BioProject

Dendritic and axonal morphology reflects the input and output of neurons and is a defining feature of neuronal types, yet our knowledge of the diversity of neuronal morphology, in particular distal axon projection patterns, remains limited. More...

Dendritic and axonal morphology reflects the input and output of neurons and is a defining feature of neuronal types, yet our knowledge of the diversity of neuronal morphology, in particular distal axon projection patterns, remains limited. To systematically obtain complete single neuron morphology on a brain-wide scale, we established a platform with five major components: sparse labeling, whole-brain imaging, reconstruction, registration, and classification. We fully reconstructed 1,741 neurons from cortex, claustrum, thalamus, striatum and other brain regions. We identify 11 major projection neuron types with distinct morphological features and corresponding transcriptomic identities. We further reveal extensive morphological diversity within each of these major types, some of which cluster into more refined morphological subtypes. We analyze this diversity at different levels following a set of generalizable organizational rules governing long-range axonal projections, including molecular correspondence, divergent or convergent projection, axon termination pattern, regional specificity, topography, and individual cell variability. We illustrate how these rules manifest in different projection neuron types. Although clear concordance with transcriptomic profiles is evident at major projection type level, fine-grained morphological diversity often does not readily correlate with transcriptomic subtypes derived from unsupervised clustering, highlighting the need for single-cell level cross-modality studies. Overall, our study provides a systematic demonstration of the crucial need for quantitative description of complete single cell anatomy in cell type classification, as the cell type-specific morphological diversity reveals a plethora of ways different cell types and individual neurons may contribute to the function of their respective circuits. Overall design: single-cell RNA-seq data from retrogradely labeled neurons (i.e., Retro-seq), including 1,134 retrogradely labeled neurons from SSp, SSs, MOp and MOs that mapped L2/3, L4 and L5 IT subclass, 300 retro-seq cells that mapped to Car3 subclass, and 1,699 non-retro-seq scRNA cells form Car3 subclass used for clustering analysis. Less...