwissenschaften und Mathematik ; 500 Naturwissenschaften und Mathematik > 570 Biowissenschaften, Biologie
• Fakultäten: Graduate School of Systemic Neurosciences (GSN)
• Sprache der Hochschulschrift: Englisch
• Datum der mündlichen Prüfung: 27. Oktober 2020
• 1. Bericht­erstatter:in: Sirota, Anton