Long-range functional connectomics

Laubender, David (2022): Long-range functional connectomics: towards a functional connectome underlying the generation of stimulus selectivity in mouse visual cortex. Dissertation, LMU München: Fakultät für Biologie

Abstract

Ever since Hubel and Wiesel proposed their famous model on how circular receptive fields in the visual thalamus are transformed into the elongated and oriented receptive fields of simple cells in layer 4 of primary visual cortex (Hubel and Wiesel, 1962), this circuit has become a classic example of how stimulus selectivity emerges in the neocortex. They suggested that this transformation is implemented by a simple feedforward connectivity scheme, by which sets of geniculate neurons with their circular receptive fields aligned in visual space specifically converge onto individual cortical simple cells. Consequently, postsynaptic simple cells obtain their elongated receptive fields by simple linear summation, thereby becoming selective to orientations in space. However, despite being reproduced in almost every neuroscience textbook, direct evidence in favor or against this circuit model is still lacking today, more than half a century later. During my PhD, I aimed to obtain a comprehensive dataset for determining the role of the circuit logic underlying the generation of orientation selectivity in simple cells of the mouse primary visual cortex. To decipher the functional logic of connectivity, two components are necessary: the neurons’ function and their inter-connectivity, which coalesce in an approach called functional connectomics (Reid, 2012). In order to yield the necessary single cell resolution for functional connectivity, a functional readout of in vivo two-photon calcium imaging should ideally be combined with subsequent high-resolution 3D electron microscopy-based connectomics. So far, each of these have been studied in isolation due to technological limitations, or have been confined to local functional connectomics (Bock et al., 2011; Briggman et al., 2011; Lee et al., 2016b; MICrONS Consortium et al., 2021). In this thesis, I developed a novel long-range functional connectomics pipeline. I tailored and deployed this experimental pipeline to acquire a comprehensive, multimodal dataset. Specifically, I mapped functional receptive field properties, like ocular dominance, spatial receptive fields, orientation and direction selectivity, in layer 4 cells and dLGN axons in thalamo-recipient layer 4 of binocular primary visual cortex of mice using dual-color, deep in vivo two-photon calcium imaging. I found that functional response properties differed significantly between both populations, with cortical layer 4 cells being more binocular, having larger, more elongated and retinotopically confined receptive fields, sharper orientation and direction tuning compared to dLGN axons, in line with the circuit proposed by Hubel and Wiesel. The respective neurite morphologies of both layer 4 cells and dLGN axons were imaged in 3D, the mouse subsequently transcardially perfused, and a biopsy containing the imaged tissue extracted. For determining the underlying connectome, the biopsy was stained for electron microscopy with heavy metals, resin infiltrated and embedded. I re-identified vascular landmarks using micro computed tomography and tracked the position of the functionally imaged field-of-view using vascular triangulation. Next, I ultra-sectioned the biopsy from upper layer 5 to middle of layer 2/3 into > 10000 consecutive 35-40 nm thin sections without a single section loss. Lastly, I imaged the neuronal ultrastructure of a 1 x 1 x 0.33 mm3 volume centered on the functional field-of-view using 3D multi-beam scanning electron microscopy, and aligned the resulting petabyte-sized rawdata into a coherent 3D volume. Although the connectomic analysis is still ongoing, I am convinced that this dataset provides the data quality and richness required for providing comprehensive evidence on the functional logic of geniculo-cortical connectivity in mouse binocular primary visual cortex. Such a functional connectome will finally allow us to make conclusive statements on the generation of orientation selectivity in cortical simple cells.

• Dokumententyp: Dissertationen (Dissertation, LMU München)
• Keywords: functional connectomics, geniculo-cortical, LGN, V1
• Themengebiete: 500 Naturwissenschaften und Mathematik ; 500 Naturwissenschaften und Mathematik > 570 Biowissenschaften, Biologie
• Fakultäten: Fakultät für Biologie
• Sprache der Hochschulschrift: Englisch
• Datum der mündlichen Prüfung: 9. Mai 2022
• 1. Bericht­erstatter:in: Bonhoeffer, Tobias