Spatial transcriptomics reveals human cortical layer and area specification

Area of Science:

• Neuroscience
• Developmental Biology
• Genomics

Background:

• Human cerebral cortex development involves distinct molecular and structural layers and areas.
• Single-cell transcriptomics advanced understanding but lost spatial context.
• Spatial resolution is crucial for understanding developmental processes.

Purpose of the Study:

• To investigate molecular, cellular, and cytoarchitectural development of the human fetal cortex with spatial single-cell resolution.
• To create a comprehensive spatial atlas of human cortical development.
• To uncover the mechanisms of cortical layer and area specification.

Main Methods:

• Multiplexed error-robust fluorescence in situ hybridization (MERFISH) for spatial transcriptomics.
• Deep-learning-based nucleus segmentation for cell identification.
• Integration with single-nucleus RNA sequencing.

Main Results:

• Established a spatial atlas of over 18 million cells across eight cortical areas and seven time points.
• Identified the early establishment of the six-layer cortical structure before visible cytoarchitectural layers.
• Discovered continuous and discrete modes of cortical areal specification, including an abrupt boundary between V1 and V2 visual cortices.
• Revealed early synaptogenesis upregulation in V1-specific layer 4 neurons.

Conclusions:

• Spatial relationships are critical for molecular specification of cortical layers and areas.
• The study challenges gradient-only models of cortical arealization.
• This work establishes a paradigm for spatially resolved developmental brain atlases.