Transcriptional profiling at the single-cell level offers exquisite resolution for studying genotype-phenotype relationships. However, isolating cells from tissues requires an enzymatic digestion step, which provides an opportunity for gene expression artifacts to occur. In addition, obtaining individual cells from certain sources, such as neuronal tissue, can be extremely difficult because of the number of cellular interconnections. To bypass these single-cell isolation problems, Grindberg et al. developed a method, reported in a recent issue of PNAS, to perform RNA-seq from isolated nuclei. All steps of the procedure are performed at 4°C, reducing the chances for expression perturbations during sample processing. To investigate whether an individual nucleus can suffice as a source of mRNA for cDNA synthesis, the authors isolated nuclei from murine neural progenitor cells (NPCs). Cells expressed cytoplasmically localized yellow fluorescent protein (YFP), and the nuclei were stained with propidium iodide so that nuclei isolated from homogenized cells could be verified to be clean of contaminating cytoplasmic material during sorting by flow cytometry or upon isolation by micromanipulation. Although the overall yield of cDNA was lower in nuclear versus whole-cell samples, the average number of transcripts detected by each method was virtually identical, representing roughly 78% of all protein-coding sequences in the NCBI RefSeq database. Importantly, expression patterns appeared to be the same whether a single nucleus or cell was analyzed, with 98% of the transcripts being detected at similar levels. Based upon the successful tests in NPCs, the authors applied their single-nucleus sequencing method to murine hippocampus dentate gyrus tissue. As measured by number and quality of reads, the procedure performed equally well when isolating nuclei from tissue as it did when starting with cultured cells, and the authors conclude their technique should be widely applicable to vertebrate systems. Since about half of all transcribed sequences have been found only in the nucleus, procedures that rely on isolated nuclei may be especially suited to the study of noncoding RNAs and their role in regulatory pathways governing proliferation and development.
R.V. Grindberg et al. 2013. RNA-sequencing from single nuclei. Proc Natl Acad Sci U S A. 110:19802-7.Landscape architecture
Gene regulation studies are often performed using sequence elements that have been removed from their genomic context. In reality, cis-regulatory elements (CREs) do not function in isolation, but act in concert over large distances in complicated genomic landscapes. A recent article from Bessa et al. in Genome Research introduces a new tool to study CREs in their native environs. The key to their “expression disruption” system is insertional mutagenesis with a Tol2 transposon construct containing a strong insulator, which isolates regulatory neighborhoods by sitting between enhancers and promoters, blocking their interactions. The researchers introduced the insulator-containing transposon into zebrafish embryos. Random integration of the insulator would be expected to disrupt the local genomic landscape; if such a perturbation occurred, phenotype changes should enable inferences about the regulatory mechanisms governing the affected gene. To screen for embryos with informative integration sites, the authors designed the construct to contain a GFP (green fluorescent protein) gene on one side of the insulator and an RFP (red fluorescent protein) gene on the other. Each reporter has a minimal promoter sequence, so each functions as an enhancer trap, picking up regulatory elements upstream (GFP) or downstream (RFP) of the integration site. Screening for different expression patterns of the two reporters hones in on the most disruptive insertion sites, which can be investigated further to detect regulatory anomalies affecting nearby genes. In all, Bessa and colleagues generated 223 transgenic lines from injected embryos, of which 78% had different patterns of GFP and RFP expression. The authors used inverse PCR to map 59 of these cell lines and, by comparing expression patterns of the reporter with those known for nearby genes, were able to identify a likely target gene in about half. After generating homozygous embryos for the most interesting lines, the authors performed in situ hybridization of the inferred target gene and showed disruption of normal expression patterns. Tellingly, these patterns could be rescued by Cre-mediated excision of the insulator. These mutant embryos can be used in assays to investigate physical interactions between chromatin loops, and can also provide phenotypic information, both of which can help characterize a gene's cis-regulatory landscape. Preliminary evidence suggests that the same insulator-containing construct is also functional in mouse embryos, so it may be applicable in other animal models. Indications are this expression disruption strategy will be a promising tool for investigating genetic variants important to human disease, many of which occur in noncoding sequences.
J. Bessa. A mobile insulator system to detect and disrupt cis-regulatory landscapes in vertebrates. Genome Res. [Epub ahead of print, November 25, 2013; doi: 10.1101/gr.165654.113].