Highly Multiplexed Single Cell In Situ RNA Detection

Citation

Shah, Sheel Mukesh (2017) Highly Multiplexed Single Cell In Situ RNA Detection. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/Z9X63JXH. https://resolver.caltech.edu/CaltechTHESIS:12152016-144548062

Abstract

Identifying the genetic basis of cellular function and identity has become a central question in understanding the functioning of complex biological systems in recent years. Single cell sequencing techniques have provided a great deal of insight into the transcriptional profiles of various cell types. However, single cell RNAseq studies require cells to be removed from their native environments resulting in the loss of spatial relationships between cells and suffer from low detection efficiency. Moving forward, a central question in further understanding large biological systems consisting of many disparate cell types will be how do these cells interact with each other to form functional tissues. To accomplish this goal, a method that keeps the tissue architecture intact is required. Single molecule fluorescence in situ hybridization (smFISH) is one such technique, but suffers from a lack of multiplex measurement capability as only a very few genes can be measured in any given sample and has low signal to noise ratio. Here I present a method that overcomes the low signal to noise ratio by using an amplification technique known as single molecule hybridization chain reaction (smHCR). smHCR coupled with the existing sequential FISH (seqFISH) method, which overcomes the inherent multiplexing limit of smFISH, provides a powerful tool to measure the copy numbers of 100’s of genes in single cell in situ.

The mouse brain contains 100,000,000 cells arranged into distinct anatomical structures. While cell types have been previously characterized by morphology and electrophysiology, single cell RNA sequencing has recently identified many cell types based on gene expression profiles. On the other hand, the Allen Brain Atlas (ABA) provides a systematic gene expression database using in situ hybridization (ISH) of the entire mouse brain, but lacks the ability to correlate the expression of different genes in the same cell. Using the smHCR-seqFISH technique to measure the expression profiles of up to 249 genes in single cells in coronal brain sections, we have identified distinct cell clusters based on the expression profiles of 15000 cells and observed spatial patterning of cells in the hippocampus. In the dentate gyrus, we resolved lamina-layered patterns of cell clusters with a clear separation between the granule cell layer and the sub-granular zone. In CA1 and CA3, the data revealed distinct subregions, each with unique combinations of cell clusters. Particularly, we observed that the dorso-lateral CA1 is almost completely cellular homogeneous with increasing cellular heterogeneity on the dorsal to ventral axis. Together, these results demonstrate the power of highly multiplex in situ analysis to the brain, with further application to a wide range of biological systems.

Thesis Files