The BRAIN initiative has put forth the development of a neuronal cell-type census as the first step towards mapping the structure and components of neuronal circuits. Several groups have used single-cell RNA sequencing (scRNA-seq) to shown that RNA transcript levels are cell type-specific, and by isolating and sequencing the RNA from individual cells they can create a catalogue of cell types within the brain. Likewise, single-nucleus methylcytosine sequencing (snmC-seq) has shown DNA methylation (mC) patterns within the genome are highly cell type-specific and may also be used to define cellular identity. With whole-brain census efforts underway, technology development for spatial mapping of cell types within the brain has become a major focus. Several groups have shown the potential of techniques which leverage scRNA-seq data and RNA fluorescent in-situ hybridization (RNA FISH) or in-situ sequencing to create spatial maps of cell types in cell cultures and tissue sections, though no such efforts have been reported for spatial mapping of cell types using snmC-seq data. Therefore, this proposal will seek to test and develop a method that leverages the cell type- specific snmC-seq data that is being generated for the mouse brain atlas to develop a 3D map of these cell types within the frontal cortex of mice. Fluorescent in-situ sequencing (FISSEQ) has been used to sequence RNA molecules at base resolution and can also be used to sequence DNA in-situ, but this method has yet to be adapted for in-situ methylcytosine sequencing. In addition, this method is hampered by the inclusion of two steps that limit the efficiency with it creates molecules that can be readily sequenced. Firstly, in-situ sequencing using FISSEQ calls for the RNA to be reverse transcribed into cDNA, the creators of this technique have identified this reverse transcription as a major factor limiting it?s efficiency. Therefore, a cell-typing strategy wherein no reverse transcription is necessary should be inherently more efficient. Secondly, the FISSEQ protocol calls for the production of circular cDNA through intramolecular ligation, but the formation of end-to-end loops is highly energetically unfavorable, as in-situ sequencing requires crosslinking of DNA and proteins within the cell to fix them in place. This proposal calls for the development of a method wherein DNA is chemically converted for methylcytosine sequencing in-situ, circularized through the direct ligation of hairpin adaptors to blunt DNA, followed by targeted in-situ sequencing. Successful completion of this project will result in a method to spatially map cell-types by their methyl-cytosine patterns, bypassing the inefficiencies introduced by reverse transcription and intramolecular ligation. This technique called methyl-cytosine in situ sequencing (MIS-SEQ) will be paired with tissue clearing techniques and light sheet fluorescence microscopy in order to sequence thick mouse brain slices. The combination of these techniques will allow in-situ methylcytosine sequencing and spatial mapping of cell types in the entire mouse frontal cortex from 6 slices (~430 ?m sections).
The advent of single-cell genomics and transcriptomics has generated a large catalog of defined cell types, but current high-throughput sequencing approaches do not address the spatial context of these cell types, and spatial methods for cell typing do not leverage the large amount of epigenomic data being generated. The work proposed here will result in spatially resolved sequencing technology that will allow researchers to leverage data being generated by the BRAIN funded Center for Epigenomics of the Mouse Brain Atlas (CEMBA) for three-dimensional mapping of cell types in the mouse brain using in-situ methylcytosine sequencing, and will generate a complete map of the cell types in the mouse frontal cortex as a first step towards this goal. This approach will result in the creation of a technology that enables researchers to create epigenetic maps of each cell type that can be connected with existing brain gene expression atlases and connection maps, directly addressing a technological barrier towards using epigenomic data to address the second BRAIN Initiative priority research area (Maps at multiple scales).
