Jim Eberwine and I am involved in various aspects of single cell genomics.
We have been using the idea that RNA of a cell is a key tunable part of the molecular state space and whole transcriptomes can be used to trans-differentiate one cell type to another (Kim and Eberwine, 2010, Trends in Cell Biol. DOI: 10.1016/j.tcb.2010.03.003). Using laser-mediated phototransfection, we showed that differentiated neurons in rats can be made to acquire the stable phenotype of differentiated astrocytes (Sul et al. 2009 PNAS doi: 10.1073/pnas.0902161106). We are now working to generalize this process to other cell types and to develop new combinatorial process to systematically explore the idea of cell type tuning.
Genomic factors for subcellular localization
Many molecules in a cell act in a spatially localized manner. In particular, specific RNA in neurons translocate to dendrites and axons and serve as templates for local translation in response to external signals. We are studying the sequence/structure factors that mediate this localization at the whole genome scale. We have found that many mRNA contain subsequences of introns that are retained when exported to the cytoplasm (CIRTs; Cytoplasmic Intron-sequence Retained Transcripts). Some of these CIRTs seem to contain signaling factors including localization signals. Our hypothesis is that the CIRTs mediate subcellular processes and are cytoplasmically spliced out before translation.
Single cell RNA variability and its implication for age-related diseases
Our bodies are made up of organs, organs are made up of tissues, and tissues are made up of cells. As our body ages, not all parts deteriorate at the same pace. During degeneration of bodily function from age and disease, not all organs are affected equally, not all tissues in an organ are affected in the same way, and in fact, not all cells within a tissue are affected in the same manner. For example, it is well known that age-related degeneration in Alzheimer’s disease involves a subset of neurons within a subset of brain regions. Different organs carry out different functions and different tissues also have specialized roles—thus, that they are affected in different ways by age and disease may seem natural. But, at a superficial glance, cells within a tissue seem to all carry out similar function and occupy a similar niche, so why is there a difference in their response to age and disease?
Recently, using single-cell genomic technologies, our labs have discovered that heterogeneity of function extends down to single cells. Using a group of neurons isolated from rat hippocampus that seem identical by standard measures, we showed that individual gene expression and the molecular products of such expression might differ substantially between each cell. Surprisingly, the genome-wide gene expression data seem to suggest that cell types differ not so much by which genes are expressed at what concentration, but by a broad pattern of co-variation. In addition, the molecular products of gene expression show much wider variation in form than previously suspected and this form seem to be related to how much of the molecule is available for further processing (i.e., functional action).
From the above data, we hypothesize that heterogeneity in aging and age-related dysfunction may be due to heterogeneity in genomic function of individual cells. Thus, understanding the role of genomic elements that show cell-to-cell variability may lead to understanding why certain cells show disease traits and while others stay normal. Furthermore, we propose that cellular heterogeneity may have obscured our understanding of the roles of different molecules in age-related dysfunction. Uncovering variability at the single cell level will lead to new models of how molecules function in a cell and to the discovery of novel targets of therapeutics. In this project, we will use single-cell RNA sequencing to identify genomic elements that show cellular variability and then assay the contribution of such cell-specific elements to aging and cell degeneration in Alzheimer’s disease models.