Characterizing Neuronal Diversity in the Adult Human Brain by Single-Nucleus RNA Sequencing

Understanding the human brain requires translating basic cellular physiology and neuroanatomy into an extensive interconnected network of ~100 billion neurons.  Integral to this is the identification of neuronal subtypes based on functional gene expression profiles.  However, characterizing the transcriptome of individual human adult neurons has been hampered by the difficulty in applying existing single cell RNA sequencing methodologies to the readily available repositories of postmortem tissues.  Major challenges remain in the isolation of neurons that are intact, uncontaminated by glia, or which have minimal RNA degradation.  To circumvent these issues, we have adapted the Fluidigm C1 microfluidics system to amplify total RNA from single neuronal nuclei for sequencing.  Using this method, we have produced 3084 quality-filtered data sets from individual neuronal nuclei across six anatomically distinct regions of the prefrontal cortex.  This sampling depth permitted unbiased classification of 16 distinct neuronal subtypes encompassing all of the cortical layers and revealed heterogeneity in expression profiles between functionally distinct brain regions that would be masked in bulk tissue studies.  As such, we demonstrate a robust and scalable method for detection and categorization of single cell subtypes using difficult or limiting tissues through examination of their nuclear transcriptome.